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:
The Treasury Department released partial guidance for the new “foreign entities of concern” restrictions on clean energy tax credits.

The Treasury Department published long-awaited guidance for claiming the clean energy tax credits on Thursday, ending the state of limbo in which project developers have languished since the One Big Beautiful Bill Act passed last summer. Well, sort of.
Trump’s tax law put new restrictions on many of the clean energy tax credits, limiting eligibility to projects that could prove they had minimal material inputs or oversight from a handful of countries labeled “foreign entities of concern,” i.e. Russia, North Korea, Iran, and, most problematically, China. The problem was that it was hard to suss out exactly how to follow these rules. The Treasury Department would have to provide clarification, or in the parlance of federal tax law, “guidance.” Without this, developers might unintentionally break the rules, get audited, and then owe the government a bunch of money — a risk that financiers are not keen to take.
Now, developers have, shall we say, partial guidance. The FEOC rules have two main components, and a notice published by the IRS Thursday covers one of them.
The guidance clarifies how to calculate the material assistance limits, which ask for proof that a certain percentage of the material inputs to the project did not come from a FEOC-owned or -influenced company. For a solar farm, for example, that includes the photovoltaic cells, the frame, the glass, the sealant, the circuit boards, etc. These limits apply to the clean electricity investment and production tax credits (48E and 45Y), as well as the clean manufacturing credit (45X), and went into effect on January 1 of this year.
The notice the Trump administration published this week demystifies the material assistance math for some project types, but not others. It says the Treasury will be publishing more on this by the end of the year.
Then there are foreign influence and “effective control” restrictions that have to do with the ownership structure of the project. Those apply to any project attempting to claim a tax credit, including carbon capture (45X), nuclear, (45X), and clean fuels (45Z), that started construction as of January 1, 2025. Even though these rules have been in effect for longer, the Treasury has yet to clarify how to follow them. The notice suggests the department will publish this along with the additional information on material assistance.
To be clear, development did not halt or even really slow as a result of this missing guidance, although that may have been starting to change. Many companies were able to avoid the arduous material assistance calculations by starting construction on their projects last year, before the new restrictions went into effect. They were also allowed to use past IRS guidance, including tables breaking out the various components of a project and their relative weights for determining the amount of domestic content in a project, which they could then apply to determine the amount of non-FEOC-produced materials as a temporary solution.
As for the ownership restrictions, “you just err on the side of caution,” David Burton, a partner at the law firm Norton Rose Fulbright, told me.
I spoke to Burton late last night right after he had gotten through reading the new 95-page IRS notice, and he walked me through some of his initial takeaways.
What are the questions that companies had about FEOC prior to this that this document clears up?
It’s pretty specific on how to calculate whether or not you meet the material assistance percentage restriction. So for instance, if you have a repowered project, there was a question of, do you have to apply material assistance to the new stuff you’re adding? Or do you also have to apply it to the old stuff? And the rules clarify, it’s just the new stuff. If you have a solar project that you’re repowering by replacing the modules but you keep the old inverters, the new modules are subject to material assistance, the old inverters are not. So it clarifies that type of thing.
I think there’s going to be a lot of accountants doing spreadsheet work based on these calculations in the notice and the various elections and choices, trying to find the most advantageous path. I think it’s too early to tell if there’s some opportunities that the industry might benefit from, or some landmines that we weren’t anticipating, because there’s just … the calculations, there’s too many of them, they kind of link together, and it’s very complicated. So we need a little more than a couple hours to go through all that.
What do you mean by elections and choices?
You can use the domestic content safe harbor tables, or you can get a certification from a supplier. The notice says that if a supplier gives you a certification that says it’s not a prohibited foreign entity, and it’s not aware of any prohibited foreign entities in its supply chain, you can rely upon that. Or if it gives you a certification that says, I’m not a prohibited foreign entity, but 20% of the supply chain that feeds into my product is, you can rely upon that.
You’re unlikely to get top-to-bottom certifications that totally answer the question, but it is helpful.
We’ve talked in the past about how far up their supply chain companies will need to look to calculate material assistance. Does it answer those questions?
It does provide guidance on those questions, but really only for the technologies that are covered in the domestic content notices. So for instance, fuel cells or combined heat and power: If you’re not wind, solar, storage or some other technology, it doesn’t provide that much help. But it does clarify for wind, solar, and storage how to do the calculation. It provides some guidance for technologies other than wind, solar, and storage, but it’s still going to be pretty hard, I think.
What is still missing from the guidance? What are the open questions that certain projects might still face?
Foreign influence and foreign control, the notice doesn’t cover. For instance, there’s a rule that if 15% of your debt is held by Chinese banks, you don’t get tax credits. The notice doesn’t tell us how to apply that rule — how that applies if one lender transfers to another lender and syndications of debt, all that kind of stuff. It doesn’t even tell us at what level to apply that test. Do you apply it at the project company? Or at the ultimate parent company at the top of the ownership chain? So it gives us none of that.
How often are you running into that with clients?
Every deal where the project began construction after 2024 has that question. Most of the time it really shouldn’t be an issue, but you have to ask, who owns this entity? Who’s on your board? Who has the right to appoint people to your board? We’re starting to write a lot of memos about this stuff, but there’s not a lot of guidance.
How do you deal with that without guidance?
We have a statutory language, so it’s not like no guidance at all. You just err on the side of caution, and you err on the side of it being overbroad, and then you end up asking the parties involved a lot of due diligence questions. And they’re like, really? We have to answer your 1,000 questions here?
Do you think that, based on this guidance, this is workable for companies? This doesn’t seem to be the sort of backdoor way to kill the tax credits that some people initially feared.
I think it’s workable. I think it’s relatively even-handed. I think they are trying to make them administrable. Not easy, not simple — again, full employment for accountants. But at least you can spreadsheet it. It’s better to have to build a complicated spreadsheet than just be like, well, we don’t really know what the rule is, we’re not sure what the path is here.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Facing down a sea change, the automaker has staked its next EV bet on a compact, sporty pickup.
“Full fathom five, your father lies,” the invisible spirit Ariel sings early in The Tempest, as a handsome and grieving prince listens, rapt. The song tells of a shipwrecked skeleton transforming into something else — its eyes have become pearls, and its bones pink coral — as it undergoes, yes, a “sea change.” It is the first time that phrase appears in the English language.
Ford is now facing its own kind of sea change. Over the past decade, the automaker has doubled down on its most profitable and exciting vehicles — pickups, SUVs, and the Mustang muscle car — and dropped from its line-up the cheap, boring cars that once made it famous. It embraced, then backed off, the transition to electric vehicles, in part because it failed to make money from them; and it began to reckon with the surge of cheaper, cleaner, and “far superior” EVs from Chinese producers that are transforming global auto markets around the world.
Locked into its aging but reliable line-up, yet unable to innovate at the low end, Ford might seem like the epitome of a company facing disruptive innovation. No wonder its stock has traded flat from where it was five years ago — even as the broader market has surged by more than 70%.
Its solution is an EV skunkworks, run by Tesla alumni, where it can develop a new “universal EV platform” to undergird future vehicles. Today, we got a peek at the first car to emerge from that secret shop: an all-electric compact pickup that will hit the roads by the end of next year. Its name? The Ford Fathom.
We know very little about the Fathom, as our correspondent Andrew Moseman wrote today. It will retail for just over $28,000, and even with mandatory delivery costs and other add-ons will stick to this side of $30,000. That makes it only a smidge more expensive than the gas-burning Ford Maverick, a sporty, compact, and popular pickup that starts around $27,000.
Ford promises that the Fathom will have as much seating capacity as Toyota’s RAV4, America’s best-selling car that isn’t a truck. (Ford’s own F-150, of course, holds the true No. 1 spot.) Those dimensions suggest the Fathom will sport a four-door crew cab, like the Maverick, making it more acceptable to families with kids — or young professionals who want to give their friends rides on the weekend. It will also have a frunk.
Beyond that, though, we don’t know much. We don’t know its range, for instance, and its price point shouldn’t inspire too much confidence on that front. Nor do we know, frankly, whether Ford can pull it off: When the automaker announced its first electric truck, the F-150 Lightning, in 2021, it claimed a price point of less than $40,000. Eighteen months of inflation later, it actually sold them for closer to $55,000 — and it still lost money on every EV that it made. Fixing the latter problem is part of why the skunkworks exists in the first place, and Ford now has an additional half-decade of experience making EVs. But consumers hoping for a miraculously priced electric pickup from the Blue Oval have been burned before.
If the Fathom disappoints, though, then consumers will soon have other options. The American car market is about to be deluged with sporty, compact pickup trucks — a welcome change from just a few years ago, when the segment was almost entirely dominated by mid-size and half-ton models. The Jeff Bezos-backed startup Slate will start delivering two-door, all-electric pickups starting at $25,000 at the end of this year. The automaker Stellantis, which owns the Dodge and Jeep brands, says it wants to bring another compact pickup — it’s almost more of a ute — called the Rampage to North America soon.
That’s welcome news for me — I love these little trucks — but I’m a little worried I’ll be outside my pickup-buying years by the time they actually make it to market. In the meantime, I’ll keep you posted on other updates about the Fathom. Will “sea nymphs hourly ring its knell”? No, but it will have Apple CarPlay and Android Auto.
The company confirmed its plans to market research company Cleanview.
The data center buildout has hit a new inflection point. It has long been true that artificial intelligence is fueling climate change by driving up power demand; more recently, tech companies have started directly financing new natural gas plants in their quest for AI glory. Now one is backing the biggest fossil fuel-fired power plant ever to exist in the United States — exclusively to power an AI data center complex.
That company is Amazon, according to the market research company Cleanview, which reported on Friday that the tech giant is building an AI data center campus in Texas powered by an up to 7.65-gigawatt off-grid natural gas plant.
That’s larger than any other power plant in the country — fossil or otherwise. The next biggest plant is the Grand Coulee hydroelectric plant in Washington State, at 7 gigawatts, followed by Arizona’s 4-gigawatt Palo Verde nuclear plant, and the West Count Energy Center, a 3.7-gigawatt natural gas plant in Florida.
The new power plant’s developer, Pacifico Energy, announced in January that it had secured permits from Texas regulators for the project, dubbed “GW Ranch.” The site is also permitted for up to 750 megawatts of solar and 1.8 gigawatts of battery energy storage.
It was not clear who the customer for all this energy would be until earlier this week, when Cleanview uncovered construction permits Amazon filed showing that the company owned the GW Ranch site. The company confirmed to Cleanview that it acquired the site and planned to buy power from Pacifico’s plant.
Not only will this natural gas plant be larger than the one in Florida, it will also use far less efficient technology. Pacifico’s permit says it plans to build 35 “simple cycle” generating units, which are typically installed in rarely-used peaker power plants and waste a lot more fuel potential than the modern “combined cycle” natural gas plants that serve as baseload power for the grid today. These more efficient turbines are essentially on backorder for years, and power-hungry developers have increasingly turned to the simpler versions as a quick fix as they race to bring facilities online.
According to its permit, the GW Ranch plant is allowed to emit as much as 33 million tons of CO2 per year. That’s twice as much as the most-polluting power plant in the country, the James H. Miller Jr. coal plant in Alabama, emitted in 2023, the most recent year for which data is available.
In a statement to Cleanview, an Amazon spokesperson said the company “believes in paying the full costs of powering our operations,” and that this Texas project “does just that: it’s powered by new on-site generation that won’t raise electricity costs for Texas families and designed to transition to grid-connected service as interconnection timelines allow.”
Some researchers disagree on that point, however. In an opinion piece for Utility Dive, Energy Innovation director Jeffrey Rissman and senior fellow Eric Gimon argue that the proliferation of off-grid natural gas generation for data centers will increase costs for regular people more than if the data centers connected to the grid, because they will be competing with utility companies for gas supply. “Data centers can buy gas in bulk and sign long-term contracts (as we’ve seen in Texas, Pennsylvania and New Mexico), giving them access to cheap gas, even if this unfairly drives up prices for everyone else,” they write.
Jane Flegal, a senior fellow at the Searchlight Institute, has also argued that building off-grid natural gas plants to serve data centers locks in emissions for decades because the plants don’t face competitive pressure from other resources. When a new natural gas plant is hooked up to the grid, by contrast, there’s a far greater chance that cheaper, cleaner resources will displace its generation over time.
The Rhodium Group recently developed a scoring system to help investors differentiate between projects that are likely to accelerate the energy transition, those that will have little effect one way or the other, and those that will actively slow it down. They used it to assess options for powering data centers, and found that off grid natural gas plants scored the worst, falling at the bottom of the latter category.
Regardless, Amazon still, somehow, asserts that it is committed to achieve net zero emissions by 2040.
The smoke pouring into Seattle from Spokane is particularly bad, but there’s also no such thing as good smoke.
I wrote this story from inside a cloud of smoke. Owing to some funky meteorology in the Seattle area this week — a pressure ridge paired with a thermal trough — the region’s usual westerly winds reversed, causing smoke from the fires burning in the eastern half of the state to pour through the mountain passes and river valleys of the Cascades and pool over the populous Puget Sound lowlands, where I live.
Though it’s cleared up some today, I’m still running my air purifier on full blast because I know what’s in the lingering smoke. Unlike the still blazing wildfires in Ontario that are burning through mostly uninhabited forests, the smoke in the Seattle area this week came to us partially from Spokane, where the Old Trails fire razed at least 700 buildings and homes last weekend. That means that beyond the usual organic matter associated with wildfire smoke, the pollution that has hung over Seattle has likely also contained particles and chemicals from burned plastics, batteries, cars, and household appliances.
But how can the average person be sure whether their wildfire smoke is the bad kind or the worse kind? (At least assuming that well-adjusted people do not obsessively watch the animations on AirNow.Gov, as I do.) I turned to Coty Jen, an associate professor of chemical engineering, and Albert Presto, a research professor of mechanical engineering, both of Carnegie Mellon University, to learn more about the chemistry of wildfire smoke.
“There is no safe smoke,” Jen said, setting me straight immediately. “It’s all bad. It will piss your body off.”
While it’s true that some smoke is more toxic than other smoke, what you might call the “all natural, organic” variety will still spike hospital emission rates and exacerbate pre-existing respiratory diseases, even if it is mostly burning trees.
Under ideal conditions, when cellulose or lignin — the main structural components of trees and plants as well as leaf litter and soil, the largest sources of carbon during a forest fire — heat up and combust, the chemical reaction creates carbon dioxide, water, heat, and light. But wildfires don’t burn cleanly, and the chemical reactions often stall midway through that process due to things like oxygen availability and temperature variation, producing intermediate products like carbon monoxide or partially broken-down bits of carbon, often called soot. The tiniest of these particles can be smaller than 2.5 micrometers across — 30 or more could fit across the width of a human hair — and are measured collectively as PM2.5, a catch-all term that refers to the size of the particle rather than what it is. What’s important, though, is that these particles are small enough to penetrate deep into our lungs and potentially enter our bloodstream, factors that add to the known mortality associated with PM2.5 exposure.
Different kinds of forests create different emissions — heavy duff, or leaf litter, which is common in pine forests, creates some of the densest smoke conditions. Wetter fuels also burn “dirtier,” creating more pollution. Different topographies also impact air quality in myriad ways; it’s no surprise that some of the worst pollution from the Spokane wildfires pooled in mountain valleys as a warm overhead layer of air trapped the particles near the ground.
Even “natural” wildfires can be extra toxic; burning eucalyptus, which grows in Southern California, is not something you want to inhale. Pine smoke can cause mutations in bacterial DNA, a common lab test for a substance’s potential to cause cancer. Wildfires that smolder are worse than those that burn fast; researchers have found that PM2.5 can be up to 70 times higher when fuels aren’t actively on fire. “You can even see this if you’ve ever built a fire yourself,” Presto explained. “There’s a period where everything is big and flaming, and then, if you’re burning a log, it eventually goes down to smoldering. The emissions are different.”
In the case of something like a house burning down in a wildfire, however, it’s not only cellulose and lignin combusting. “We’re good at engineering materials that are extremely robust, but when they burn, they release very exotic compounds,” Jen said. She pointed to the common plastic PVC, which is used for everything from exterior siding to plumbing and window frames. When it combusts, PVC releases chlorine, “which is very bad for you,” Jen told me. “It’s like how bleach is bad for you — it’s a pretty nasty chlorine compound. PVC isn’t releasing bleach, but it is releasing radical chlorine molecules that produce some crazy compounds.”
If you’re following a smoke event at home, the answer is “not really.” PM2.5 is measured in micrograms per cubic meter, which tells us how much small stuff is floating around, but not what that small stuff is. “It is pretty difficult to measure all the different compounds that wildfires, or broadly any pollution, will emit,” Jen said. “The easiest way to quantify it is to literally suck air onto a filter and measure how heavy it got.”
Measuring what exactly is in that mass requires instruments that cost in the ballpark of half a million dollars, which is not financially feasible at every air monitoring station, Jen went on. But while there are certainly academic applications for that kind of knowledge, a person trying to decide whether or not to go for their run in wildfire smoke doesn’t need that level of granularity.
“Some smoke is definitely more dangerous,” Jen said. “But as innocent bystanders, it’s not like we can pick and choose what smoke floats over to us. You just have to live with it, so the best mentality is to treat all smoke as bad.”
In a 2026 Science Advances study that attributed more than 24,000 deaths per year to wildfire smoke in the U.S., researchers found no safe threshold for PM2.5 exposure. Every 0.1 microgram per cubic meter increase in a county’s average annual PM2.5 from smoke was associated with nearly 5,600 excess deaths nationwide, even though most counties saw only trace amounts of smoke — about 0.4 micrograms per cubic meter a year. While it’s “orange sky” days, when the pollution spikes into triple-digit AQI numbers, that get the most media attention, even low exposure that you can’t smell or even see can be affecting your health.
PM2.5 is just one component of wildfire smoke — the other is gases, including benzene and formaldehyde. Many gases chemically transform as they move from where the fire is to where you inhale them. “The atmosphere is extremely oxidizing — it likes to add oxygen molecules onto compounds,” Jen said.
Some of those compounds react faster than others, “so it depends how downwind you are,” Jen went on. That’s why people closer to a wildfire — maybe a day or less downwind — get the distinctive campfire smell, mainly from the “young” vapors and volatile organic compounds. But for people on the East Coast who were subject to the Ontario smoke several weeks ago, the smoke had to travel several days to reach places like Pittsburgh and New York, and by then the sharper-smelling compounds had transformed into new pollutants like ozone.
The AQI only measures a few specific gases that are considered “criteria pollutants” under the Clean Air Act, which means, as Presto told me, “during these fires, you could emit a whole bunch of different other gases that don’t have an AQI number.”
Instead, you can look at the PM2.5 number to get the gist of how prevalent wildfire gases are. “If your PM 2.5 is high, it’s impossible for the bad gasses to be low,” Jen said. “The way we think about it is, there’s a bunch of junk on the particle, and if the same junk’s not also in the gas, it will evaporate off the particle into the gas. They always exist together.”
You might notice by now that I’ve written little about the actual AQI number, that score that appears on your weather app and runs from zero to 500 (or, confusingly, even higher). That’s because while the AQI is a great communication tool, it doesn’t offer us much in the way of the science of wildfire smoke.
The AQI measures five different pollutants — PM2.5, ozone, carbon monoxide, sulfur dioxide, and nitrogen dioxide — with the EPA setting specific concentration thresholds for each one, as my colleague Emily Pontecorvo has explained. “If local concentrations of any one of them tick up above those protective standards, the AQI will jump from green to a more alarming color,” she wrote. “The higher the level of pollution is, the higher the AQI and the darker the color will be.”
If you want to impress your friends, though, you ought to zero in specifically on the PM2.5 concentration — again, because the prevalence of the tiniest particulates is a good indicator of all the other gunk you can assume is in the air, too. (You can find the specific PM2.5 concentration usually by clicking for more information about the AQI on your weather app or checking IQ Air’s widget.) For example, at the time of this writing, my local PM2.5 concentration is 50 micrograms per cubic meter, more than triple the World Health Organization’s 15 micrograms per cubic meter threshold for 24-hour exposure. (The EPA’s 24-hour threshold is much more lenient, at 35 micrograms per cubic meter.)
When I asked Jen how she stays sane knowing all she knows about smoke exposure, she laughed. “I have just generally become more terrified of all campfires and all barbecues, but people already think I’m weird, so I might as well add to it,” she told me.
In all seriousness, though, she told me the answer is air filters, and her confidence in their ability to work. When wildfire smoke rolled through Pittsburgh, she had two running that she moved from room-to-room with her family, as well as a whole-house air filter. “We were getting PM2.5 concentrations in our house of about 80 micrograms per cubic meter when it was 150 outside,” she said. “But with the air filter on, we could drop that down to less than eight.”
Jen pointed out, though, that many people do not run their air filters properly. Filters are rated at their highest blower level, “so for them to be effective, you need to crank them to their highest setting to get all the air through,” she said. Most people keep their filters on auto or low because they’re so loud — myself included, until I learned otherwise.
Additionally, while an air filter is a rather large appliance, it really ought to be placed in the center of your room to be the most efficient, rather than up against a wall. (Again, my bad.) “When these wildfire events happen, the most effective place for the air filter is where you are, and you have to run it loud, which kind of sucks,” Jen said. “But it is better than breathing in gross air.”