May 10, 2024 – After this week's wrap-up, Financial Sense Newshour's Jim Puplava speaks with Mark Mills, Executive Director of the National Center for Energy Analytics about the collision between policy goals and economic realities when it comes to green energy, growing electricity demand from AI, data centers, and onshoring of semiconductor manufacturing. Mark explains how the US is not building enough electrical plants to keep up with demand, let alone enough wind and solar, so it will have to be provided by natural gas and coal, which is also seeing exploding demand globally. Mark also discusses the role of nuclear power, small modular reactors, and why places like Germany and California are seeing outright deindustrialization due to their aggressive green energy policies by not factoring in the difficulty, if not near impossibility, of switching away from the use of hydrocarbons and nuclear for electricity production. This is an eye-opening interview packed full of economic and financial market insights that you won't want to miss with one of the smartest and brightest minds when it comes to understanding the current conflict between policy, energy, and reality.
Website: National Center for Energy Analytics
Related article: The Magical Thinking Behind the "Energy Transition"
To speak with any of our advisors or wealth managers, feel free to Contact Us online or give us a call at (888) 486-3939.
Stay ahead of the market! Subscribe to our premium weekday podcast
Transcript:
Jim Puplava:
Well, one of the big events in the stock market over the last twelve months is AI. AI stocks have dominated the S and P last year and this year. But one thing that AI takes, it's called electricity. How is this going to work? Let's find out. Joining us on the program is Mark Mills. He's founder of the energy think tank and National center for Energy analytics. He's also its executive director. Mark, last year, AI dominated the S and P. It's doing it this year, and they're projecting by 2028, you probably have a better figure than I do, that AI will consume 25% of our electricity in the grid and that the cloud will be probably another 10%. We're not even talking EV's here. Where is this electricity going to come from?
Mark Mills:
Well, the thing about energy is you can only consume what you produce, and that sounds obvious, but if you don't have means to produce it, it won't be consumed. So we're not planning to build enough electric power plants in the United States to supply growth, so growth from the economy. So if we get growth more than we have now, which is very anemic, if we get real growth, it creates electric demands. If you are able to successfully repatriate manufacturing, which is the goal not just of this administration, but it's a bipartisan goal, by and large, that's an electric intensive business. Chip fabs, EV's are electric intensive by definition.
Mark Mills:
And now everybody's rediscovered the obvious. You have to plug computers in. They consume electricity. And when you increase the appetite of computing, which AI does, you need lots of electricity. So the combination of that has caused most of the grid operators and utilities to realize they've been underestimating the near term demand for electricity.
Mark Mills:
So that's a long way of answering your question. Where will electricity come from? Well, those who are bullish on wind and solar hand wave that we need more electric transmission lines to build more wind and solar plants, a long way away from where the loads are. Cause they aren't coincident. In other words, the power that's available from the wind is typically not where the data centers are or people live.
Mark Mills:
And then, of course, the problem with that theory is that it's not doable at any price. We are not building enough windmills and solar panels to supply the magnitude of electricity the country needs now or will need in the near future. So we'll either not produce it, which means we'll stifle growth, or we'll produce it by burning natural gas. And depending where you live in the world. Coal.
Mark Mills:
Coal construction is booming globally outside of the United States, and natural gas use for electric generation has been rising dramatically, and I think will continue to rise dramatically as reality sets in. So there's two separate issues that people are conflating, but they're closely related. What will the future demand for electricity look like? That obviously matters. That's the technology and economic forecasting game, has nothing to do with how we make the electricity.
Mark Mills:
So when I wrote my book, the cloud revolution, it's only a couple of years old now. A lot of it's about artificial intelligence and about the infrastructure of the cloud, which at the time I wrote the book, it's still largely the case. The global infrastructure that we call the cloud, which is all the computing that makes possible. Your smartphone, that makes Airbnb possible, that makes mapping possible. It gives you advice about shopping, all those advisory services, as well as how we store our clever videos and pictures, all that stuff collectively that collectively uses twice as much electricity as a country of Japan.
Mark Mills:
The global cloud. Or put in terms that maybe make more sense to people, the global cloud already uses more electricity, more energy than global aviation. Now, this where we are now with artificial intelligence. Artificial intelligence analogies are always problematic, but they're instructive. Artificial intelligence in terms of energy demand and in terms of its utility, that is, the two things are related.
Mark Mills:
If something's more useful, people will use it more often. It'll proliferate. Anything that we build or operate uses energy. So you need to have those. So imagine it's 1950s, and it's right at the time.
Mark Mills:
I think it was 1957, when Boeing introduced a 707. I think it was 57. I have to. Well, I'll tell you in a second. My brain will kick in and I will know the exact date it was.
Mark Mills:
Anyway, the 707 was the first commercial jet aircraft. There weren't any commercial jet aircrafts before that. And that kicked off a huge boom in global aviation use. Now, it was 1954. It happened before that.
Mark Mills:
There was lots of global aviation. There was global aviation. There was a global aviation industry with hundreds of manufacturers, thousands of airplanes flying. Never mind world War two, I'm talking about civilian uses for 30 years, it was already a 30 year old industry. When the jet engine came along and we accelerated the utility value of flying, we made it safer, easier, cheaper, faster jet aircraft could fly above the weather, all those kinds of things.
Mark Mills:
You would not have to have been a genius to figure out that that would cause a boom in aviation and a huge increase in the amount of fuel needed to fly people places, obviously, but nobody at that time wrung their hands over trying to figure out where the energy would come from. We just wanted to have the wonderful benefits of global travel. AI is to aviation what the jet engine is. It was. We've gone from the propeller era of computing to the jet engine era of computing, because AI adds functionality that didn't exist before.
Mark Mills:
It enhances the advisory capacity, if you like, of computing, but it's incredibly energy intensive. So Google has pointed out, and this is not new, it's just a technical literature that AI enhanced search of. Just something as simple as Google search. Google search is not bad. We all use it, right?
Mark Mills:
It's a pretty good tool. But if you add AI to Google search, the tool gets better. In effect, it learns what you're interested in. And as you iterate, just like the graphics programs, if you iterate, do the prompts right, it gets better. Right.
Mark Mills:
Google search, when you add AI to it, increases the energy cost for each search by tenfold. Not 10%, but tenfold. When we add AI to anything, it increases its energy appetite because it's functionally the equivalent of going from a propeller to a jet engine. Or maybe because we, we have this phrase, the information highway, the information superhighway that was. Everybody thinks Al Gore coined it.
Mark Mills:
He didn't coin it. It was coined by a korean american artist a decade earlier in a museum of modern art display. He did, where he created the phrase information superhighway. This was in the era of television and telefaxing and all the rest. But if you thought about the energy appetite of computing, that on the information highway and car truck terms, it's the equivalent of going from highways that drive mini coopers to highways using semi trailers.
Mark Mills:
That's the magnitude of energy increase associated with using AI in cloud functions versus conventional compute. And as you know, the world can't seem to spend enough money throwing AI and do into systems and data centers. And of course, that has sparked this incorrect claim that you led with that AI. The correct claim is AI and data centers together are a major vector now for electric demand for utilities in the United States that they didn't anticipate those in industry anticipated. I did in my book.
Mark Mills:
I write a lot about that in my book. But the market didn't believe it was happening until it happened, and now it's happened. But the truth is, it's not just AI, it's the data centers themselves. In fact, roughly speaking, the global data center industry, as I said, is already consuming more energy than global aviation. Now we're throwing fuel on the fire, pun intended, to make it grow even faster, which has caught everybody flat footed.
Mark Mills:
We've been developing AI for 25 years. This is a 25 year overnight success to make AI useful. There's bad uses for it, too, deepfakes and all that. But it's extremely useful tool for discovery, for chatbots, for graphics, for all kinds of things. It's not unanticipated.
Mark Mills:
We knew it was coming, but yet once it's happened, then everybody's surprised. That's the job. I mean, airplanes, we knew jet engines were coming. There were jet engines invented in world War two. They just weren't very good.
Mark Mills:
The B two is a jet engine, the german missiles, bombers. But to make them useful took another 1520 years, and they're extremely useful. The same is true for AI.
Jim Puplava:
You know, it's amazing because Amazon is talking about getting micro nuclear plants to power their data centers, because they see this coming.
Mark Mills:
Well, so let's be honest about nuclear power first. I'm a nuclear bull. I love nuclear energy. I think it's the only phenomenology of true interest that is better and different than burning hydrocarbons. But you can't use nuclear power plants to fly airplanes.
Mark Mills:
We don't have the engineering to do that. And you can't use nuclear power plants to make steel, so you still need hydrocarbons for lots of things. Make fertilizer and polymers, operate trucks. But when it comes to making electricity, power plants, there's nothing as magical as a nuclear power plant. It's incredible.
Mark Mills:
You shrink footprint, you shrink energy. Sorry, material use. You shrink all the metrics that we care about. They're high reliability. The problem is we're not allowed to build them in America.
Mark Mills:
Really? And the only nuclear power plants we know how to build, nobody seems to want to build the big nuclear power plants. Gigawatt scale power plants. We've only built one new power plant system in the United States in two and a half decades. That's the Vogel plant in Georgia, which just came online and over budget.
Mark Mills:
So all this excitement about new small nuclear power plants to power data centers, I'm thrilled at it. Well, that means that of the dozens and dozens of new designs, and there are dozens of them, for new classes of nuclear power plants, small ones, mid sized micro ones, portable ones, all these. They'll get built now, eventually, but they don't exist yet. They're PowerPoint plans. They're prototypes.
Mark Mills:
This is profound. We'll call it virtue signaling to be unkind. The only way they're going to power those data centers is by building gas turbines. That's what they're going to build. If they want to build a data center today, you could put an order in for a small nuclear power plant.
Mark Mills:
Good luck with that. They haven't been built yet. I'll say it again, they haven't been licensed, they haven't been proven. We have to build dozens of them and run them for years and figure out how to manufacture them at scale. All that will happen.
Mark Mills:
I think it's written in the stars. It's guaranteed we'll do it because we know it's possible. But it would be a shock if it took less than a decade, because it just takes time to build power plants of a first of a kind. Design them, certify them, get the supply chain in place. We don't have the ability to make the fuel for them at scale right now.
Mark Mills:
You don't just dig up uranium and shovel it into a boiler to make a nuclear power plant, right? You actually have very sophisticated, complex processes needed to refine and convert the uranium into the right form, enrich the uranium, and then package it in a form that is useful in a nuclear power plant. These are all hyper specialized industries which don't exist at all in some cases or at the scale needed in most cases, they will again. But there is no amount of money you could throw at it to accelerate the fact that you have to build the first of a kind, see if it actually works, how it works. Iterate it to fix the problems that inevitably come up in building big machines at scale.
Mark Mills:
And by big, I mean small reactors. Small reactors are big. Let's calibrate. A data center could have 50 to 100 demand. Commonly today, that's a steel mill, that's a steel mill, that's a giant petrochemical plant with demand.
Mark Mills:
We build dozens of petrochemical plants of steel mills, and we built thousands of data centers at that scale. Thousands. And now with AI, there's talk of data centers that set complexes that will be 500 mw each. Some are even talking about thousand megawatt gigawatt data centers. So you could park it next to an existing 1000 megawatt nuclear power plant.
Mark Mills:
But that means that electricity is not going to be available for anybody else. So you got to build a net new power plant somewhere. You could steal the nuclear electricity by paying a premium for it if you want a virtual signal as a data center operator, but you're going to leave the rest of the grid needing to get the power from somewhere, and it's not going to come from windmills and solar plants because they can't replace that power at that level of reliability and cost at scale.
Jim Puplava:
And the other thing the wind and solar doesn't, it's intermittent power. So what are you going to do with the data center if the wind doesn't blow or the sun doesn't shine? I want to talk about, as we're talking about electricity here, we've set some very aggressive climate goals, 2030, 50% carbon emissions, 20 5100 percent. I want to bring up what has changed with the war in the Ukraine and sanctions and how fuel, for example, being fed into Germany is not working. Germany got rid of its nuclear power plant, shut them down after Fukushima.
Jim Puplava:
Then we had the ukrainian war sanctions on russian fuel. Now Germany is reactivating its coal fired plants to get electricity. So lets talk about that, because a lot of this stuff isn't working.
Mark Mills:
Trey. Well, I think the entire narrative of the ability to abandon hydrocarbons, oil, gas and coal is first, it was predictably not going to work and it predictably cannot work. But that's a prediction. And what's happening now is reality is settling in. This is not to say, lets be really clear that there isn't use for, and nor is there used for more wind and solar.
Mark Mills:
There'll be lots more windmills built, lots more solar panels built, lots of uses for it. But that's a different thing than the claim, as you know, that we're going to abandon oil, gas and coal. So you have two things going on in parallel, which is epitomized by what's happening in Europe right now and in Germany in particular. One is the discovery, belatedly that you can't replace hydrocarbons. So Germany is the poster child for this spending, billions and hundreds of billions of dollars.
Mark Mills:
They double the size of their electric grid in the last 20 years, Germany did, by shutting down nukes, which slightly shrunk their grid, and adding lots of wind and solar. So they doubled their grid size almost entirely with wind and solar to supply a 7% increase in electric demand. So over those 20 years, the demand for electricity in Germany has gone up seven, not even 10%. But they doubled their grid. You don't have to be an economist to know that that would increase the cost of electricity in Germany.
Mark Mills:
It did. It tripled the cost of electricity in Germany, which is they had to use taxpayers money to subsidize industries and industrial electric rates to prevent industries from shutting down and abandoning the country. That's what they've done. And they were able to do that in part on the backs of importing cheap russian gas, gazprom. And, in fact, Germany's finance minister has publicly said that they financed, they hid the costs of wind and solar on the backs of cheap russian gas.
Mark Mills:
Well, so the second thing that's happened to your point is the ukrainian invasion by Russia has triggered the embargo on russian gas. So they've collapsed the amount of gas they're purchasing from Russia. Europe still gets pretty close to 15% or 20% of its gas from Russia. Still used to get overwhelming majority, but it's not gone. But to replace that, they've had to shut industries down, replace gas in boilers with oil.
Mark Mills:
So many steam boilers are using oil instead of gas, import it from somebody, because they don't drill for oil in Germany and import LNG largely from the United States. By redirecting our LNG exports from Asia to them, all of which has caused costs to go up and expose the real costs of wind and solar. And at the end of the day, as they say, to use that odious, overused expression, at the end of the day, not even 10% of Germany's primary energy comes from wind and solar. They haven't decarbonized their energy economy. What they've done is deindustrialized themselves or begun the process of deindustrialization because of the high costs and poor reliability of their energy plans.
Mark Mills:
So the ukrainian invasion sort of amplified and slightly accelerated that. Let's be clear about one thing, by the way. The embargo on russian oil and gas had a minimal effect on Russia's ability to make money selling oil and gas. At 70% of their economies, exports are oil and gas, and they continue to export roughly the same amount of oil and gas now as they did before the invasion. They've just switched buyers.
Mark Mills:
The buyers are China, India, Pakistan. And here's a real shocker for most people. Saudi Arabia. Saudi Arabia is buying russian crude. Now, why would the Saudis buy russian crude?
Mark Mills:
Well, because of the sanctions. It sells at a discount. Russias have to sell their oil at discount because they're sanctioned. I mean, we can't force, but they have. That's what they're doing.
Mark Mills:
So if you're Saudis, you buy russian crude at a discount, which they use to make electricity in Saudi Arabia and sell the rest of the world the oil at full price. Call me a cynic, but that doesn't sound like a very effective punishment for Putin. And it certainly hasn't hurt the Saudis.
Jim Puplava:
No, certainly prospering I want to talk about something. You've written an article about this and I follow this and that. That's the IEA, which was set up after the oil embargo in the seventies for energy security. But it's as you have written and I've read their stuff, their annual reports, it's become more of a policy arm for pushing the green agenda. The thing that drives me nuts, Mark, is they come out with these annual peak demand, peak demand every single year, at least I think for the last decade they've underestimated demand.
Jim Puplava:
So oil prices go up. They come out with, oh, demand is going to fall off and then they have to later revise it. Let's talk about that. Because governments in financial markets rely on the EIA and the IEA when they're looking at the price of oil in the markets.
Mark Mills:
An interesting problem. I mean, as you said, the International Energy Agency, the IEA, was formed at the recommendation of Henry Kissinger right at the time of the 73 74 arab oil embargo. They did that for two reasons. One is to, they realized they didn't have really good information, transparency, what's going on around the world in oil and gas markets in general. So they wanted a central repository that his mission was to collect and disseminate information and make forecasts.
Mark Mills:
And also they wanted some venue where you could have some coordinating mechanism in the event of an exogenous event, an embargo, a war, a natural disaster, you might want to coordinate as best you can how to shift where supplies go, who needs it. You can do arbitrage swaps if you like. So that ladder motive is a nice motive. It's been used a few times. It's not been super effective, but it's real.
Mark Mills:
The first motive is a very important one. And to your point, what's happened over the years, and this is not a secret, it was an overt shift emission. The director of the, executive director of the IEA and its member countries. So we're a member along with Japan and Australia, but we're a member of one of, I think that has more than two dozen member countries, they have directed their mission towards advocating for an energy transition. Now that's their right to do that.
Mark Mills:
I mean, it's an independent organization. We don't run it. We're arguably with Japan, the biggest single source of funding for it, but we don't run it. So they now have a confused mission. And this is reflected in what's happening.
Mark Mills:
The IEA organization, the staff doing the data and reports produce very good work. I use it all the time. But you have to ignore the executive summaries, and you have to especially ignore the policy pronouncements and aspirational goals of the IEA's administrative superstructure. You just have to ignore what they're saying completely and read the reports. If you read the reports, you'll find things that are orthogonal to the claims, to your point.
Mark Mills:
So they keep predicting peak oil demand. And in fact, they published in 2019, before the great lockdowns, that that was the year that oil demand would peak. It would shrink from there. We'd reached peak demand and it would shrink because of EV's and efficiency and its fuel switching away from oil. They have a whole set of reasons, but if you looked at the underlying oil reports that the people who run the division that did oil tracking and forecasting, they didn't see that in the data.
Mark Mills:
They saw oil demand absent an exogenous event like the lockdowns, rising. Well, as you well know, post lockdown, we've already blown past the 2019 peak and are seeing a new peak in global oil consumption already this year, and we'll see another peak next year. It's still going up slowly because it's a big market. So what's going on? I mean, what's going on is the member countries have overtly said that the IEA should advocate for an energy transition, and they are.
Mark Mills:
So the head of the IEA runs around talking about how we have to stop exploring, literally said, we have to stop exploring for oil and gas, exploring, never mind producing now, immediately, not in the future, in order to reach these goals. He's stating a truth, that in order to reach the goals, you'll have to do that. So it's not like he's lying, he's stating a truth. That's what you have to do. Well, the world isn't doing that, and it's not going to do that.
Mark Mills:
Here's a factoid, a fact, rather, the have in the head. It's not a factor, a fact to have in one's head. Roughly 82% of the world's energy is still supplied if you do it. The big bucket, right? But oil, gas and coal, and of the other, the melange of the rest of the things, wind and solar combined are about 4%.
Mark Mills:
So all the focus on wind and solar, we're still after trillions of dollars. At 4% of world energy, oil, gas could coal, 82%. The percentage is slightly smaller than it was two decades ago. Two percentage points. But it's in absolute terms, is bigger.
Mark Mills:
But that percentage actually misleads you in an important sense. 100% of the world's economy, 100% of every product and service, everything that all of us use uses hydrocarbons. Not some, not 82%. 100% of every product and service. Your smartphone, your data centers, your watch, your shoes, your clothes, the heat in your room, your light.
Mark Mills:
100%. At some point in the development, production and operation has to use hydrocarbons. It can't not, cannot avoid using hydrocarbons. Our economy is anchored fully with hydrocarbons. 90 of the actual direct fuel use, 97% of all direct fuel used to move all goods and services is oil and gas.
Mark Mills:
Of the remaining three percentage points mixed of biofuels and electricity, of global transportation, 100% of the machines built to use biofuels and electricity use hydrocarbons to get built in the first place, 100%. There are no exceptions. So it's the idea being promulgated by the IEA that we have to transition away from that. You're allowed to have ideas like that, but we have to be honest about whether it's remotely possible and the constant consequences of even trying. It's not remotely possible.
Mark Mills:
These are PowerPoint proposals to switch. They're just silly ideas. Frankly, there's no other word for it. That we can abandon this and make it all wind and sort of lots of transmission lines and lots of batteries. They're just silly ideas.
Mark Mills:
Arithmetically silly ideas. And trying to do it is economically destructive. I'll say one last point about what I mean by economically destructive. Let's just take as an example the Inflation Reduction act, that Orwellian misnamed act, which the progenitors of which proudly claimed after it was passed, that it's the Green New Deal. Since the vast majority of the money is the fund for climate change technologies, wind and solar batteries, hydrogen and stuff, the actual cost of it was claimed to be below 400 billion when it was passed.
Mark Mills:
The CBO re budgeted it after its pass, recosted it out at closer to 800 billion. The wood Mackenzie credit Suisse. I think two or three other banks have looked at this more carefully. What it's really going to cost. Goldman Sachs looks like somewhere between one and a half and $2.5 trillion.
Mark Mills:
And what is the purpose of this? This is money that hasn't been spent yet. So think about this. We are told, correctly, that the injection of one to $2 trillion of rescue money into the economy during lockdowns caused inflation. Obviously, you print money to not create a bigger economy, but just a rescue, you cause inflation.
Mark Mills:
And it also disrupted supply chains because of the lockdown. The combination was structural and monetary inflation, which is sort of relaxing. Now, we inflated the economy, but it's not inflating as fast and further. But if you're in the Fed and you're looking at the inflation Reduction act, what you're looking at is doing it again. Here's another one and a half to $2 trillion of money that has yet to be spent, which has to be printed, which is going to cause monetary inflation, and it's also going to disrupt supply chains.
Mark Mills:
By that I mean the energy supply chains, forcing them to switch from inexpensive, reliable energy to expensive, unreliable energy, which will cause structural inflation. So we are fueling another round of inflation that will start soon. So you say, how come it hasn't happened? Because they haven't had time to blow through the money yet, but they're starting and they're trying real hard. This is a really bad idea, in my opinion.
Jim Puplava:
And speaking of bad ideas, my own state of California, 2035, you're not going to be able to buy a gasoline engine or car. And then the Biden administration just issued an order that we're going to do that to our transportation trucks, the very vehicles that get us all the stuff we buy in grocery stores. How are you going to get a semi truck and make that an ev and make that work? And what will it cost? I know a lot of times you go to a charging station, you may be there for an hour or so because you got to wait in line. It takes you a half an hour. What are you going to do with a semi? Talk about stupidity.
Mark Mills:
It really is dopey. The electric trucks epitomize a characteristic of the silly planning that's in place. So you asked the question correctly. How are you going to electrify a truck? Engineers can do stuff like that.
Mark Mills:
There are electric trucks out there. They're already built. Caterpillars built electric mining trucks, giant ones. Cummins has built electric. I think Cummins built an electric trailer.
Mark Mills:
You could build this. Tesla has its electric semi trailer. You build a big enough battery, big electric motors. You have an electric truck. This is obviously, this is not the question.
Mark Mills:
The question is, in sort of two parts, how much does that truck cost? Just dollars a lot. A lot more than a conventional truck. And then how does it operate? Well, it has very low range because the batteries are so big.
Mark Mills:
They weigh tons and tons. So you have to refuel more often. And to your point, refuel take far longer to refuel. So a typical refueling of a car, which is faster and easier with gasoline, it's about five minutes, the amount of time it takes. Pull up, put, say, 1520 gallons in a tank and leave.
Mark Mills:
You could time it yourself, three to five minutes. Assuming you don't go in and get a coffee, you're out of there. If you're on a fast charger for an electric car, then you can conceivably get a two thirds, three, four, fill up in 20 minutes, 30 minutes, assuming you haven't had a wait in line. But you do the arithmetic here. That's five times longer.
Mark Mills:
So if you want to have the same utility function of the fueling station, you have to have five times as many fueling ports. Otherwise people will be waiting in line. That's just arithmetic. You can't have as many people during peak times refueling. But that's the least of it.
Mark Mills:
I mean, the truck will have the same problem. By the way, you can refuel a truck, not in five minutes. They're big tanks. They hold 100 or 200 gallons. So you got to take 1520 minutes to fuel a truck.
Mark Mills:
But if you want to fuel a battery, that's that big your hours for fast charger, not minutes, but that's the least of it. This is what nobody's taking into account. There's a profound difference in electrical engineering and physics between power and energy. Power is the an engine in your car, 200 hp. That doesn't tell you anything about how much fuel it uses.
Mark Mills:
The energy is the fuel consumed. It tells you something about it. If you use a big engine and go fast, you'll use more fuel, but one's power, the other's energy. The infrastructure needed to deliver power is not the same as infrastructure to deliver energy. So you have to deliver oil with an infrastructure and build a machine that has power, the engine.
Mark Mills:
The same is true in delivering electricity to a battery. The electricity has to be produced somewhere. In power plants, it doesn't matter whether solar or wind or building natural gas, but the equivalent of pipes and tanks are wires and transformers. And the interesting thing about electricity is, as you go to fast chargers instead of the charger in your home, which is three to 5 kw, which is about equal the peak load of a typical residence. Overall power demand, you need several hundred kilowatt to several megawatt chargers, 100 times the power.
Mark Mills:
100 times the power requires very different wires in very different, very big transformers that are very expensive. So the fueling station on the road that you, we all stop at, that has tanks in the ground, a fuel has an electrical demand, a load, a power load equal to like a 711 or a sheets. It's like a convenience store level power demand. If I convert that to superchargers for trucks and cars to charge them fast, the electric power demand of that one fueling station will rival a steel mill, and I need thousands of those along the highways. So I need, this is not, again, not fuel.
Mark Mills:
It's the power demand. I need thousands of steel mill level power demand and the infrastructure to support that. Massive transformers that don't exist at scale yet, massive upgrades in distribution systems, massive upgrades in local transmission that will cost hundreds of billions of dollars, which is not in the inflation reduction x large s of giveaway. It's not in any utility plans. So what will happen is you won't get to charge your car, you won't be able to charge your truck.
Mark Mills:
There'll be nothing available, or you'll wait in a very, very long line. So these mandates I can predict, I take the bet. The mandates quietly go away or get overtly reversed because you cannot run the economy. It will not be possible to deliver systems and goods at any prices at the scale we now do by making everything run on batteries. It's not going to happen.
Mark Mills:
It's not going to work. So it will end. The only question is whether it ends ugly or ends quietly. I hope it ends quietly.
Jim Puplava:
Well, along these lines, as we're talking about going to clean green, all this stuff takes raw materials, and there's a book out. You just reviewed it for the Wall Street Journal. It's called the war below. Lithium, copper, and the global battle to power our lives. Let's talk about that, because I don't care if you're building a windmill, a solar panel, whatever you're doing, making stuff takes raw materials.
Jim Puplava:
This little iPhone takes raw materials, which means we got to mine it. So let's talk about that, because nobody wants mining, or if they do want it, just not in my backyard. Go do that someplace else.
Mark Mills:
Yeah, well, we've been mining metals for all of history. In fact, mining predates written history and think copper is the oldest known mined metal. So we've been mining quarrying rocks for a long time, too. But to get metals, you have to dig stuff up out of the earth. Mining, you need to refine it.
Mark Mills:
And there is nothing that exists in our society that doesn't begin with a mine somewhere on the earth. Every one of the metals and materials on the periodic table we mine. The only things we don't mine are things like wood, which we grow, you know, and leather products. We grow that. But our modern society is based entirely on mining.
Mark Mills:
So if you were serious about this and people are now finally getting serious about it, you would count up the tons of stuff that you have to dig out of the earth to create the same road mile, same hour of heat, same hour of light, same hour of computer time. Not the electrical energy required. Not the energy required, but that's a different metric. Count the tons of rock you have to dig up. Now, the tons of rock are determined by something that is sort of an arcane geology term, which is the ore grade, the percentage of the mineral you want, that is in the rock.
Mark Mills:
So ore grades for iron ore to make steel, for example, are very high. That is, iron ore occurs with iron as a percentage of a high as 50%. It can have an iron mine when it digs up a ton of the earth, or half of that is actually iron itself, which is incredibly high concentration for most metals. The concentration of the metal in the rock is a couple percent, 1%, 0.1%, 0.001% gold state. The extreme one, gold occurs at a, you know, thousandth of 1% of gray at the rock, the ore.
Mark Mills:
So you dig up, you know, kilotons of rock to get ounces of gold. That's why it's expensive. It's not expensive because. Because it's special. That's why it's expensive.
Mark Mills:
You know, it's rare and it's inert, but it's special because it's rare. And because it's rare, low concentrations takes lots of rock, lots of energy, lots of mining to get, or you have to crush lots and you have to, by the way, let's finish the food chain. You got to dig up the rock. Let's use copper. Copper.
Mark Mills:
The average copper ore grade is about 0.2%, 0.2%. So you can do the arithmetic on this. If you need a pound of copper, you got to dig up. You know, you're pushing tons, tons of rock now. You have to crush the rock.
Mark Mills:
You have to grind it up with big machines, by the way, which are powered by coal and natural gas and diesel fuels, to crush the rock. Then you have to dissolve the rock, literally dissolve the rock with chemicals, which are petrochemicals, which are produced with hydrocarbons, with lots of energy, to extract the minerals from the wheat, from the chaff, if you like. Then you have to further refine it, which takes energy, and then convert the refined metal into the specific form you need it for the product, because in many cases, the form you need it in has to be an alloy or combined with something else is not the same. Lithium is just like that. You have to go from raw lithium to lithium carbonate to the lithium class of lithium you put in the batteries.
Mark Mills:
All that takes energy and time so we can distill it to a couple of examples. We know this is true because it's in the geology of engineering, the world we live in. So, roughly speaking, at the overall level, to deliver the same unit of miles on the road, to deliver the same unit of heat or light or computing power, we have to increase mining for metals, the tonnage of stuff, by roughly tenfold. That's 1000% in, 1000% increased, depending on the metal. It ranges from 400% to 7000% increase in mining to deliver the same useful road mile, the same useful computing method.
Mark Mills:
The world is not now mining enough material to do that. Nor are the world's miners planning to mine enough material to do that. Nor are the world's governments planning to eliminate all the rules and regulations that make it difficult to open mines, especially in the United States. The battery in your electric car, the one you had before, like a Tesla class, it's 1000 pounds. A thousand pound battery requires roughly 500,000 pounds of mining somewhere on earth to make that one car.
Mark Mills:
And that mining activity and all the rest of the supply chain I described, leads to carbon dioxide emissions. So this is the dirty little secret. It's not only requiring massive disruption of the earth ecosystems, potential for pollution, social disruption, environmental impacts in countries that want to do the mining, like in Africa, South America, China, but it leads to CO2 emissions. So the assertion that an electric car cuts CO2 emissions is a guess and an estimate. It does reduce CO2 emissions in most cases.
Mark Mills:
But in almost all cases, we don't know by how much. And we also know that it's never zero. And more importantly, we know that there are many cases, for many parts of the supply chain where the electric vehicle will actually emit the same or more CO2 than the internal combustion engine when you make a real world estimate of the emissions upstream. So when Norwegians buy their Tesla, they love their Teslas. Roughly speaking, before it's driven 1 mile, that individual Norwegian has emitted 40 tons of CO2.
Mark Mills:
By virtue, they get to own the CO2 missions because they bought the car. And if they had just driven a highly efficient internal combustion engine over its lifespan, it will emit about 40 tons of CO2. So as a CO2 trade, it's a kind of a silly trade, because you're pretending that the emissions aren't happening. Of course they're happening. They're happening elsewhere and we're told it's just 1 AtM, it's all the same atmosphere.
Mark Mills:
So if the CO2 emissions are happening in China where they burn coal to make the aluminum, that's essential to make the EV because EV's have more aluminum, not just more copper because the battery is so heavy you have to lighten the vehicle with a lot of extra aluminum. Which I'll state again, two thirds of the world's aluminum is made in China on coal fired grids. We're not planning to expand our aluminum foundries in America again. We've abandoned them years ago and nor is Europe actively expanding its aluminum capacity sufficiently. China is however, and they're also expanding their coal fired generation.
Mark Mills:
So we've got this entire panoply of myths built around the sustainable energy transition that are literally myths. In some cases they're not just. They're not myths, they're just misunderstandings. In some cases there are misperceptions. In some cases it's myopia to continue with alliteration and in some cases they're just silly.
Jim Puplava:
Misinformation too.
Mark Mills:
Yeah, and misinformation. Yeah, absolutely.
Jim Puplava:
Let's talk about the implications of this because we began where the IEA is saying stop exploring for oil, stop doing this, we're going to make this transition. We have ESG, which is saying we don't want you doing this kind of stuff. In fact, I think the SEC is looking at these companies and they're going to have to start reporting their ESG things. So at a time we need more minerals, we need more electricity, we need more things to make this work. We are saying we're starving the industries that we're going to need to make this happen.
Mark Mills:
Yeah, well, yeah, you said it right. That's exactly what's going on. Now there's some fuel money being pushed into this as well because the Inflation Reduction act contemplates giving loans and grants to some miners, some mining, it contemplates providing incentives to repatriate manufacture for these things. So it's there, I mean there's a genuflection to that reality. But the scale of what's being offered is trivial compared to what's needed.
Mark Mills:
First of all, millions and tens of millions of dollars versus billions and hundreds of billions required. And more importantly, there's a, we'll call it an operational schizophrenia to it. By that I mean on one hand we have the inflation Reduction act saying, we want you to build factories here again like chip fabs and lithium refineries, and we'd like you to open a lithium mine in America and well give you a loan to do that. Great. I'm not a huge fan of subsidies.
Mark Mills:
I think it's a bad idea. I'm not a huge fan of the government picking winners and losers. But sometimes you have to, to stimulate things. Ill stipulate, you know, there can be compromise in that. But the problem is at the very same time, let's do mining.
Mark Mills:
As an example, this administration has a multi agency working group under the auspices of the Council of Environmental Quality and the White House looking at mining. And they're making recommendations as we speak and started at the end of last year and continue this year on how to improve and reform in the mining sector. All of it? Not some of it. All of the recommendations will make it more difficult and more expensive to mine in America.
Mark Mills:
All of them. There is no pulling back. There's no rationalization of the kinds of legislations or regulations needed to make these kinds of industries expand when money is given up, for example, to incentivize the construction of a chip fab. At the same time, they're incredibly electric intensive. The famous tip chip company TSMC in Taiwan, that company alone, I think, uses a fifth of all of Taiwan's electricity.
Mark Mills:
The manufacturing of chips is massively electric intensive. We have some of the cheapest electricity in the world in the big manufacturing capacities. But the administration's policies are going to make electricity more expensive and less reliable, simultaneous with giving money to industries to bring electric intensive businesses here. These are incommensurate goals. They don't work well together as obviously what incommensurate means.
Mark Mills:
It just is a very bad idea. So as a consequence, I think we're already seeing the slowing down of any appetite to apply for deploy the money. There's been some pullback, not to mention the fact that the legislation and the implementation of the legislation for loans or grants to industries have all kinds of strings and hooks in them. ESG hooks, Dei hooks, social equity hooks, who you have to give money to. In many cases, the social programs have merit at sort of a philosophical level, but they're operationalized in ways that are extremely difficult, in fact, onerous to most businesses and are antithetical to a free market.
Mark Mills:
So we have rhetorical virtual signaling to what we need. There's an acknowledgement we need the materials and the mining, for example. We need more electricity. AI is going to need lots more electricity. Elon Musk has been running around the country for a year saying we're going to need lots more electricity, or for two years.
Mark Mills:
And he's been proven to be right in the data. But what you don't have going on simultaneously is the significant expansion of the capacity to produce electricity at scale to your point, and reliably. Reliably means that if you have a data center that needs electricity 24/7 that they either have to have backup oil fire generators or running on gas turbines, because when the wind's not blowing, you're not going to be able to store those quantities of electricity in batteries at a price anybody can afford. And we know that because we know the engineering arithmetic. It doesn't require being naive or by being profoundly insightful to guess the real costs.
Mark Mills:
We know what it costs to build batteries is public data. We know it'll cost build batteries two or three years from now because they're not going to change significantly. And we know how much electricity is needed to run 100 megawatt operation. And if you have to run it for hours at a time, constantly, not episodically, not once in a while, but continuously, you start talking about quantities of batteries. If you're doing this at a regional level that start to approach trillions of dollars, I mean, staggering numbers of it's just unimaginable, and we aren't printing enough money to do that.
Mark Mills:
And the claim all the technology will get better, it'll make it cheaper. Of course it will get better, and it will get a little cheaper, but not profoundly. None of these things are on exponential cost declines. In fact, almost all of them are now asymptoted to their low cost. And some of them are all rising in cost as the input costs go up for materials, and, of course, as input costs go up for capital interest rates.
Jim Puplava:
This is anecdotal, but Taiwan semiconductors. Building a new plant in Arizona. One of the reasons they have two nuclear power plants.
Mark Mills:
Well, sure, cheap power. Cheap power, good labor. Good labor policies in the state, by and large, access to engineering talent in the region, not just the university system there, but the neighboring university systems. But yeah, cheap power. Cheap power matters.
Mark Mills:
You put manufacturing. The repatriation of the petrochemical industry was in the news a lot a few years ago. It's not been in the news as much. Our energy costs to make polymers, which are essential to modern life, are the lowest in the world. So there's been a mass migration into the United States of building new petrochemical capacities to produce this for the world.
Mark Mills:
It's a big export market for the United States creates great jobs. In fact, it creates better jobs in building data centers, because data centers, once they're built, the jobs are in the construction. They're essentially no jobs to operate it. These are pretty much big autonomous beasts that have a few dozen people that have to operate it.
Jim Puplava:
One of the things, and when I say this, people think I'm nuts, but I could see in the direction that we're going right now as these AI, Google, all these companies are heading in this direction. And I think back in 2022, in the summer when we had power outs and our governor here said, hey, if you have an Ev, just don't charge it from nine to nine, don't do it. But I could see electricity rationing. What are they going to do when all this stuff starts to converge? At the same time, these data centers, these cloud, and we're not doing the mining, we're not building nuclear power plants.
Jim Puplava:
We're going to wake up, I think, one day like Germany, and just say, oh my God, we need to reverse this.
Mark Mills:
Well, that's what I hope. We don't go as deep down the rat hole as Germany has gone. Some states already are.
Jim Puplava:
Certainly my own state.
Mark Mills:
Yeah, California is like Germany, deindustrializing. New Jersey, significantly deindustrialized because it's pledged to follow the California path. So it has the effect you'd expect. By the way, two things happen. You do reduce the amount of energy used because the overwhelmingly more intensive, energy consuming sectors are the industrial.
Mark Mills:
So if you drive industries away with high cost energy and high cost labor, but especially high cost energy, you achieve your goals of reducing your energy intensity, but you also get lower value jobs, so the economy slows down. So I know the data for Germany, New Jersey and Germany, they're very similar. California, I think, is similar, that the growth rate of the economy is slower than the US at large. And the only part of the California economy that's experienced significant growth rate is, as you know, is the one pocket of the Silicon Valley software large ass. And that domain is the outlier if you take that out.
Mark Mills:
Because most of their energy intensive activities, the manufacturing, the stuff they invent, takes place overseas. The data centers that they inspire, are mostly not in California. So the energy intensive part of that industry is outsourced outside of California. So California has the illusion of becoming more energy efficient, but it's doing it on the backs of deindustrializing and moving its economy, except again, for Silicon Valley, away from high value manufacturing jobs that not only pay high wages, but ripple into the economy. To lower value, lower wage service jobs, which is why you get these distortions where the government wants to inflate what you have to pay, low level, entry level service jobs.
Mark Mills:
Well, this is a vicious cycle. I don't think every state's gonna like. Not every state's following that path. I think the United States by and large, the american political class and american citizens won't like to become like Germany and California at large. I just don't think that this is just, this is not physics or engineering.
Mark Mills:
I'm now engaging in political psychology as a. As an immigrant that came from Canada to America. I think Americans are different. I don't think they'll put up with it en masse. And we're already seeing evidence that's true.
Mark Mills:
Right. People say they want certain things, but they're not willing to pay the price for it. Europeans seem to be far more willing to pay a far higher price than Americans to try these things out for size. They're now revolting, too. But the revolt's very difficult to effect in Europe because, as you probably know, the political party's differences in many of the countries with respect to energy policies are non existent.
Mark Mills:
But they both are genuflecting to the need for an energy transition and pushing for high energy costs. I think it'd be very tough to pull that off nationally. In the United States, I think we'll see political blowback. I think it's starting. I think it's already happened to some extent in California.
Mark Mills:
It's beyond obvious. The reason that your governor stopped the shutdown of diabolic canyon was that he was told. I'm sure I don't have this firsthand, but I'm sure he was told by the California grid operator, if you do that, we'll have rolling blackouts constantly. We can't maintain the lights. So the rationing comes by forced rationing, rolling blackouts and brownouts.
Mark Mills:
So instead they keep it running and simultaneously allowed the permitting by circumventing state rules of more gas turbines to make sure that when the sun sets, because a lot of solar power in California, that you have the ability to meet the consumers industry's electric demand. So it doesn't mean you can't do both. My point is they're doing both. There's been a quiet, slow drift to realism in California, a faster drift towards realism in Germany. Canada's undergoing a bit of a political revolt right now.
Mark Mills:
They have an overt carbon tax in Canada which is very high. I forgot the exact number. It's something on the order of several thousand dollars per middle class household in after tax income. Or put differently, it's like 20 or 30% of their after tax income is being taken away by a carbon tax, which is achieving nothing. I mean, it's nothing.
Mark Mills:
There's no dumber place to use solar power than latitudes. Most of Canada represents, by the way, the same latitude as Germany. Let me give you an example of the operational insanity of putting solar panels in places like Germany or Canada or northern United States. The capacity factor is a measure of merit for power plants, the period of time in which the power plant is available to produce electricity. So a really good gas turbine has a capacity factor.
Mark Mills:
So like 75 or 80%, counting maintenance and fuel, it's available to operate 75, 80% of the time. Nuclear plants have 90% capacity factors. They can operate 90% and often do operate 90% of the time. In theory, a solar panel has a capacity factor of about 30%. It can operate 30% of the time on average, in theory, right.
Mark Mills:
But depends on where you are, how much cloud cover, and what latitude. The more northern you go, the longer. You know, shorter your days in the winter. Right. Longer in the summer.
Mark Mills:
So the average capacity factor for solar panels that's installed in the world is around 20%. So they're available fifth of time. So you could do the math in your head. That would mean to replace a watt of gas turbine, I need four watts of solar power. So I need.
Mark Mills:
It's actually four times more expensive. But more importantly, the actual capacity factor in Germany of their operating solar panels is below 10%. It's about 6.5%. So this means. This is crazy.
Mark Mills:
It means you spend $100 on a solar panel and you're getting essentially dollar ten worth of power value in capital terms. That's why the cost of electricity goes up. It doesn't matter if the sun is free. Panels aren't free, and they cost a lot when you have to scale them that way. So I think those are irrelevant realities.
Mark Mills:
When solar is a minor share of your electricity at the state level or national level, they become very relevant as the percentage rises, which is what's going on in California. It's what's gone on in a lot of Europe. And most of the serious analysts are now asking more different questions. There's an odd curve. The marginal cost of electricity from solar and wind as you scale them up, declines as you might imagine it would.
Mark Mills:
You get a little better at it, better at operating, installing it. But then it gets to a point where it constitutes a significant share of the energy you need in any given day, you have to begin to do something to operationalize the episodic nature of the power. Like operate other power plants, build other ones, build more transmission, build batteries, all those things that cost money. So the cost curve that tilts back up the net actual real cost of its own electricity as the penetration rises, curves up just because of operational realities. So there's an optimum point.
Mark Mills:
An optimum point. The share of your energy you get from episodic power varies geographically how far? North or south, where it's windy, not windy, those factors. But one thing that's very clear, it's not all your power. It's not even half your power for 99% of the parts of the world that humans inhabit.
Mark Mills:
In trying to do it, though, you cross that chasm, you start increasing costs of energy. And that's when you get, I think, not only just a deindustrialization phenomenon, which makes people unhappy because it jobs, it's the economy. Taxes have to go up to make up for the difference. So it's like the proverbial frog in the pot, turning the temperature up on the boiler for the frog. It's not like it's suddenly ugly.
Mark Mills:
It just gets slowly ugly until suddenly it's really obvious. And that's sort of where you are.
Jim Puplava:
Well, my own state's example, we have the highest unemployment rate. We have the highest income tax rates, we have the highest utility rates. And, I mean, it's the jobs and the companies that are leaving the state, and we're getting low paying jobs, and it's reflected in the tax revenues. Well, Mark. Well, Mark, as we close, we've covered a lot of ground.
Jim Puplava:
Tell our listeners about your new national center for Energy analytics, which you're the director.
Mark Mills:
Yeah. You know, I've been asked a lot over the years as I write and talk about these issues, why we're, you know, why there's so much misinformation out there. There's disinformation, too, but a lot of it's just misinformation. People are not aware of facts. We can have debates about intentions, and we can debate about aspirations, but it doesn't matter what your reason for avoiding carbon dioxide is.
Mark Mills:
It doesn't change the thermodynamics or the economics of batteries or cars, how fast or how hard it is to charge a car fast. I mean, those underlying factors. So what's happened is to compress it into the bottom line is a group of individuals with deep pockets, including trustees and the organization of the Texas Public Policy foundation, which is a state based think tank, said they would back the creation of a new national energy think tank. This is the only to my, I think the brag. I think there is no other independent public policy think tank that focuses only on energy, not environment and therefore energy.
Mark Mills:
Not on policy in defense, therefore energy not an economics and therefore energy, but flip it. Energy and its environmental effects. Energy and its geopolitical effects. Energy and its economic energy, its social. So since energy is essential to civilizations operations, the absence of a think tank that lends on the world is just energy.
Mark Mills:
First cost, availability, technology doesn't exist. Thats what we created. So we have over a half dozen scholars now. We have an impressive advisory board. People can find the if you type in the National center for Energy analytics, the website will come up in Google.
Mark Mills:
If you type my name, it will come up Mark P. Mills. So we just launched it. It's brand new last month. We'll expand our programs and our footprint.
Mark Mills:
We'll launch not just studies and articles, but engage in the public policy debate, all of it from the perspective of trying to put up incredible solid information. Frankly, to push back on the economically destructive execution of the narrative that we have an energy transition. We don't plan to debate your motive for that. I mean, there's lots of reasons people don't like oil and gas and coal. I get that they're allowed to have their opinions and you know that all expression, but you're not allowed to have your facts.
Mark Mills:
So let's at least get the facts out about how the world uses these energy sources and how hard it is, or in some cases easy to replace them in some applications. So we're sort of realism based pursuits. The website has a quote from Philip K. Dick, which I'll paraphrase as saying that reality is that which continues to exist when you stop believing in it. So it's sort of our mantra that I want to push reality out and then we can debate aspirations and what the cost of those aspirations are.
Jim Puplava:
Well, I wish you much success. Hopefully you can return reason to the mythical mandates that we seem to continuously pursue. Mark, thanks so much for being so generous of your time. It's always a pleasure having you on the program.
Mark Mills:
Thank you for the generous provision of a time to an opportunity to preach the gospel of reality. Thank you.
