Everyone has a strong opinion on energy right now.
If you’ve followed energy for a while, none of this is new. There have always been strong opinions — renewables versus fossil fuels, subsidies versus markets, activists versus infrastructure. But the intensity feels different right now.
People are arguing about everything: the speed of the transition, how to fix broken electricity markets, whether renewables raise or lower power prices, whether AI data centers are about to break the grid.
So who’s actually right?
This week, JP Morgan’s Michael Cembalest joins the show to weigh in on some of the top fights in energy. Michael is the chairman of market and investment strategy at JP Morgan Asset and Wealth Management. He writes the “Eye on the Market” newsletter, and every year he publishes a deep dive on energy market trends. This year’s report is called “Fighting Words.”
We talk with Michael about the fallout from the war with Iran and why the global economy may absorb it differently than past crises. We also dig into gas markets, electricity prices, data centers, CCS, green hydrogen, and sustainable aviation fuels — and what they all reveal about the reality of today’s energy system.
Credits: Co-hosted by Stephen Lacey, Jigar Shah, and Caroline Golin. Produced and edited by Anne Bailey, Sean Marquand, and Stephen Lacey.
Want to watch this episode? Subscribe to our YouTube channel.
Ready to accelerate your career in clean energy? Yale’s Financing and Deploying Clean Energy Certificate is a fully online, 10-month program built for working professionals. It delivers real-world skills in clean energy policy, technology, project finance, and innovation — all in just five hours a week. Enroll hereand use the discount code OpenCircuit26 on your application to save $500 on tuition. Applications close April 20, 2026.
Explore the new era of AI innovation in the fifth season of Where the Internet Lives, an award-winning podcast from Google and Latitude Studios. Follow and listen to Where the Internet Lives on Apple, Spotify, Google, or wherever you get your podcasts.
Join Latitude Media on April 13-14, in San Francisco for Transition-AI 2026, a two-day, in-person conference on the digital and energy infrastructure buildout needed to support AI load growth. Our podcast listeners get a 10% discount on this year’s conference using the code PODS10. Register today here!
Transcript
Stephen Lacey: Michael, I expected a little green from you on St. Patrick’s Day.
Michael Cembalest: Well, actually this is green. Remember that thing from a few years ago, like what color is the dress? This is green. It’s just the wrong light.
Jigar Shah: Well, and the original color for St. Patrick’s Day was blue.
Michael Cembalest: Oh, I didn’t know that.
Jigar Shah: Yeah. I did a trivia night with my son the other day, that was one of the questions.
Stephen Lacey: All right, there you go. Well, I did a little research because I realized I didn’t know anything about St. Patrick’s Day and it turns out that he was a very humble missionary who loved his enemies, and so I guess I can’t think of a better person to celebrate for today’s recording about the fights in energy right now.
Michael Cembalest: Perfect metaphor.
Stephen Lacey: From Latitude Media, this is Open Circuit. If you’ve been following or working in the energy business for a while, you are no stranger to a fight. There have always been heroes, villains and strong opinions; scrappy renewables versus entrenched fossil fuels; subsidies versus free markets; activists versus pipelines and coal plants, but lately it feels like the fights have multiplied and intensified. People are arguing about everything. The speed of the transition, how to fix broken electricity markets, whether renewables raise or lower power prices, whether AI data centers are about to break the grid. And honestly, I never thought in 2025, the sitting president would still be stoking a fight over whether CO2 warms the planet, but here we are.
And this week, we’re going to face some of those fights head on. Michael Cembalest of JP Morgan is back on the show. Every year, Michael publishes one of the most widely read deep-dives on the global energy system, cataloging the trends, numbers, and contradictions shaping the transition. This year’s report is fittingly called Fighting Words. So today we’re going to walk through some of the biggest showdowns in energy right now and arm ourselves with plenty of data. That is all coming right up.
Stephen Lacey: Welcome to the show. I’m Stephen Lacey. I’m the executive editor of Latitude Media. Jigar Shah is my co-host. He’s a longtime clean energy investor, podcaster and provocateur. Jigar, how are you?
Jigar Shah: Ready. Ready for the fight. Although I suspect we’re going to agree a lot more than we’ll fight.
Stephen Lacey: Well, joining us in the third chair this week is Michael Cembalest. He’s the chairman of market and investment strategy at JP Morgan Asset and Wealth Management and the author of the widely red Eye on the Market newsletter. If that is not in your media diet, Eye on the Market, then you are definitely missing out. And every year he publishes this sprawling analysis of the global energy system packed with charts, arguments, sharp observations, and it is red meat for us here on Open Circuit. So Michael, welcome back to the show.
Michael Cembalest: Thank you very much.
Stephen Lacey: The title of your opus this year is Fighting Words. You frame it around tensions, debates between decarbonization, energy security, affordability. What made you focus on fights? What’s different about this moment?
Michael Cembalest: I mean, I just… The vast majority of the people that I meet and read are in one of three corners. Renewables only, fossil fuels only, and nuclear only, and I keep expecting the Venn diagrams of those things to start to converge and they don’t. There are some people on LinkedIn and I’ve curated my LinkedIn feed to where I get a lot of really good energy stuff, but there are some people that will only tell you one side of the story. And for instance, in Europe and in the UK specifically, look at how much coal has been displaced, but there’s never any discussion about needs other than the grid, and there’s never any discussion of power prices. And everybody’s just kind of talking about one segment of one part of the pie.
Stephen Lacey: Who’s winning the argument right now? Who’s the loudest?
Michael Cembalest: It depends who you’re listening to. There are people that wake up every morning and scan the universe for tidbits of information that confirm their point of view, and those are the worst people to listen to. And I’m going to mention, I’m going to name names here. There’s a guy… And this has nothing to do with energy, but there’s a guy named Gary Marcus, and-
Stephen Lacey: I know. I read Gary. Yes, I know him.
Michael Cembalest: Gary wakes up every morning and tries to find information that is related to flaws in agentic AI and evidence that our artificial intelligence isn’t really a thing, and in my view, he’s lost a lot of credibility because you’re not going to get any news from him that’s positive about what agentic AI is capable of doing. And in the energy space, there’s too much of that. There’s too many fossil fuel people that are publishing reports on bird deaths from wind turbines, and as if that was kind of the ultimate climate barometer. And so let’s look at the data. The energy transition is moving along at a linear pace in most countries. In some countries, that linear pace is a little bit faster than others, but something in the neighborhood of 0.7 to one and a half percent a year, and I’m referring to the annual rate at which renewables are displacing everything else.
And so the transitions are moving. There’s a lot of noise and yelling and screaming, but the transitions are moving and obviously are set to move a little bit more slowly now that the President and his team have cut subsidies for wind, solar, and EVs. The US is at the lower end of the range of about 0.5, and in Europe and in China, we’re looking at a faster pace, but that pace is still about one and a half percent or so. And I’m talking about renewables as a percentage of useful final energy, not just the grid, but the whole thing. So it’s continuing to move forward. There are a few things going on that suggest that that pace may accelerate a little bit, but it’s still a linear transition. It’s not an S-shaped technological Silicon Valley kind of transition.
Stephen Lacey: So let’s talk about what’s happening on the ground. I mean, last year we talked about a bunch of different constraints when you were on this show. We were facing equipment shortages, obviously the tariffs had been recently announced, there were transmission bottlenecks, there still are major transmission bottlenecks, long interconnection queues, coming federal policy swings. A year later, what’s your view on what actually mattered most and were there any that turned out to be less impactful than people feared?
Michael Cembalest: The tariffs are a big deal because by and large, the semiconductor industry has ended up exempt from tariffs. 80% of all goods related to what OpenAI and the hyperscalers are doing are exempt from tariffs. That’s not the case with respect to the grid. And so everything related to switches and motherboards and transformers and copper wire, and to me, that’s the biggest surprise of what the administration is doing. I can understand them wanting to tariff consumption as part of this argument that Bessent has, that there’s too much consumption relative to production and investment, but they’re tariffing all the transformers and other equipment that would otherwise be the widgets of faster electrification and things like that. So that’s having a cost. And if you look at the PPI, things related to the grid have the highest rate of inflation of almost all the 40 or 50 categories in the PPI report.
So that definitely mattered. And look, I think the bill that they passed that cut the subsidies will matter. We’re already seeing some early evidence that EV adoption rates in the US are slowing in contrast to what’s happening in the rest of the world.
Stephen Lacey: Jigar, what do you think mattered most and least from last year?
Jigar Shah: Well, I think that I, like you guys have been talking about, try to get to the truth first and then we try to evaluate everybody’s narratives. It’s clear that on the physical supply chain, just understanding not the cost, but whether we can actually make enough widgets, that I have now come to the conclusion is not a problem. So for instance, when you look at India, Malaysia, Thailand, the amount of transformer manufacturing capacity that goes under-utilized in those markets is so large that we can easily drop our availability back down to 52 weeks, and you’ve got companies like Air Energy and others in the United States that have gotten venture capital to do that. So that’s really more of a trust game. People in this sector don’t buy something out of a catalog that they don’t know anything about. They want to meet the people who make the transporters and use it and inspect it and make sure that they really trust it.
So those kinds of things have to happen. I also think that that’s true of the battery supply chain, it’s true on the solar supply chain, it’s true in a lot of other things. It’s most certainly not true on the natural gas supplies. So I think you’ve seen an extraordinary tightening of that supply chain and Mitsubishi is like they’re going to double their production. GE has said that, “We’re going to do bottleneck stuff, but we don’t trust you guys because we lost a lot of money in 2006,” and Siemens I think is in the same place. So we’ll see where that goes.
I think the other thing that I find fascinating around where we are in this moment is that the turf fights that Michael’s talking about actually is leading to bad decision making and conversations, and you see that within the PJM conversations around data centers, you see that within ERCOT and MISO and the way in which these different electricity markets are operating. They really do have this fights and not data-driven fights. And so that part of it was surprising to me and was not something that I would have expected from people who are professionally running these markets.
Michael Cembalest: To me, one of the most fascinating fights of the year was a paper that I think started at Duke. It’s a guy named Tyler Norris. He started at Duke-
Stephen Lacey: He’s one of the most widely cited…
Michael Cembalest: Now, citations don’t always equate to accuracy. One of my favorite ratios is the number of academic citations on carbon capture divided by the actual metric tons of carbon capture, which is the highest ratio in the history of science, and green hydrogen’s number two. So Tyler wrote this piece and then followed up with a similar broader piece now because he’s at Google and the premise is that with capacity factors averaging 70 to 75% within the combined cycle gas network in the United States, that there was ample room to add more data centers to the grid, increase efficiency, as long as those data centers agreed to be curtailed or have their own backup generation at moments of peak demand. And IMM, which is this monitoring organization that’s the oversight entity for PJM, came out with a scathing response, basically saying that this was all nonsense and that you had to have dedicated capacity for all these data centers.
And what struck me was they didn’t really have any economic skin in the game. They really truly believed that this Google thesis is nonsense, and I spoke to Tyler about it, he said, “Look, we’re still trying to figure out why they had the response they had because they obviously don’t agree.” But to me, that’s one of the really critical issues as we sit here right now, because if data centers are going to be closed off to joining the grid and they’re going to pursue their own behind-their-meter generation, 80% of that’s going to end up being gas. Well,
Jigar Shah: And this is where the Department of Energy not weighing in is a huge problem. So Google has proven this thesis to a number of players through contracts, so whether it’s in Nevada or whether it’s in Minnesota that they announced recently, they announced a deal with DTE in Michigan, and all of those contracts are under this thesis. Here is the available capacity, here’s the number of hours that we need to shift, we’re going to fund the form energy battery or the grid enhancing technologies or whatever it is that we need to do to free up this capacity and that’s going to be cheaper. And so that is exactly what their thesis is, that’s exactly what they’re signing contracts for.
And the independent market monitor and PJM has just gone off the reservation. I mean, they are now in a place where they are anti-Bitcoin miners and so they’ve limited the amount of demand response that Bitcoin miners can do even though Bitcoin miners are fully in effect in demand response markets in Texas. They’re anti… Just anti-demand response. I get it, that is their religious cross to bear, is that they don’t like any demand side measures, but there needs to be an arbiter of the science. DOE should come in and say, “Okay, here’s what their arguments are. Here’s what these arguments are. Here’s where the data is, and then that’s-“
Michael Cembalest: Well, I think they did the opposite, because that White House meeting with the PJM governors recommended that PJM hold these kind of specific auctions instead, where the generators would have to fully pay for essentially amortizing the cost of new generation. So yeah, there hasn’t been a lot of leadership on this front. And again, I’m surprised because usually when you see people retreating to certain corners and you look under their hood, usually there’s underlying economic self-interest that’s behind it, whereas in this case, it really is a fundamental disagreement in terms of how the grid functions. And I take the IMM report a little bit more at face value than you are.
Jigar Shah: The reason I care so much about this is because when you look at the data from 2010 till today, and you look at retail electricity costs, which is what’s animating everybody, is this affordability narrative. Generation costs are basically below the cost of inflation adjusted basis of where they were in 2010. Transmission costs are also at or below where it was in 2010. So the entire game has been distribution costs, and so if you don’t believe in demand response and you only believe in new generation, upgraded transmission to ship that generation and upgraded distribution to take at the last mile, then you’re saying that there’s actually no way for us to arrest the rate increases that we have today because you need new technologies by which to get more out of the distribution group we’ve already paid for.
And for a lot of utilities, that’s batteries, for some places it’s demand response, for other places it’s other things. But for the independent market monitor saying, “We don’t believe in that entire body of science,” I feel like, okay, someone’s going to have to step in here and it’s not going to be me, but someone’s going to have to step in here.
Stephen Lacey: Well, there’s another secondary fight below the surface, and that is the debate over how demand response will even be used within these data centers. So is it going to be compute level flexibility? Are folks going to use batteries essentially to curtail? And I think that Tyler Norris basically pointed out that if you curtail for a couple of hours a year, you could open up a hundred gigawatts of grid capacity around the country. And Google has been working on demand response for five years now. I mean, they developed this product called Carbon Aware Computing, and then they used it during the European energy crisis after Russia’s invasion of Ukraine, when there were gas supply shortages, they’ve done it in the Midwest at different data center sites. So I think that they believe that compute level demand response is a possibility.
But obviously it’s a certain type of workload. And I think a lot of people are very skeptical of the idea that you are going to go into these AI data centers and ask them to flex compute when they have to maximize the value of these GPUs. And so the question is, where is this flexibility going to come from if we really believe that it is a true resource at the data center level?
Michael Cembalest: The evolution of the conversations I’ve been having with data center owners has changed. When it started a couple of years ago, it was demand response is impossible, don’t ever mention it again, it’s the plague, and then over time they’ve kind of softened, but the sense I get, and I also have talked to a lot of the people within JP Morgan who run and operate our own offsite data centers, is a lot of creative things can be done with frontier model training workloads to adjust those, but most inference workloads, particularly those that support the payment system and other kind of 24/7 batch processing are very difficult to curtail, and so you would need some pretty immediate backup onsite generation to handle that transition.
Jigar Shah: I completely agree with you that no one is flexing inference loads. The whole thing was such an exercise in ridiculousness. We set standards for industrial load in the 1940s, and the data centers are violating those standards. You cannot shift your use of electricity from 20% utilization of your chips to 100% utilization back down to 20% in milliseconds. That is not allowed, that is illegal. You cannot use power that way. They need to see certain ramps, they need to see those things. So I think software will be used to fix the illegal nature of how data centers are using power, but I think the notion that you’re going to take chips that are depreciating in three years and say, “We’re not going to run them for [inaudible 00:20:02].”
Michael Cembalest: By the way, according to the hyperscalers, they’re depreciating them over six years, but…
Jigar Shah: I think that for a lot of others, it’s three years, but in general, I’m just saying, I think you’re always going to find it cheaper to build batteries, to build physical infrastructure that takes the load off the grid, then you are going to be shifting a lot of these AI related loads. The cloud, et cetera, that’s different.
AUDIO: Are you ready to accelerate your career in clean energy? Of course you are. Yale’s financing and deploying clean energy certificate, a 10-month online program is designed for working professionals ready to become the next leaders of the clean energy economy. You’ll dive into real world frameworks and clean energy policy, project finance, technology, and innovation, all in just five hours a week. If you’re passionate about clean energy and ready to level up your career, the Yale Center for Business and the Environment is the place to start. Join this year’s cohort by visiting cbey.yale.edu to learn more and enroll today. And be sure to use the code OPENCIRCUIT26 to save $500 on tuition. Applications are open through April 20th, 2026.
AI is pushing economic and technological progress into a full out sprint, and the energy sector is at the center of it all. Explore the new era of AI innovation in the fifth season of where the internet lives. An award-winning podcast from Google and Latitude Studios. What does it take for a data center to be a good grid citizen? How is AI revolutionizing agriculture, medicine, art, and manufacturing, and what does that mean for the future of energy demand and digital infrastructure? Find Where the Internet Lives wherever you listen to podcasts, or click the link in the show notes.
On April 13th and 14th, Latitude Media is offering the chance to hear from the experts on the front lines of the AI energy infrastructure build out. TransitionAI 2026 is a two-day in person conference in San Francisco, addressing the challenges of building at gigawatt scale in the face of AI load growth. The lineup of speakers includes Meta’s former director of energy strategy, Peter Freed; regulatory expert, Arushi Sharma Frank; Tim Hughes of STACK Infrastructure; and Google’s former head of energy development, Caroline Golin. And our podcast listeners get a 10% discount. Use the code PODS10 when you check out. See the full agenda and register today for TransitionAI 2026 at latitudemedia.com/events, or click the link in the show notes.
Stephen Lacey: Let’s turn to the biggest story in energy right now, which is the fallout from the war in Iran. Closure of the Strait of Hormuz has created the biggest oil supply disruption in history in terms of volume, and the big question is, how will this play out relative to other major oil shocks like in the 1970s? What is different this time, Michael, is that the energy intensity of global GDP has fallen. So why does that matter for potential outcomes here?
Michael Cembalest: I mean, it’s a mitigant. I mean, the oil intensity of GDP globally and in the US peaked in the 1970s or right around the time of the Iranian revolution. Then it fell by a third by 1990s Gulf War, but you could still make the argument at the time that oil intensity of GDP was pretty high and it’s fallen another two thirds or so since then, which is a byproduct of more efficient everything from cars, planes, refrigerators, lighting, window standards, et cetera. That’s the primary issue. The secondary issue has been some switching from coal to gas, and then the tertiary benefit has been from the advent of renewables in certain places. So you now have a much lower oil intensity of GDP, and then you have countries like China.
I mean, I’m amused by the administration’s contention that China should be offering some protection the Strait. By volume, yes, they import a lot, but as a share of their energy consumption, they’re among the lowest countries in terms of exposure to all of this because 55% of useful final energy in China’s coal. They get another 13% or so from renewables and another three or 4% from nuclear. So China in certain ways is more insulated on a normalized basis than other countries are.
Jigar Shah: Well, they also have their new overland pipeline from Russia, bringing in gas from that-
Michael Cembalest: Right. And China also produces 50% or so of its gas consumption.
Jigar Shah: And slowly increasing fracking within the country.
Michael Cembalest: Yeah. So China’s actually way more insulated here than a lot of people think. And so we had a webcast yesterday for our clients and one of the charts that we showed is, let’s synthesize oil imports and gas imports, oil and gas consumption as a percentage of overall primary energy consumption, and then oil and gas consumption as a percent of GDP. And then we looked at all the countries and you end up with parts of Northern Europe and Southeast Asia that actually have the highest exposures, but not China and the US is at the lower end. That said, a 50% increase in Brent has translated into 150 to percent increases in jet fuel. And one of the crazy things about the energy market is that oil is a global market. So for as much as the US happens to be a marginal oil exporter at this point, the US is suffering many of the same economic shocks from high oil prices as every other country in the world, even the ones that import all their oil. Whereas the gas markets, very different.
So LNG and pipeline import prices in Europe and Asia have gone up a meaningful amount and Henry Hub has barely changed. So when you kind of decompose all of this and normalize the numbers, you get a better picture. And one of the things that we did was to look at our sensitivity index cores and they match up pretty well with changes in regional equity markets since this invasion began. In terms of where we go from here, I’m trying.
Jigar Shah: Where else does it go but down?
Michael Cembalest: Yeah. I mean, I’m trying to think of a weirder spectacle than what took place over the weekend in which the President, after kind of belittling, insulting and tariffing most of the OECD for the last 12 months, then reached out to them for help on Strait of Hormuz protections and everybody kind of shrugged. I can’t remember the last time I saw something like that from a geopolitical perspective.
Stephen Lacey: And then he came out and said, “Oh, it was just a test.” We didn’t really mean it. This was just a test for our-
Jigar Shah: A test of what? The emergency broadcasting system.
Michael Cembalest: Right. So I’m really kind of puzzled by that. We had a briefing here the other day with some former general joint chiefs of staff who kind of walked us through what it would take to do destroyer accompaniment of tankers and container ships, and it’s hard, it’s real hard. And they reminded me that the Lend-Lease program in World War II, 30% of those ships got sunk. So the attrition rates, if challenged, by drone swarms, mines, missiles, and speed boats is substantial. So I kind of understand why nobody’s raising their hands to do it.
Stephen Lacey: I’m kind of curious about what you guys think about how this may feed into the US gas market. We are at capacity for LNG export facilities, and so that’s one of the reasons why not more gas isn’t going overseas and prices haven’t resisted here. Do you expect that to change in any way? Is this a long-term installation?
Michael Cembalest: There aren’t as many long term contract counterparties left to sign, which make some of the newer liquefaction and gasification products harder to finance. So Freeport LNG and a bunch of those other companies took advantage of the market a few years ago, got a lot of those contracts signed. I think some of the projects that are on the drawing board are more uncertain because of how they’re going to get financed. It’s a couple of billion dollars each to build those kind of facilities.
Jigar Shah: Well, just to poke a little bit at your comment, I just want to make sure that the assumption that you’re making is that no one in their right mind would build a fully merchant LNG facility, that you would need an offtick to make, and I think you’re right to make that assumption, and I think that when you look at the Strait of Hormuz and 20% of global LNG being trapped and Qatar really shutting down their facilities, I don’t know a single country that I’m talking to right now that is excited about addicting their economy to natural gas rates. And so if they’ve got an import terminal, they’re saying, “Great, we’re going to use it. We’re going to buy natural gas and we’ll build a power plant at the import terminal. We’ll build an industrial park at the import terminal.” But they are now canceling the billions of dollars that they would have to spend to build an inland natural gas distribution infrastructure.
They’re saying, “Why would we move from coal to a 50-year investment and an addiction to natural gas?” This is destroying the LNG thesis. The LNG thesis that was basically, “We need to help the environment by converting people from coal to gas.” People are looking at the supply chain and saying, “I don’t want to invest in a 50-year investment cycle [inaudible 00:30:01]-“
Michael Cembalest: Hang on. Couple things. Yeah, so LNG has now reached like 10% of US production. And so for all the reasons we’re discussing, I think it’s going to be hard to go much higher than that. Plus you’ve got domestic lobbies in the chemical industry that are pushing really hard on the DOE to prevent that number from going too much higher because-
Jigar Shah: Oh, I heard from them when we were servicing.
Michael Cembalest: Right. So the other thing too though is there’s a little bit of confusion. 20% of oil moves through the Strait and 20% of LNG moves through the Strait. The difference is that LNG is only 14% or so of total gas consumption. So it’s only a 3% number. So only 3% of global gas consumption is trapped because LNG only represents 10 to 15% of global gas consumption.
Jigar Shah: No, I’m not disagreeing with you. I just want to make sure that people understand that this has been a project that we’ve been pursuing since 2009. Hillary Clinton was Secretary of State and we went to all the countries around the world and said, “We figured out fracking, we’re going to have a ton of natural gas. We want all of you to invest billions of dollars to build natural gas import terminals, LNG import terminals,” and the DFC at the time, OPIC funded a lot of those import terminals, all that stuff. And so I’m just saying that this conflict has caused everyone to pause and say, “Do we still want to continue that 50-year project or are we going to stop here, we’re still going to use tons of LNG. No one’s going back on the stuff they’ve already said, but are we going to actually pipe gas to the rural communities in our country and build all those pipelines out when it takes 50 years to advertise that investment?”
Stephen Lacey: Yeah, do you buy that, Michael?
Michael Cembalest: I don’t know. I don’t know. When you look at the numbers, the geologically recoverable amounts of gas around, not just in the US, but around the world that the potential gas committee keeps revising, those numbers just keep going up. So from a purely mercantile economic perspective, the gas thing still has momentum because of the amount that are increasingly deemed recoverable around the world.
Jigar Shah: The question is whether you believe the electrostate technologies that China is pushing, that you talk about in your report are going to compete with natural gas consumption because the state has to invest a huge amount of if-you-build-it-they-will-come money to actually addict those rural communities to natural gas.
Michael Cembalest: Right. Yes. And the real critical question to me is, is electrification compared to solar, wind, hydro batteries? So I’m going to put a proposition out there and we’re not going to solve it on this call, but even assuming an LNG import price of somewhere around $10 at BTU, the solutions that we’re talking about are more expensive than gas. So, and let’s just take that as a premise, okay? And I’ve run my own numbers, other people have other numbers. Whenever MBAER has numbers that they publish, they come out in a different way.
I think sometimes their conclusions are predetermined, but the question is how much of a premium is it worth paying for in order to have the national security benefit of not being subjected to the kind of price wings that we’re seeing? The biggest mistake that I’ve seen people make, and this is where we’re probably going to all fight with each other, the biggest mistake that I’ve seen people make is looking at these nonsense marginal LCOE numbers from Lazard and saying, “Okay, wind, solar, batteries is a much cheaper solution than gas,” except it doesn’t do anything to tell you about the systemic cost of meeting grid demand in a more electrified system.
Jigar Shah: I’m not arguing that point. What I’m arguing is that… Remember, we’re talking about the 50 emerging markets in the world that import oil and energy. That’s who we’re talking about here. We’re not talking about OECD countries. They’ve already built their LNG import terminals, they’ve already built other inland natural gas infrastructure. Right now we’re talking about Kenya or Tanzania or Nigeria or other places. So now the question becomes, they don’t have the money to begin with to build out all of this natural gas infrastructure. They’re getting pressured by XYZ lobbyists or this person or that person to do it. It was hard enough for them to raise the money to be able to do this from private sector capital sources, the guarantees, the this, the that, the IMF, whatever.
And so now the question becomes, if you’re an industrial facility inside Kenya, my sense is that the government is less interested in spending all this effort to get you subsidized natural gas subsidized by the people of Kenya than they are facilitating an off-grid system. This is not about Ember versus Lazard. This is just about, it’s really fricking easy to raise $20 million to actually build a micro grid in rural Kenya right now. I can find 50 people to invest in that. It’s not easy to get a billion of dollars to build out if-you-build-it-they-will-come natural gas.
Michael Cembalest: Okay. Yeah, I’m not sure how much discussions of rural Kenya, Tanzania, and Mozambique are going to be moving the needle on global energy consumption.
Jigar Shah: That’s where the growth is. That’s where the population is. Everyone else has got a negative burden.
Michael Cembalest: Population, yes. But no, no, I disagree with you. I disagree with you. Population, yes. Energy usage per capita, no. So I think you’re wrong on that one.
Stephen Lacey: Well, so in this environment, on the electricity side, what’s going to win? Is it going to be a lot more coal burning? Do you think that sort of a configuration of renewables and batteries will benefit?
Michael Cembalest: Look, solar adaption is skyrocketing, skyrocketing, and particularly when paired with batteries, and I would expect that trend to continue, and that’s happening in Pakistan and it’s happening in developed markets, it’s happening everywhere, and courtesy of the Chinese government, which has mandated their solar companies to operate at deeply unprofitable margins that would make them go out of business in any other country in the world. So we have this peculiar situation where you’ve got a bunch of money-losing Chinese solar companies that have built massive amounts of capacity, which are facilitating this solar transition, and from a climate perspective, this is fantastic.
Jigar Shah: Well, but let me push you a little bit there. I have used the same argument that you’re using, so I totally agree. So to be clear, I’m not disagreeing with the fact that these companies are unprofitable, but I think if you benchmark them to the manufacturers in India, Turkey, Ethiopia, Jordan that are not unprofitable, you’re adding five cents a watt to them. So instead of 10 cents a watt, these are 15 cents a month modules. At that price, it’s still [inaudible 00:37:24], and the battery costs are coming down as well. So if you add another $10 a kilowatt-hour to the cost of the battery, they’re more profitable, it’s still pencil. I think the bigger thing that was in your Eye on the Market report, which I found fast, was you were saying basically that the trade credit that the Chinese government is providing all these countries exceeds the US marshal claim.
Michael Cembalest: Yeah, it’s enormous. The amount of money that they’re pushing out is enormous. And the chart that was most resonant to me was the one that looked at the amount of capacity in solar, EVs, and wind, even compared to Chinese consumption, was massive. So they have built so much capacity, there’s almost nothing for them to do with it, except provide export benefits to move that stuff overseas, and so-
Jigar Shah: And the part that I think you didn’t really cover in your Eye on the Market report, which I think I care more about, particularly on the national security and the economic security side of things, is that for many of these oil-importing countries, you’ve got James Gutman who is one of the co-authors of the New Joule Order that Carlyle put out last year. What he’s saying is that these 50 emerging market countries that are the largest importers of oil and natural gas right now are basically not wanting to addict themselves to this importation because you’ve got to use your foreign currency reserves to import this stuff and they’d rather use that foreign currency reserves to build data centers or do things that are more productive for their economy. So there’s this big trend around energy sovereignty and localization of these supply chains.
Michael Cembalest: I think to me, there’s the easy part and the hard part. The easy part here is adding wind, solar, battery, and hydro when you can do it to the grid to displace gas and coal when possible. The heavier lifting comes in when you’re talking about the other two thirds of energy consumption, which is not electricity-based, at least right now, but has to do with thermal heat used in industrial production, for home heating, for offices, and for transportation. And so that’s really where the rubber hits the road in terms of whether or not you can meaningfully accelerate the electrification of these economies the way that China’s doing. So China is right now showing the fastest pace of electrification in the world. The pace of electrification is much slower elsewhere. It’s stagnant in the US and in a lot of the emerging economies, it’s also pretty stagnant.
Jigar Shah: But if you take what you’re saying to its logical conclusion, I think you’re making a comment there on the net-zero challenge and the decarbonization challenge, but the argument that I’m making is that the technology exists and I think the paybacks exist to match China’s electrification level. It may not be 100%, but I think that everyone in the world can increase their electrification levels by 10 percentage points in an economically productive, profitable way. The technologies exist to do that, and that, if you believe that, is a game changer, because you’ve got that [inaudible 00:40:55], you’ve got everyone rethinking everything on the energy security, national security point with a point that Jason Bordoff from Columbia is making is that this kind of disruption, which was what occurred in 1973, had a far bigger impact on moving people’s behavior than the Paris Agreement in 2015. They’re entering another period of this instability.
Michael Cembalest: Yeah. What I want to continue to do is to focus on the specifics of what we’re talking about, which is when we’re talking about electrification, a big part of what we’re talking about is the benefit of things like industrial and residential and commercial heat pumps, which have like a three to one advantage-
Jigar Shah: Or electric vehicles.
Michael Cembalest: Or electric vehicles. And the issue is the cost per joule of electricity in a lot of countries in the world is a lot higher than the cost of a joule of natural gas or petroleum. If what you’re saying was true, we would be witnessing a faster transition than what we’re witnessing. I’m judging this simply by looking at what we’re seeing as effects on the ground on a country-by-country basis, and we track this for 95% of the world’s energy consumption on a country-by-country basis, and we’re looking at the speed with which things are being electrified and the speed with which renewables are displacing other things as a percentage of final energy. And if the incentives to electrify were as great as they’re sometimes advertised, I believe that we would be seeing a faster transition, but that’s…
Jigar Shah: I guess what I’m saying is, I think where Vaclav Smil has been proven to be wrong over and over again, is that when he looks at his historical numbers, he is-
Michael Cembalest: I don’t know how that came in here, but okay.
Jigar Shah: Well, because he’s the main perpetrator of this school of thought, in my opinion, is that he’s basically saying energy transitions take a lot longer than everyone thinks they will, and I think that he believes that what we need to say is that when you look at your range of 0.5% to 1.5% electrification, that we mean to say that electrification has to happen at 5%, but we don’t need to say that. I think that the underlying data shows that for places that are petrostates, basically, which is what the United States has become, you’re at 0.5% because you’ve got cheap natural gas at Henry Hub. But for places that are importing these fuels, you are closer to 1.5%. The difference between 0.5% and 1.5% is massive on an annual basis where you’re increasing, and that I think is why China claims to have a lot of overcapacity or people accuse them of having a lot of overcapacity, but it’s not clear to me that they do have a lot of overcapacity.
I think with this Strait of Hormuz issue and the Ukraine in 2022, there are a lot of countries that are oil importing and LNG importing countries saying, “We’re going to be closer to the 1.5%. We’re not going to be at 0.5% and we’re going to make sure our policy aligns with that because it’s in our financial and economic security best interest.”
Michael Cembalest: You also have some unfortunate countries like Taiwan, which are not just oil importers and gas importers, they’re also coal importers and I think something like 95% of energy consumption in Taiwan is imported through two easily blockadable ports, but that’s a separate story.
Jigar Shah: [inaudible 00:44:37] a couple of nuclear plants which they’re now trying to turn back on.
Michael Cembalest: Yeah.
Stephen Lacey: I just want to step back and ask a really simple question that I think a lot of people might be asking themselves, which is, are we in an energy transition or are we seeing energy addition. This is sort of the debate around how Vaclav Smil has sort of looked at previous energy transitions, they’re taking a lot longer than expected, and in fact, we’re actually just adding renewables on top of continued growth of fossil fuels globally. So what are you seeing?
Michael Cembalest: I try to avoid adjectives. So when I look at the chart that we have that monitors all of these different… We have like 160 charts in the paper. I don’t use adjectives to define any of them because they are what they are and one person’s adjective is going to be different from somebody else’s adjective. So I think it’s important for our clients to simply understand the transition math that’s going on. I think that to answer your question, you have to look at China, because China uses more energy than the US and Europe combined and is the pacesetter on a lot of things going on, whether it’s fission or fusion or solar or trying to reverse desertification. They’re the ones that are developing these new sodium ion batteries. They’re the pacesetters here.
And you’re going to get trapped in a war of words, because if you look at shares of energy consumption, renewables are displacing fossil fuels. If you look at joules of energy consumption, they’re not. And so you get trapped in this kind of debate that I don’t think is particularly useful when you’re trying to understand the economic benefits and incentives of the transition itself.
Jigar Shah: Well, this is why I thought your graphic was so important where all three sides are shooting at each other. I think if you talk to Chinese planners, what they will tell you is that solar and wind can handle the growth of our economy, but they cannot handle the replacement of our coal. That has become crystal clear to them. And so that is why they are scaling up nuclear as fast as they can, and they have done a good job as you talk about in the Eye on the Market report. And I think that if they are successful at scaling up nuclear, the Chinese have said that their stated goal is to replace their coal with nuclear, not with solar and wind. Solar and wind is good for growth, but it’s not good for replacing the base load because they also use all the heat. The thing about the Chinese is they need the heat from that nuclear power plant too. They don’t want just the electronics.
Michael Cembalest: My energy report is actually underneath this computer, so I don’t want to take it out because it’ll be like a Jenga disaster, but I will say, based on all the Chinese plants under construction, nuclear, even if we pencil them out by 2030, their wind and solar generation will massively eclipse the nuclear generation.
Jigar Shah: Through 2030, but through 2040, I mean, you can imagine it going much faster, and the Chinese are also installing gen four reactors now too, which optimize heat production. Because the Chinese are really efficient at using all of the parts of the coal. It’s not just the electrons coming out of the coal. They also have combined heat and power really from these coal plants, and so they need something that replaces that combined heat and power function.
Stephen Lacey: All right. I want to move on and see how many other fights, controversies we can get to, and I think to go back to data centers, this is an important one. Obviously, a lot of people are very concerned that data centers are driving up electricity costs. US power prices, as you point out, have only risen by two cents per kilowatt-hour in real terms since 2022, and electricity spending is only 1.4% of household expenditures. Is there a disconnect between fears over data centers and actual impact?
Michael Cembalest: I don’t think so. This was the longest section in the piece because it deserved to be dissected carefully and dispassionately. There are certain locations within data center alley where some academic work has been done to suggest that there is an impact from data centers showing up on residential power prices, in other places that hasn’t been seen. What was mentioned earlier is the right thing, which is some of these things can be very disruptive in terms of the speed with which they’re either ramping up or down their loads, and that has to be solved for.
I have a feeling that the backlash that’s taking place, because there are already some data center projects that are being protested and canceled and things like that, I have a feeling that sooner than later we’ll kind of end up in an equilibrium where the data center and the hyperscaler companies are going to end up having to foot whatever somebody else tells them the bill is. And remember, power on a levelized basis is only 10 to 15% of the cost of one of these data centers over its lifetime. Jensen Wang is making most of the money. So I think the hyperscalers will be willing to pay a little bit more for the certainty of power because economically it’s more important for them to get these facilities up and running.
Jigar Shah: But I think you missed the political angle to this, which is that the fact that the data centers are willing to allow the utilities to do dumb things and to accommodate their load in the most expensive way possible doesn’t mean that Republican governors find that to be good for their states. I think that at some point, and this is again, where DOE has failed to provide the leadership that they’re supposed to provide, someone needs to come in and say, “What Google is doing is actually genuinely cheaper for rate payers, and they should pay for 100% of their cost of whatever that form energy battery was or whatever it was, or it’s not, and we should build a budget behind the [inaudible 00:51:14] aircraft and ship engines behind our thing.”
But I’m just saying that we’re allowing people to sort of figure it out at the local level where they have no expertise whatsoever. A mayor of a city has no expertise, a county commissioner has no expertise, a governor of a state has no expertise. And so the DOE needs to come in and say, “Yes, what Google is suggesting or what GridCARE is suggesting or whatever these groups are suggesting is a viable way to do this as opposed to letting a thousand Twitter handles blue.”
Michael Cembalest: Yeah, I mean, I’m not holding my breath for this Department of Energy to kind of establish those kind of frameworks. Everything has become extremely ideological and I think the whole thing that’s going on with the XII projects in Tennessee and Mississippi are example of that where they got a waiver from the NOX restrictions just because they put the turbines on mobile devices, so all of a sudden they got some kind of waiver. I agree, there’s a lot of experimentation that’s taking place on a local basis and there’s no kind of coordinated policy approach to this and that’s unfortunate.
Stephen Lacey: Well, we’ve had some extensive conversations on this and I want to get your view on what it means for data centers to pay their own way. We had the signing of the rate payer protection pledge at the White House a couple of weeks ago and the policy conversation is increasingly focused on like how to get data centers to pay their own way. And as we’ve talked about, you can ring-fence the cost of data center costs, but if you’re taking a very CapEx-heavy approach to grid modernization and you’re failing to maximize the current grid, rate payers could still pay a lot more money. And then to your point, like there are widely different views on how to build energy on site. Do you want to put, I think, gas turbines on mobile tractor trailers or do you want to build the long duration battery and batteries in people’s garages and add demand response and try to get capacity that way? What do you make of the argument right now about how data centers should pay their own way?
Michael Cembalest: Look, my job is less about some of these policy solutions. The big picture for me is two years ago, the data centers were simply kind of driving their RV into your garage and plugging in, and now there’s an increased focus on how is this working? How should this work? So I think we’re in a better place than we were a couple of years ago where nobody was really paying attention to this. And I think there’s a lot more people that are focused now on… And if they weren’t really paying any of the integration costs, whether they’re paying 70 or 80%, I’m kind of indifferent. That’s a big move and will make things better and will be fairer for local rate payers if we get closer to a point where people are identifying the real incremental generation transmission and distribution consequences of having these data centers built.
Jigar Shah: But I think what you’re glossing over on this, I think, is that the electric utility sector is the one sector that has extraordinary amounts of unstructured data that they have not made a decision about, and the one technology that can do that is AI. And they are the one sector that are not using AI. And so the question becomes, we have all of these people who are running the grid using a slide rule, we have the ability to take all the smart meter data, all of the sensor data, all the stuff that they have, which is not in their grid operations module, and we can answer these questions in an exact way. And instead, we’re still handwaving. And I was like, “Why are we handwaving when we have super computers at all the national laboratories? Why are we handwaving when we have all of these PhDs that work there that know how to answer the question?”
Michael Cembalest: Yeah. I mean, I’m just taking it as a given that when I see a five to seven year queue for hooking up wind and solar to the grid, that we’re dealing with extremely complex routing systems and-
Jigar Shah: But we’re not, that’s what I’m saying. In Texas, they can get them online in 18 months. So it is a human problem, it is not a technical problem, and the reason I care about this so much is I think it translates into private credit and the Blue Owl stuff. And I think it goes into what’s going on with OpenAI and whether they’re going to be in business in four years and like-
Stephen Lacey: But how does it connect to those? Can we explain more why it connects to those?
Jigar Shah: Because initially we thought all of these people were AAA credit and that when they came in and they provided deposits into the grid and then they built an AI data center that they really were going to pay 100% of their own way, but now those contracts are dependent upon them paying their own way for 25 years. If they go bankrupt and then no one else decides to go in their spot because one day we all decide, which I think is happening any day now actually, that these 1,000 megawatt data centers are stupid, they’ve always been stupid and we need to go to inference on the edge and build a bunch of five megawatt data centers and 100 kilowatt data centers because that’s where inference is better, the latency is lower, then you’re going to get stuck with all this infrastructure for this 1,000 megawatt data center with a bankrupt counterpart. And then who puts the bill on that? The rate payers that are there.
Michael Cembalest: Right. [inaudible 00:57:30] part two. Yeah, agreed.
Stephen Lacey: I mean, there are a couple of big questions in that, which is how do we feel about the financial health of these AI labs and tech companies that are pursuing a zero-sum game in AI? Have your thoughts changed on the bubble dynamics at all over the last year, Michael?
Michael Cembalest: A little bit. If you roll back the clock six months, the hyperscalers and Oracle were financing almost everything through free cash flow and they finally hit the wall at the end of last year. In the fourth quarter, out of nowhere, we had this explosion of data center-related financing and the Blue Owl Meta one was really explosive for the markets in a lot of different ways. I mean, first of all, it was structured to be off balance sheet, which is kind of ridiculous because it’s obviously full faith in credit of the lessor, but it was the first time that you saw a lot of data center financing in the debt markets, which is a sign I think of things to come. And just in the last six or eight weeks, we’re starting to see free cash flow margins drop pretty sharply on the hyperscalers as well as Oracle.
So we’re getting closer to where there’s more and more debt financing that’s going to be needed to continue this build out, and that makes me nervous because that’s what so far has distinguished this build out from the fiber build out and from other, the natural gas build out and other big capital build outs in the past. The data center industry still says, “What is everybody worried about? We have 98% occupancy and pre-leasing of data centers,” but that’s now. I mean, I don’t know where we’re going to be in a few years.
Stephen Lacey: I have so many different sort of scenarios and data points that I want to walk through, but I want to give you the opportunity to pick something from the report. What is something that surprised you? Are there any storylines that we haven’t talked about now that you think are really important to address?
Michael Cembalest: Well, again, at the risk of getting some of your listeners upset, we had a section on SMRs, small modular reactors, and I think to me, simple. The premise is you can definitely build them. Some of them are using the same light water, boiling water reactor technology. Others use different technology in terms of gas cooling and things like that. Small modular reactors can be built. The big debate that matters now are two. Number one, how much is it going to cost? Because you’re starting to see TVA and other entities start to price them out. And I would argue that anything in the neighborhood of $130 a megawatt hour would be a home run. I strongly doubt that even the nth of a kind SMRs can get there, but that’s what the industry has to prove. And just for context, a high utilization combined cycle plant in the US, assuming four to $5 gas prices is somewhere about $80 a megawatt hour. So that’s still a big premium, but at 130, okay, maybe there’s some kind of benefits to it.
And unfortunately, we’re just a long way away from the kind of proof statements that would let us know where those are going to end up. And if you remember with Oklo and NuScale and things like that, most of the contracts that you read about are cancelable by the people signing them at the 11th hour if it turns out that the costs are too high. The other thing is this administration likes nuclear and really wants to do things to help nuclear. They have been gutting almost every government agency related to science, public health and energy, and so I have to… And there’s a chart we have that kind of shows the cliff-like nature of across multiple, like 10 or 12 different government agencies. They’ve canceled all sorts of special technical advisory committees.
And so the question I pose in the report is this administration capable and qualified to be overseeing the build out of the kind of power technology that if something goes wrong that’s potentially catastrophic when they’re gutting the science and public health agencies that you would assume would be part of that review process. And so it’s an unpopular question and it’s not a riskless question to ask for anybody and given the current environment in Washington, but that’s the question I’m asking.
Jigar Shah: Yeah. I mean, weirdly, I’m less concerned about safety on the nuclear front, just because I think that there’s such a culture of safety in nuclear and I’m more concerned about just the deal making. I think that we were told six, nine months ago that the Koreans and the Japanese were going to fund VC Summer through a fund that Howard Lutnick put together that Brookfield was going to operate. I don’t know about you, but right now I don’t think Korea and Japan is in the invest in America mood. And so I don’t know that that deal is going to come-
Michael Cembalest: Yeah. I mean, okay, a lot of people who I trust and respect, like Allison Macfarlane, who’s the former head of the NRC, have raised some really big red flags. I would suggest that anybody interested in the SMR space, just take a look at some of the stuff that she’s been right.
Jigar Shah: You’re right. I mean, people should pay attention there.
Stephen Lacey: And the early nuclear industry realized that you have to build bigger.
Michael Cembalest: The premise of the whole renewable transition over the last 15 years has been the learning curve and people should remember the original learning curve in nuclear was don’t build them in 50 megawatts. You got to build them to a gigawatt. And those lessons were learned. And so can you really modularize some of these things? I think the jury is out, is the nicest way that I can say it, as to whether or not…
Jigar Shah: But I do think that’s a big topic that we should… We don’t have to talk about now, but I do think it’s an important thing to highlight, which is that a lot of the cost productions have come from the learning curve, whether it’s solar or battery storage or offshore wind [inaudible 01:04:23]. And so the question is, should that be a gating item for nuclear? Should we say that the United States is not capable of doing large, complicated construction projects in a cost-effective manner, so we should be moving to whatever size nuclear plant can be manufactured in a factory, whether it’s Kairos or whether it’s Aalo or whether it’s Radiant or whether it’s whatever, like those are manufactured units.
So should we be limiting ourselves to things that you can manufacture? I’m sure it’s going to come in at a dollar a kilowatt-hour for the first unit and then the second unit will be 50 cents a kilowatt-hour and the third year will be 25 cents a kilo-hour. But then the Trump administration is trying to figure out what to do with 500 billion dollars of excess money that they tried to put in the defense department, put one of those things at every one of their bases.
Michael Cembalest: Yeah. Now we also have a defense action forum here conference that I work on every year. And I do think the military will probably be the first place of deployment for mobile SMRs that are going to be used in remote bases and things like that. But if you look at the history of nuclear submarine, I mean, they’re price-insensitive.
Jigar Shah: I’m not suggesting we’re ever going to get to three cents a kilo hour, but I could imagine that if you’ve produced a hundred units per year out of a factory, that you could get to 15 cents a kilowatt-hour, which is not horrible.
Michael Cembalest: Yeah, it’s not horrible. That’s a Manhattan Project lift to get to the point where the US can be producing 100 to 150 SMRs.
Jigar Shah: I endeavor United States to be able to do a Manhattan Project lift again.
Michael Cembalest: Right.
Stephen Lacey: All right, let’s run through a few other sectors really quickly. I realize we might lose a little bit of nuance in depth here, but I just wanted to go through a couple other sectors and have you just describe it, maybe react to it with one word, and then we’ll have you briefly explain it.
Michael Cembalest: Okay.
Stephen Lacey: Carbon capture and storage.
Michael Cembalest: Wake me when we get there.
Stephen Lacey: What are the big limiting factors there right now?
Michael Cembalest: The big limiting factors have always been the same, which is that the largest emitters… And this is another example of where people get distracted by little things in the corner. The largest emitters are coal and gas plants, where the CO2 is call it five to 12% of the flue gas. And so thermodynamically, it’s very expensive to strip out something that is five to 12% of the air handling. It’s an inherently expensive thing to do. And the last 12 months have seen a lot of data coming out that supports that and all this direct air carbon capture stuff is a waste of everybody’s time.
Stephen Lacey: Hydrogen.
Michael Cembalest: Similar story, which is the thermodynamics are just really difficult and a lot of analyses don’t fully load all of the costs because the electrolyzers are just one piece of it. It’s a complicated process and even if the cost of electricity goes to two cents a kilowatt-hour, you still have all the other costs and the market’s telling you something. Every year, way more green hydrogen projects are canceled in favor of other things than are being adopted. And so that’s been a waste of people’s time and attention as well. If you could replace the existing brown hydrogen, which is about a hundred million tons a year, with green hydrogen, that would be great. But all of this other hydrogen economy use cases in transportation and home heating and stuff like that is very farfetched.
Stephen Lacey: All right, we’ll lump all the fuels together, sustainable fuels, aviation, motor, shipping.
Michael Cembalest: I don’t have a huge horse in the race. We just have models that are tracking what’s going on and we’re not seeing anything. And so relative to the targets, they’re kind of not moving. And so a lot of that has to do, I think, with the difficulty in aggregating the kind of feed stocks that are necessary here, whether it’s used cooking oils or… And certainly green hydrogen is a very expensive way if you’re using that pathway. And so all the pathways have turned out to be pretty expensive. So whether it’s sustainable shipping fuels, e-methanol, the amount of subsidies those industries would need to compete with existing fuel costs, even if you had a carbon tax in place, are extremely high.
Stephen Lacey: So I just want to revisit sort of the interplay that we talked about at the top of the show as we round out here, and that is there’s this tension right now between decarbonization, security priorities, and affordability. When we go back and read your energy transition paper for 10 years from now, do you think there’s going to be any countries that have mastered all three of those?
Michael Cembalest: To me, the low-hanging fruit that, as I understand it, is if you bucket energy consumption into kind of residential and transportation and commercial office heating and industrial use, industrial energy consumption globally is the biggest piece. A lot of that’s been outsourced from Western countries to developing countries, but globally, that’s a very big number. And within industrial energy consumption, there are certain things like cement which require extremely high temperatures of a thousand degrees celsius or higher. But based on some analyses that I’ve seen, something like 30 to 50% of all industrial energy consumption is taking place at temperatures that would allow them to be electrified. And so like to me, that’s an interesting one because I think five to seven years from now, maybe we look back and somebody like me would say, “I underestimated the speed with which that energy consumption could be essentially electrified from gas and oil and then with renewables providing the power to do that.” So to me, that’s the most interesting one. I don’t know if you guys agree with that.
Jigar Shah: I live for that moment, Michael. I totally agree. I mean, look, I think that 10 years from now, several things I think will be true. One is that China will have perfected the ability to build nuclear plants, not only within China, but also for 50 emerging markets around the world, and they’ll be exporting that technology around the world, and I think that is a big deal. I think the second big thing is as a result of the electrostate stuff that they’re exporting around the world, I think you will see zero oil demand growth 10 years from now. And so they will have figured out how to either through efficiency or through displacement make sure that we stay at a hundred million barrels a day of production. I’m not suggesting it goes down to 50, but I am suggesting that it’s not going to keep going up by a million, million and a half barrels a day-
Michael Cembalest: In which case, that oil intensity of GDP numbers continues to plummet.
Jigar Shah: That’s exactly right. And then I think the third piece, which I am most excited about 10 years from now, is I think we will have universal electrification as per the sustainable development goals [inaudible 01:11:50] UN. I think the 700 million people around the world today who are in extreme energy poverty, we will have solved that problem within 10 years, and that gives me extraordinary hope because when you think about where the greatest amount of GDP growth comes from, it’s from taking those people and bringing them into the economy, and I think providing them basic levels of electricity is essential to accomplishing that goal.
Michael Cembalest: Right.
Stephen Lacey: To grossly oversimplify this, if we were to look back and compare the impact of decarbonization versus security as the primary force driving energy transition, which of those do you think wins out?
Michael Cembalest: I think there was a pendulum here, and your co-host is already getting ready, but I think there’s a pendulum and just let me tell you just from my perspective what it’s been like to be doing this job. So for the first 10 years I wrote the energy paper, nobody read it, because from 2005 to 2015, I enjoyed it, most of our clients weren’t paying attention. Then all of a sudden they started to pay attention and under the Biden administration, the pendulum swung really far towards prioritizing decarbonization over national security to the point where Manchin, who was, without which there would be no energy bill, wrote an open letter complaining about how the various cabinet agencies were misinterpreting the spirit of the original bill and over-prioritizing decarbonization over national security. So now we get this kind of crazy pendulum to the point where even when the private sector’s willing to do an offshore wind project in New York Harbor, the administration says, “Nope. Boy, would it be a shame if somebody took your permit away,” boink, and then it gets yanked for no legitimate reason. And eventually I think it was reinstated.
So now we have this kind of pendulum swing. And think about the irony of this. The bill they passed last year cuts subsidies for wind, solar and EVs, which are the things that have the clearest benefits, but left them in place for green hydrogen and carbon capture, which kind of is very puzzling to me. Eventually, I think the pendulum will kind of swing partially back. I had a section that I always like including that from MethaneSAT and other similar studies showing that the emissions that are reported to the Department of Energy from the oil and gas industry, for better, for worse, are grossly different than the satellite measures. And so I do think eventually you’ll have a partial swing back here and where we can have conversations about both, but we have to kind of go through… Unfortunately, it’s not that different from the immigration debate.
The United States had 800,000 people a year come in of undocumented workers every year for 20 years. That number then went to three and a half million under Biden, and now it’s zero. So as a country, we’re going through this kind of withdrawal reaction, violent reaction pendulum to these policy issues, and I’m hopeful, and I don’t use that word a lot, I’m hopeful that by 2028 and thereafter, we’ll kind of recognize that there’s middle ground that we need on all of this stuff.
Stephen Lacey: I think that’s very well said. Jigar, any final thoughts on sort of how the potential swings among those frameworks or which framework dominates?
Jigar Shah: So I think what Michael said is exactly right. I mean, Joe Manchin became a good friend of mine while I was serving and he and I shared a view that the Biden administration was not doing a good job on the national security front and that was what he passed into law. And so, I obviously worked hard to get the critical mineral stuff funded and we did, and we got a lot of the nuclear stuff funded and other things. And so I think that it’s important to note that the electorate cares a lot about climate. I think that the decision makers in general, in every part of the world, care much more about national security and economic security than they care about climate. It happens to be that the Venn diagram between those things is so overlapping because the technologies of the electrostate are so cheap now that you can do most of this under the national security and economic security framework.
And on top of that, I think when you think about where this is going to go, I think, when the pendulum swings back, is we’re going to move away from restricting supply. And you’re starting to see that already with Democrat talking points, is that the notion that we’re going to kill pipelines and kill the ability to drill for oil or whatever else was always a dumb idea. The way that you actually hurt the oil and gas industry is by getting people to use less of their product. It’s by destroying demand, by having superior solutions that people want to use. I mean, I like my electric car because I can beat any internal combustion engine car off the line because it goes zero to 60 in two seconds, and that’s awesome. And I think that in general, that is where this whole thing is going to go is through superior products, not by restricting choice.
Michael Cembalest: Yeah, and I’m encouraged… There was another section in the piece. I’m encouraged by the bits and pieces of specific virtual power plant and demand response programs are picking up and you’re… At some point soon, we’ll all start hearing about the retirement of peaker plants, and I think that will help build some momentum for the understanding that some of the old infrastructure can be safely retired without jeopardizing grid safety and things like that. To circle back, I think I’m now just from this discussion, understanding that this issue about data centers in a way, even though they’re only three, four, 5% of total power consumption, is critical because how do you handle those moments of peak demand on those really hot days and you have a power spike for either heating or air conditioning? How do the data centers kind of fit in with that big picture and how does the whole system work so that you can go ahead with electrification in the electrostate without worrying that you’re going to break the grid? And so data centers probably are a central part of how you do that.
Stephen Lacey: Absolutely. We’ve been talking a lot about data center development models and what it’s doing to intentionally unlock the demand stack and change the way utilities build infrastructure, and these hard questions about market reform and the business models to serve the capacity to [inaudible 01:18:51] data centers is really critical, and so that’s why it’s become such a big part of the conversation.
Michael Cembalest: The only thing that we haven’t talked about is geothermal. The Secretary of Energy is very hot on geothermal and there are some interesting productivity taking place with enhanced geothermal drilling, borrowing some fracking techniques, but it looks marginal. It looks marginal. The vast majority of geothermal energy is used for thermal heat rather than power, and it’s a great thing in the toolkit if we can improve it because it’s homegrown and doesn’t have any national security issues, but they’re very site-specific and to me, it’s not scalable like wind and solar and batteries.
Jigar Shah: Well, I think that’s another thing that we’ll revisit in 10 years.
Michael Cembalest: Okay.
Stephen Lacey: This is great. There’s so much more in the report that they couldn’t even scratch, a lot of country level transitions, in the case Germany regret over the nuclear phase out, the EV manufacturing business, whether it’s going to be structurally low margin business. There’s just so much in there. So go check out Michael’s energy paper, that’s the Eye on the Market newsletter and I’m so glad people are paying attention to it now, Michael.
Michael Cembalest: It’s a pleasure to kind of speak on this podcast. It’s wonderful to talk to people who care a lot about this stuff. So I really appreciate being asked to join.
Stephen Lacey: Absolutely. Well, we’ll link to it in a show notes. Michael is the chairman of market investment strategy at JP Morgan Asset and Wealth Management. Thanks again, Michael.
Michael Cembalest: So long. Thank you.
Jigar Shah: Thanks, Michael.
Stephen Lacey: Jigar, thanks so much. Good conversation as always.
Jigar Shah: Always.
Stephen Lacey: Open Circuit is produced by Latitude Media. The show is edited by me, Sean Marquand and Anne Bailey. You can find all of our episodes of Open Circuit on YouTube, subscribe to Latitude Media. And you can find the audio version, of course, anywhere you get podcasts, and transcripts@latitudemedia.com. Thanks so much for being here. We will catch you next week.
