You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:

America runs on natural gas.
That’s not an exaggeration. Almost half of home heating is done with natural gas, and around 40% — the plurality — of our electricity is generated with natural gas. Data center developers are pouring billions into natural gas power plants built on-site to feed their need for computational power. In its -260 degree Fahrenheit liquid form, the gas has attracted tens of billions of dollars in investments to export it abroad.
The energy and climate landscape in the United States going into 2026 — and for a long time afterward — will be largely determined by the forces pushing and pulling on natural gas. Those could lead to higher or more volatile prices for electricity and home heating, and even possibly to structural changes in the electricity market.
But first, the weather.
“Heating demand is still the main way gas is used in the U.S.,” longtime natural gas analyst Amber McCullagh explained to me. That makes cold weather — experienced and expected — the main driver of natural gas prices, even with new price pressures from electricity demand.
New sources of demand don’t help, however. While estimates for data center construction are highly speculative, East Daily Analytics figures cited by trade publication Natural Gas Intel puts a ballpark figure of new data center gas demand at 2.5 billion cubic feet per day by the end of next year, compared to 0.8 billion cubic feet per day for the end of this year. By 2030, new demand from data centers could add up to over 6 billion cubic feet per day of natural gas demand, East Daley Analytics projects. That’s roughly in line with the total annual gas production of the Eagle Ford Shale in southwest Texas.
Then there are exports. The U.S. Energy Information Administration expects outbound liquified natural gas shipments to rise to 14.9 billion cubic feet per day this year, and to 16.3 billion cubic feet in 2026. In 2024, by contrast, exports were just under 12 billion cubic feet per day.
“Even as we’ve added demand for data centers, we’re getting close to 20 billion per day of LNG exports,” McCullagh said, putting more pressure on natural gas prices.
That’s had a predictable effect on domestic gas prices. Already, the Henry Hub natural gas benchmark price has risen to above $5 per million British thermal units earlier this month before falling to $3.90, compared to under $3.50 at the end of last year. By contrast, LNG export prices, according to the most recent EIA data, are at around $7 per million BTUs.
This yawning gap between benchmark domestic prices and export prices is precisely why so many billions of dollars are being poured into LNG export capacity — and why some have long been wary of it, including Democratic politicians in the Northeast, which is chronically short of natural gas due to insufficient pipeline infrastructure. A group of progressive Democrats in Congress wrote a letter to Secretary of Energy Chris Wright earlier this year opposing additional licenses for LNG exports, arguing that “LNG exports lead to higher energy prices for both American families and businesses.”
Industry observers agree — or at least agree that LNG exports are likely to pull up domestic prices. “Henry Hub is clearly bullish right now until U.S. gas production catches up,” Ira Joseph, a senior research associate at the Center for Global Energy Policy at Columbia University, told me. “We’re definitely heading towards convergence” between domestic and global natural gas prices.
But while higher natural gas prices may seem like an obvious boon to renewables, the actual effect may be more ambiguous. The EIA expects the Henry Hub benchmark to average $4 per million BTUs for 2026. That’s nothing like the $9 the benchmark hit in August 2022, the result of post-COVID economic restart, supply tightness, and the Russian invasion of Ukraine.
Still, a tighter natural gas market could mean a more volatile electricity and energy sector in 2026. The United States is basically unique globally in having both large-scale domestic production of coal and natural gas that allows its electricity generation to switch between them. When natural gas prices go up, coal burning becomes more economically attractive.
Add to that, the EIA forecasts that electricity generation will have grown 2.4% by the end of 2025, and will grow another 1.7% in 2026, “in contrast to relatively flat generation from 2010 to 2020. That is “primarily driven by increasing demand from large customers, including data centers,” the agency says.
This is the load growth story. With the help of the Trump administration, it’s turning into a coal growth story, too.
Already several coal plants have extended out their retirement dates, either to maintain reliability on local grids or because the Trump administration ordered them to. In America’s largest electricity market, PJM Interconnection, where about a fifth of the installed capacity is coal, diversified energy company Alliance Resource Partners expects 4% to 6% demand growth, meaning it might even be able to increase coal production. Coal consumption has jumped 16% in PJM in the first nine months of 2025, the company’s Chairman Joseph Kraft told analysts.
“The domestic thermal coal market is continuing to experience strong fundamentals, supported by an unprecedented combination of federal energy and environmental policy support plus rapid demand growth,” Kraft said in a statement accompanying the company’s October third quarter earnings report. He pointed specifically to “natural gas pricing dynamics” and “the dramatic load growth required by artificial intelligence.”
Observers are also taking notice. “The key driver for coal prices remains strong natural gas prices,” industry newsletter The Coal Trader wrote.
In its December short term outlook, the EIA said that it expects “coal consumption to increase by 9% in 2025, driven by an 11% increase in coal consumption in the electric power sector this year as both natural gas costs and electricity demand increased,” while falling slightly in 2026 (compared to 2025), leaving coal consumption sill above 2024 levels.
“2025 coal generation will have increased for the first time since the last time gas prices spiked,” McCullagh told me.
Assuming all this comes to pass, the U.S.’s total carbon dioxide emissions will have essentially flattened out at around 4.8 million metric tons. The ultimate cost of higher natural gas prices will likely be felt far beyond the borders of the United States and far past 2026.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Talking with National Grid’s Matthew Satterwhite about his new report with S&P Global.
This week’s conversation is with Matthew Satterwhite, head of U.S. policy for National Grid. This week National Grid released a report in collaboration with S&P Global I found noteworthy amidst the data center backlash, asserting that building new transmission lines can potentially reduce consumer costs. I reached out asking if we could chat about how this argument leans into the fight over hyperscale infrastructure. I found our conversation illuminating and educational.
The following Q&A was lightly edited for clarity.
Why did you make this report?
It’s all focused on our customers. We’re always looking to find ways to make sure we can provide our service in the most affordable way possible, the most efficient way possible, and we always think of transmission, but it’s fallen out of favor recently. There’s so much demand with large loads, data centers, advanced manufacturing, reshoring. There’s such a need, and a lot of the debate has been focused on what we need on the generation side. We think transmission is an answer, as well.
We focused on what we have control over — since we’re in deregulated states, the only generation we’re doing is to help states reach their renewable goals. It’s a real page-turner. We really get to the core of everything.
Can we lower customer bills with transmission? This report actually showed us that’s a good investment and helps with the resource adequacy and the constraint problems we have in the Northeast. You can bring cheaper electricity in.
With respect to concerns for everyday consumers, how much do you feel like new transmission might alleviate ordinary Americans’ concerns about rising energy prices?
When you look at the demand that’s coming, the projection is that by 2035, we’ll have to add 45 gigawatts, currently. We’re on that path right now. Transmission alone isn’t going to meet that, but the question is, how do we temper that down? What do we do as National Grid to help alleviate the need for all that demand? Can we get that somewhere else rather than in the region by building generation? It's a different version of all of the above. It’s not a generation single answer or a transmission single answer. We think transmission is a big part of that.
This also allows you to bring in cleaner energy from other places. The more robust the network is, you can have energy in different places and bring that in. It replaces the need for some of the generation to be built and pays for itself by creating a cheaper return for customers adding this.
How much of the data center backlash is affecting your transmission project planning calculus? How is it changing what lines are built in the country?
We’re focused on how we can provide the cheapest service for our customers and physics. It’s science and long-term planning. We don’t have the luxury — we can’t follow, this month we’re thinking something, someone got mad, and so we’re thinking something else. We study a lot of science and physics to figure out how to build the grid.
Do you feel like the average Joe Schmoe American sees transmission as making their life less expensive and making their electricity more reliable?
I think there’s frustration and a lack of understanding about the industry overall. There’s fear of the unknown. Are data centers really driving everything that’s happening? That’s where I think, with reports like this, the benefit of it will be that people will read this and see there’s other things we can do to address the load that we need, something different than building a bunch of generation plants.
How do the question marks around whether data centers get built affect transmission planning? How much harder is the backlash making your job?
It’s a science question. Do we do a bunch of work and then nothing happens? That’s why states put their policies out. There’s multiple studies you go through with a region and with a utility. I think that’s one reason why you see states slowing down, to make sure the policy is in check so people don’t do work they don’t need to do. It’s about having the policy to make sure, if you’re studying something, you’re doing it with a purpose.
Plus more of the week’s biggest development fights.
1. Clark County, Nevada – The first data center approved on federal lands has hit a legal brick wall.
2. Jackson County, Missouri – We have yet another high-profile case of a city councilor losing their job over voting for a data center, and this one’s a doozy.
3. Utah – What’s it take for the Bureau of Land Management to approve a big transmission line for zero-emission energy generation these days? Geothermal, baby.
4. Huntsville, Alabama – You can’t even build a tiny battery storage facility in the middle of Alabama anymore.
Where temporary moratoria could happen next.
Brace yourself for more statewide data center moratoria.
So far there are only two full state-wide blocks on data center permits, in New York and Texas. At least fifteen states have moratorium legislation in the pipeline, but few if any of those bills stand a chance of becoming law in the short term. Here are five states, however, where a broad development pause may gain momentum in the next year or two — and all of them are crucial to watch this November.
If you blinked you may have missed it: New Hampshire Governor Kelly Ayotte, a Republican, said she wants to enact a statewide data center moratorium.
Ayotte first came out in support of a pause last month at a Rotary Club meeting, declaring, “It does not make any sense at all to site a data center in New Hampshire.” She also reportedly plans to include a moratorium proposal in her upcoming 2027 fiscal budget. New Hampshire’s legislative sessions occur in the first half of the year, so we won’t see action on a moratorium bill this fall. But Ayotte’s statements suggest the Granite State — which is controlled by the GOP — could pivot to a pause very soon.
New Hampshire has very few data centers. Like, almost none. Only two project fights exist in the Heatmap Pro database, both in Portsmouth, and each has been canceled amidst opposition. Ayotte’s remarks were prompted by the fight against a hyperscale project being studied in the small town of Bow at a former coal plant that closed in late 2025.
None of this should surprise anyone familiar with New England NIMBYs. A New Hampshire moratorium also makes sense given the state’s proximity to Maine, which almost had one of its own. Ayotte, who is up for re-election this year, is likely looking at the political fortunes of Governor Janet Mills and trying to avoid potholes ahead of a likely blue wave hitting her state.
This week, Arizona Attorney General Kris Mayes, a Democrat, came out in support of a statewide data center moratorium.
Mayes told Arizonans in a public statement on Monday that she wants to avoid undue strain on the electric grid and adding to the burden of water cuts led by the Trump administration. Phoenix, where opposition grows by the day, seems to be the primary reason. This shouldn’t in any way be a surprise given the backlash to these projects, which in Arizona’s case is rooted in legitimate water security concerns.
One of the first high-profile data center conflicts I ever learned about was in Arizona: Project Blue, which had to move on from the city of Tucson after officials voted it down last summer. That led Amazon to bail from the facility, though it’s still under development elsewhere on county land. Locals are deeply concerned about the water impacts.
Ordinarily an attorney general wouldn’t have any sway on legislative or executive policy, but the state is already quite receptive to restricting data center development. Governor Katie Hobbs has enacted a three-year pause on tax abatements for data centers, and in response to requests for comment on Mayes’ statement, has told media she’s working on more policies targeting the sector. Hobbs has said she will do more in the following legislative session, but it’s not clear what.
The real decisive action here is probably going to be legislation, and that will depend on the reception any moratorium finds with Republicans in the state legislature, which is typically split in this purple state. The Arizona GOP is quite pro-industry, and Mayes’ opponent in her race for re-election opposes restricting data center construction.
You really should get to know the name Cindy Holscher for the next two months.
Holscher, a state senator, won a surprise upset victory in the Democratic gubernatorial primary this year, and currently sits within a one-point margin of her Republican opponent. How’d she get the nom? By calling for a statewide data center moratorium. “It reminds me of when the automobile manufacturers had to put seatbelts into their cars,“ Holscher told MSNOW after she won the primary. “We as a people and as a state just need to make sure there are guardrails in place.”
Kansas politics are weird. The state is best known as a conservative ideological bastion that’s pro-business. Full Republican control of the Kansas government during the Obama era led to significant social services cuts most closely associated with former Governor Sam Brownback. But after that, Kansans seemed to like moderate Democratic governors, electing Laura Kelly in 2022. Kelly is now term limited out of office.
Kansas already has a colorful patchwork of local data center and renewable energy restrictions. Land use is a big deal in this agricultural behemoth. Should Holscher win in a blue wave year, she would have a mandate to enact a statewide moratorium. Still, Republicans control the legislature, and that’s unlikely to change. My major questions are, should Holscher win, would the GOP in state government listen to Holscher’s request? Or can she do this through the executive branch?
Politics nerds are obsessing over Ohio right now. There, Trump acolyte Vivek Ramaswamy is neck-and-neck in the polls for governor with a Democratic candidate who backs a “conditional” data center moratorium: Amy Acton.
What’s a conditional moratorium? It’s in the eye of the beholder, really. Technically speaking, Governor Josh Shapiro instituted a conditional moratorium in Pennsylvania, where data center projects cannot get permits unless they meet very specific standards set by the governor himself. Shapiro did it through executive action, but in this case, it’s unclear whether the moratorium will be codified through that process or through law.
Should Acton win — or if former Senator Sherrod Brown defeats sitting Senator Jon Husted in the U.S. Senate race — I anticipate major legislative action on data centers in Ohio. Republicans there have essentially permanent control of the state legislature, and they’ve historically been pro-data center. But the freakout over opposition to artificial intelligence and hyperscalers in the senate race specifically has spooked national Republicans, who think it provided the opening Brown needed to potentially win back his seat. Acton and Brown’s political fortunes appear to be wedded to one another, linked to a general angst in the American public.
Every top 5 list needs a wild card, and mine is Oklahoma.
Currently, there’s minimal risk of a data center moratorium. I might’ve had this state higher on my list had Gentner Drummond won the runoff for the GOP gubernatorial primary, given his proclivity to side with anti-renewables activists who also oppose data centers. Instead, likely future governor Mike Mazzei is running on a more moderate, Trump-friendly approach to data centers centered on maintaining industry growth while protecting ratepayers from new infrastructure costs. His opponent, Cyndi Munson, supports a one-year moratorium.
I consider Oklahoma’s odds of having a data center moratorium about equal to the chance of a statewide wind energy ban. Momentum for anti-wind legislation began in the state legislature, and I expect the same to happen with data centers. But unlike the wind industry, which has enormous power in the state, data centers are still a nascent industry. This is a place that may take about two or three years to manifest full cultural upheaval over these projects.