Mark P. Mills joins Brian Anderson to discuss why a true energy transition is impossible.
Audio Transcript
Brian Anderson: Welcome back to the 10 Blocks podcast. This is Brian Anderson, the editor of City Journal, and today we're joined by Mark Mills. We're going to discuss energy and technology and some of his recent stories for City Journal, including his most recent, which we titled will never have an energy transition. Mark is a City Journal contributing editor and he's the executive director of the National Center for Energy Analytics. He's also a faculty fellow at Northwestern University's McCormick School of Engineering and a co-founding partner in Montrose Lane, which is an energy investment firm. Mark is the author of a number of books most recently The Cloud Revolution: How the Convergence of New Technologies will Unleash the Next Economic Boom in a Roaring 2020s. His articles have been published very widely, not just City Journal but Wall Street Journal, Forbes, USA Today, Real Clear Politics and many other outlets. Mark also served in the White House Science Office under President Reagan and early in his career he was an experimental physicist and development engineer at Bell Northern Research. Mark, great to have you on. Again, always good to get your perspective on these energy and economic issues.
Mark Mills: It's great to do back Brian, I like framing my articles for City Journal as stories because I think of them as stories telling the truth in a story.
Brian Anderson: Well, there is a kind of narrative consistency to these pieces. In a way, they're a sustained argument. Let's talk about the most recent one. President Trump recently boasted that he'd put an end to the Green New Deal and other Biden energy policies, which were quite extensive, but voices in the press continue to tout the transition to green energy as inevitable. They often cite other modern advancements, for example when a newly adopted technology makes an older technology obsolete as happened with smartphones eclipsing landline phones. I wonder, do those comparisons hold up when it comes to energy? Are we on the verge of a complete transition to new forms of green technology and currently, how extensively have we adopted green alternatives and green fuels into the overall energy mix?
Mark Mills: The short answer is no and no. Let me back up to one observation you made and I covered in my piece, which is the declaration of the end of the Green New Deal or as Trump hyperbolically called it, of course, the “Green New Scam.” I framed it as a question, not just a political question, but I was trying to illuminate here is the importance of understanding facts and what words mean. There is no energy transition, there hasn't been an energy transition, there will not be an energy transition if we look at the facts and the data. One fact, not to stay on the politics, is that Trump did not successfully kill the Green New Deal because it's embodied in two other places. He killed part of it, which is a de facto EPA mandate, and even that hasn't been killed yet because it's enshrined in legislation and the rulemaking at EPA, so they're underway and trying to stop the EPA from forcing people to buy only EVs in half dozen years from now.
Whether you think that's good or bad I think is generally bad for governments to do that obviously, but he hasn't touched the Inflation Reduction Act, which of course is the epicenter of the so-called Green New Deal and the Democrats were eager to brag that the passing of the Inflation Reduction Act, after it passed, was the instantiation of the Green New Deal as energy transition. They called it that if you like. So the politicization of this is heavy and deep and the subsidies are enormous and they have not been repealed. Some of them have been suspended, but they have not been repealed. At least a dozen states have their own Green New Deals, enshrined in legislation which have not been repealed.
Brian Anderson: We have an essay forthcoming in our spring issue by Judge Glock on the Inflation Reduction Act and going into detail about how much of the money was really poured into green technologies, and a lot of these are already being embedded in different projects and unwinding it is going to be very, very difficult.
Mark Mills: It'd be very hard politically, but what I wanted to explore was a different question whether we unwind it or not, and I obviously hope the spending gets unwound because it's something on the order of one to two trillion dollars of undeployed capital that's about to be pushed into markets, which is very inflationary to replace stuff that's already cheaper, but that's an economics argument. I looked at the history and, if you like, and the physics argument is no matter how much money you spend, will we have an energy transition? The word transition has a meaning. It means you stop using something and transition to something else. You stop using landlines, you use a cell phone. That's a transition. You've transitioned from landline to cell phones. The real question then is the energy transition framed is we have to stop using hydrocarbons, fossil fuels, oil, gas, and coal. Are we doing that?
Well, the answer is no. We're still using more hydrocarbons, more oil, gas, and coal today than at any time in history and over the last two decades of something on the order of $10 to $20 trillion of collective spending by Europe in the United States to avoid using hydrocarbons, we're using more of them. We're also using more windmills and solar panels too. That's true, but we're not transitioning from oil, gas, and coal to windmills and solar power even have we spent something close to $20 trillion and a lot of mandates and so forth, and the broader question is have we ever transitioned from any energy source over history? I did a deep dive. It struck me that I had never really thought about figuring out whether or not we've transitioned from say, grains and horses and we use humans, have used slavery as a power system, as soon as I state it it's obvious why and how, and watermills and windmills and burning wood and animal fats for illumination for centuries, for millennia, did we transition from any of those? Here's what you find out when you look at the history and the data, and it's not so easy to gather this data as you might imagine, we haven't transitioned from anything except one. There's only one energy transition and all of human history. We use more, grains are a fuel we use for horses and working animals and grains, of course, were the fuel used for slaves for millennia, sadly, and if you just look at the issue of working animals, for example, the world has more working animals fueled by grain now than any time in history. It's crazy. There's something like 200 million working animals, horses and donkeys and stuff used in the world today for power purposes, dragging, carrying, moving. Peak horse in the United States for calibration look like about 25 million horses before the industrial revolution.
So we had, if we of course don't use horses for labor much anymore. Police use them, that kind of thing. But pretty rare. The peak horse in the United States we managed, but the world didn't manage peak horse and we didn't even manage peak grain in the United States. So the grain that horses we used to grow to fuel horses for industrial and transportation purposes. Well, today the United States uses 300% more grain for transportation than we did at Peak Horse. Of course, I'm talking about converting the grain to ethanol to drive cars. Let's, this for grains for slavery. There are more slaves in the world today in absolute terms, sadly than there were at any time in history because there's more people, but animal fat and wood and watermills and wind power, all of them, all of them, we use more now than any time in history at any previous peak use of those energy sources. We use more as a percentage of our energy, they're lower because we now use a lot more hydrocarbons which weren't used extensively until the last two centuries. It's really kind of interesting when you sort of go through the, you rendering animal fat to make candles, torches and elimination. This is a very, very old business and industry, a very, very old use of a biological fuel, biodiesel if you like, but we use more of it now than ever, not less.
Brian Anderson: It's really, it's quite striking when you pose it like that. I don't think people are fully aware. It's just a matter of the growth of human population ways and the kind of insatiable need we have for energy. If the argument you're making is correct that a true energy transition doesn't seem to be possible, at least right now, what is the value of pursuing or developing green energy or should we just allow the market to dictate its deployment or is there a role for government in encouraging alternatives like wind and solar?
Mark Mills :
I guess the short answer is there's no value in the pursuit as it's framed because again, coming back to words have meaning, what people are calling green energy isn't green by what any rational definition of what green means. Green, meaning using less land for example, or using less primary materials, disrupting less of the physical environment that we live on or even producing lower levels of pollution from chemicals needed to produce the materials to make the machines, all those metrics, green energy, it takes us backwards, not forwards.
The only metric where it takes you slightly forward is in the idea of reducing carbon dioxide emissions and it doesn't really make a significant difference. Reducing CO2 emissions far less than most people realize more importantly has had no impact. All this $20 trillion again has had no measurable impact at the global level of reducing the continual increase of both hydrocarbons and atmospheric emissions of anthropogenic manmade carbon dioxide. Those are both rising, have been rising, continue to rise. We've not measurably or materially slowed anything down.
So the question that one should be asking is given how expensive this is for consumers, energy costs are rising in industrial markets because of these government mandates and subsidies. Our real costs are rising in terms of taxation of citizens to subsidize these things for industries that build them. Are you getting what you thought you been promised and the question answers itself. You're not. When you look at the data, which is a long way of saying no, it's not a good idea for governments to have this aspiration. It is a good idea for governments to think that there might be new better ways for energy technologies to be produced and deployed. Technology does get better continually, sometimes faster than incrementally.
Question about how governments should stimulate more of that coming faster is sort of the same question you have on every other industrial market. Energy's no different. What is the effective role for government for bringing along innovation that's foundational and avoiding the kleptocracy and distortions of markets that come out when you create massive subsidies? I don't mean targeted short-term subsidies, but when you create enduring massive subsidies or tax, enduring massive tariffs and taxes, those kinds of things, they distort markets and we have massively distorted energy markets to the detriment of reliability of energy supplies and their costs. That is true is clearest in Germany in the United Kingdom where they further down this ostensible, still called the “energy transition,” nobody's given up on the words, so that pursuit in the United Kingdom and Germany has led to roughly a tripling of cost of electricity for those countries in a rapid de-industrialization and economic stagnation is these are very, very expensive and damaging aspirations.
Brian Anderson: Well, certainly one of the main goals of Trump 2.0 is to bring back some form of the manufacturing economy in the US both for national security reasons but also for political and economic reasons that is viewed as a problem for the base of this party that a lot of people were employed in manufacturing and fewer are today. I wonder how that squares with the plans that the administration has afoot. In a state like California energy costs I guess are way above most states in the United States. The Democratic Party views California in a lot of ways as the future.
Mark Mills: So California is the future. It's true. In fact, if you Germany and United Kingdom are the future, which is de- industrialization. The only thing that saved California, the macro level is the fortuitous fact that Silicon Valley is there and its economic engine has hidden the destructive impact of the rest of California's policy. It's a “lighter” industry or appears to be a lighter industry. The pivot on which we can understand the truth of what happens with energy, so the Silicon Valley effluorescence of artificial intelligence and software and industries that appear light, of course all that is instantiated in data centers, which are physical pieces of hardware, they’re buildings that are the size of warehouses, there's thousands of them, they're full of computers where the software lives. Most of those data centers are not being built in California. Most of the hardware that goes into the data centers is not being built in California.
The steel that those data centers need and the concrete that those data centers need are incredibly energy intensive and they're not being produced in California. They're all being increasingly produced, all the above, in places, states or countries where energy is cheap. California is an outlier because of the existence of the headquarters, if you like, of the modern revolution in our economy, which is the digital economy. California’s model is proving what, again, Germany and United Kingdom have proven that if you make energy expensive industries will leave that are energy intensive. Gee, what a surprise and that we're rediscovering that it actually matters where you manufacture these things. So we've got this bipartisan enthusiasm for reshoring manufacturing, but apparently nobody told people who are enthusiastic that those very industries need enormous amounts of power and highly reliable power that has to be inexpensive and the policies that make power expensive and unreliable are antithetical to those goals. The Trump administration is trying to fix that, but the truth is states are also at the front lines of fixing that and a lot of states, proverbial, so-called red states are and have fixed it are trying hard and I think some of the blue states are, but the way they do it is sort of hand handed. They let electricity stay expensive for consumers but give discounts to the big tech companies that's not going to go over so well as it becomes clear to consumer advocates.
Brian Anderson: Yeah, that's probably worth a future piece and an interesting aspect of this whole debate that hasn't gotten as much attention, I think. You've also written, and this is a related question mark about the Chinese AI model, DeepSeek’s advancements seemed if we're to believe what was reported to require far less energy than rival ais with kind of comparable results. Though the news, the DeepSeek news caused energy stocks to plummet for a while because the assumption was, well, if AI can be produced with far less energy, we're not going to need all of this investment in energy. Yet in that earlier piece on DeepSeek, that the concept of Jevons paradox suggests that improvements in energy efficiency can actually lead yet again to more energy consumption. I wonder if you could talk a little bit about that and how does that apply to what DeepSeek has seemingly done and what are the broader implications for global energy demand?
Mark Mills: As you noted, this is the enduring confusion over the 19th century economist Jevons who wrote the paper “Jevons Paradox,” which I'll note for the record because I think this is true, what kids call “humble brag,” Huber and I wrote in The Bottomless Well 20 years ago this year when it was published, we wrote about Jevons paradox and if you look at Google Ngram, that's track currents of words being used in the public discourse, no one had really talked about, noticed, or written about Jevons paradox prior to when we highlighted him in our book. We wrote a chapter on the efficiency conundrum, which appears to be a paradox and that is that when you increase the energy efficiency or efficiency in general, but in this case Jevons was specifically talking about the energy efficiency of coal used to produce steam, you get more demand for the input materials, more demand for coal resulted from improving the efficiency of steam turbines and steam engines rather.
Back in the 19th century, Jevons knew that wasn't a paradox. He wrote his language as we put it in our book. The seeming paradox of that if you improve the efficiency of these coal engines that we increased demand for coal, Jevons was amongst the economists who in the 19th century was worried that England would run out of coal. And of course we revisited that silly worry a century later worrying about peak oil ,and there's still people worry about that. But Jevons Paradox is probably most deliciously exhibited in computing more than any other application of energy efficiency, because of the astonishing improvements in energy efficiency of computing that have occurred since the dawn of computing. The point that I'm making, that we made in our book, is improving the energy efficiency of computing as DeepSeek did for AI will increase the demand for our energy to do the computing because you make the computing easier and cheaper to use, a lot more of its use, then the increased use vastly overwhelms the reduction in energy use per activity because there's more activities.
Easy way to illustrate that would be to state the inverse if the smartphone that everybody listening has, if that one phone operated at 1984 computer energy efficiency, that one phone would use more electricity than the Empire State Building, but there's billions of phones because we made them something on the order of a billion times more energy efficient, so we get the hundreds of Empire State Building’s worth of electricity used just by phones in the world and that's not counting the energy used by the data centers that serve the phones or the networks that serve the phones. That's what Jevons paradox does. So back to AI: AI is incredibly energy hungry and so were the first computers, so guess what's going to happen? We're going to make AI fantastically more energy efficient and it will lead to that increase in electricity demand, which it is already doing both.
Brian Anderson: Fascinating. Mark, it's always great to have you on to walk us through these complex and important issues. Our guest has been Mark Mills. You can find his writings for City Journal on his author page and we'll link to that in our description. You can also find City Journal on X @CityJournal and on Instagram @CityJournal_mi. If you've enjoyed podcast today with Mark Mills, please give us a five star rating on iTunes. Mark, it's always great to have you on and really appreciate the time. Great to be back. Thanks Brian,
Mark Mills: And thank you for the spectacular platform of City Journal to preach the gospel of reality. It's great.
Brian Anderson :
Yeah, that's a good way to put it. Yes. Alright, well thanks very much.
Photo by Chip Somodevilla/Getty Images
