|
The Trump administration is spending more than half a billion dollars to help prop up the dying coal industry. Its also weakening pollution regulations and opening up more federal land to coal mining. All of this isnt likely to save the industryand also isnt likely to do much to meet the surging demand for power from data centers for AI. Coal power is expensive, and that isnt going to change Aging coal power plants are now so expensive to run that hundreds have retired over the last decade, including around 100 that retired or made plans to retire during Trumps first term. Offering relatively small subsidies isnt likely to change the long-term trend. I dont think its going to change the underlying economics, says Michelle Solomon, a manager in the electricity program at the think tank Energy Innovation. The reasons why coal has increased in cost will continue to be fundamentally true. The cost of coal power grew 28% between 2021 and 2024, or more than double the rate of inflation. One reason is age: the average coal power plant in the U.S. is around 50 years old, and they arent designed to last much longer. Because renewable energy is cheaper, and regulation is likely to ramp up in the future, investors dont see building new coal power plants as viable. But trying to keep outdated plants running also doesnt make economic sense. The new funding cant go very far. The Department of Energy plans to spend $625 million on coal projects, including $350 million to recommission and retrofit old plants. Another $25 million is set aside for retrofitting coal plants with natural gas co-firing systems. But that type of project can cost hundreds of millions or even a billion dollars for a single plant. (The $25 million, presumably, might only cover planning or a small pilot.) Other retrofits might only extend the life of a power plant by a few years. Because the plants will continue to be expensive to run, some power plant owners may not think the subsidies are worth it. Utilities want to move on If coal power plants keep running past their retirement age, even with some retrofits, costs keep going up for consumers. Thats something that you really see in states that continue to rely on coal for a big part of their electricity mix, says Solomon. Like Kentucky and West Virginia, who have had their cost for power increase at some of the fastest rates in the country. In Michigan, earlier this year, the DOE forced a coal power plant to stay open after it was scheduled to retire. The DOE cited an emergency, though neither the grid operator nor the utility said that there were power supply issues; the planned retirement of the plant included building new sources of energy to replace it. The utility reported to the SEC that within the first 38 days, alone, it spent $29 million to keep the plant running. (The emergency order is still in place, and being challenged by multiple lawsuits.) The extra expense shows up on consumers bills. One report estimates that by 2028, efforts to keep large power plants from retiring could cost consumers more than $3 billion a year. Utilities have long acknowledged the reality that there are less expensive energy sources. In the first Trump administration, in 2018, utilities resisted Trump’s attempts to use emergency powers to keep uneconomic coal plants open. When utilities plan to retire a power plant, there’s a long planning process. Plants begin making decision to defer maintenance that would otherwise be necessary. And many won’t want to reverse their decisions. It’s true that demand for power from data centers has led some utilities to keep coal plants online longerand electric bills are already soaring in areas near large data centers. But Trump’s incentives may not make much difference for others. The last coal plant in New England just shut down years early, despite the current outlook for data centers. “Utilities do have to take a long-term view,” says Lori Bird, director of the U.S. energy program at the nonprofit World Resources Institute. “They’re doing multi-year planning. So they consider the durability and economic viability of these assets over the longer term. They have not been economic, and they’re also the highest-emitting greenhouse gas facilities.” Even if the Trump administration has rolled back environmental regulations, she says, future administrations could reverse that; continuing to use coal is a risky proposition. In most states, utilities also have to comply with renewable power goals. There are better solutions It’s true that the U.S. needs more power generation, quickly. It’s not clear exactly how much new electricity will be neededsome of that will depend on how much AI is a bubble and how much tech companies can shrink their power usage at data centers. But the nonprofit Rewiring America calculated that data centers that are under construction or in planning could add 93 gigawatts of electricity demand to the U.S. grid by the end of the decade. The nonprofit argues that some or even all of that new capacity could be covered by rooftop solar and batteries at homes. Cheap utility-scale renewable power plants could obviously also help, though the Trump administration is actively fighting them. Battery storage can help provide 24/7 energy. One analysis of a retiring coal plant in Maryland found that it would be less expensive to replace it with batteries and transmission upgrades than to keep it running. Temporarily saving a handful of coal power plants won’t cover the new power needs. It would add to air pollution, water pollution, and climate pollution. And it would significantly push up power bills when consumers are already strugglin. Real support for an “energy emergency” would include faster permitting and other work to accelerate building affordable renewable energy, experts say. “Making sure that resources can compete openly is really important,” says Solomon. “It’s important to not only meet the demand from AI, but make sure that it doesn’t raise costs for electricity consumers.”
Category:
E-Commerce
Below, Nick Foster shares five key insights from his new book, Could Should Might Dont: How We Think About the Future. Nick has spent the last 25 years working within companies at the very forefront of emerging technology, from Apple and Sony to Nokia and Dyson. Most recently, he was head of design at Google X. He has established himself as a leading figure in the field of Futures Design. In 2021, he was awarded the title Royal Designer for Industry, the highest accolade for a British designer. Whats the big idea? We need to have a conversation about the future, but not the kind youd expect. Humans have already talked at length about what the future may or may not hold. What we rarely discuss, and need to start addressing, is how we think about the future. By understanding the ways in which people process what lies ahead, we can all become better-equipped critics of the futures we are shown or sold. To design a better future, we need better futurists. 1. Thinking about the future is more important than ever. There is a palpable sense of curiosity, uncertainty, and anxiety about the future. Google searches for What will the future be like? have tripled since 2020. I think thats because we have experienced more change in the last hundred years than at any other time in history. A hundred years ago, we had not yet invented penicillin, and less than half the homes in America had electricity. My father, who lived in the U.K., has experienced both a world before the creation of vinyl albums and after the creation of ChatGPT. Even if I think about myself, there were half as many people on Earth when I was born as there are today. Much of this rapid change has come with baggage, and many of the things were trying to fix today are the result of insufficient thought about the future from previous generations. This needs to change. To start, we need to have a conversation about our ability to think about the future. I see the world through a design lens, but Ive spent my entire career around entrepreneurs, scientists, engineers, strategists, investors, business leaders, and policymakers, and it seems that no one is very good at thinking about the future. This skill is significantly underprioritized, underfunded, and underdeveloped in almost all of us. Our thinking about the future tends to fall into one of four main varieties: could, should, might, dont. 2. Could Futurism. Could Futurism is a way of thinking about the future with wide-eyed optimism. This is probably the most familiar and publicly embraced type of futurism. If you type futuristic into Google images, youll see over-the-top visions of flying cars, humanoid robots, and towering glass architecture. This is the kind of stuff you see from futurists on conference stages or when futurists are invited onto TV, and this is also how we typically experience the future when we go to things like trade shows or expos. This futurism is built around the modernist idea of growth and change through strident progress, mechanization, and industrialism. This way of thinking about the future has largely grown in parallel with the growth of science fiction, which has fed ideas into the minds of powerful leaders. Its exciting, escapist, and intended to shock. This futurism is built around the modernist idea of growth and change through strident progress, mechanization, and industrialism. But Could Futurism has weaknesses that are often overlooked. Just like sci-fi storytelling, this kind of futurism is heroic. It treats the future as a world filled with extreme characters having extreme experiences in extreme placesand often treads a bit closer to advertising than truth. It encourages us to think of the future as a place of extreme transformation, but doesnt talk about transition or interstitial change. While it prides itself on imagination, Could Futurism is also incredibly repetitive, as evident in those Google search results. This futurism represents a placeholder for deep thinking, offering simplified icons that we can drop into our slideshows and conversations. It lacks genuine, rigorous consideration. 3. Should Futurism. Should Futurism is focused on finding some sort of certainty and assuredness in the future. In the olden times, this was mostly built around things like soothsaying and predictions by people like Nostradamus or cutting open the belly of a goat and interpreting the shape of the entrails that fell out. This futurism can help us narrow down the future and concentrate on one dot that lies ahead. In contemporary society, the world of Should Futurism is mostly dominated by corporate strategy. It is built on the idea that we can somehow take data from the past and convert that solid line into a dotted line that leads us to a position in the future. Though often useful, the downfall of this thinking is that we are creating pieces of well-styled, well-executed numeric fiction. Those dotted lines on charts are not real facts. Once the solid line turns into a dotted line, it ceases to be data and becomes a story. Youll often find people in the world of Should Futurism making bold predictions or statements about things that are definitely going to happen. They love that quote from Wayne Gretzky about skating to where the puck will be. But knowing where the puck will be is essentially a story. This type of thinking tends to view the world as a system that can be decoded, converted into an algorithm, and then utilized to create simulations. But anybody whos put any money into the stock market knows that the dotted line heading into the future is just a story. Our world is volatile, stochastic, and ultimately unmappable. The idea of using historical patterns to project futures is remarkably unreliable. To use an acronym from the Army, our world is VUCA: volatile, uncertain, complex, and ambiguous. 4. Might Futurism. Might Futurism emerges from the idea of plotting multiple scenarios out into the future, kind of like chess. This thinking is embedded, to an extent, in all of us whenever were planning an event or thinking through eventualities. But Might Futurism became more formalized around the Cold War with organizations like the Rand Corporation and people like physicist Herman Kahn. It was also grown upon by people like Pierre Wack, who worked for Royal Dutch Shell, trying to run scenarios for the future of their business. The problem with Might Futurism is, it can get tangled up in jargon and diagrams that provide countless possibilities but no real answers. In todays world, this is referred to mostly by the term strategic foresight, and its probably one of our most popular modern forms of futurism. It has an awful lot of methodologies, matrices, diagrams, and techniques for thinking about the future as a series of decision trees. The problem with Might Futurism is, it can get tangled up in jargon and diagrams that provide countless possibilities but no real answers. Its also not very good at imagining things. We often think that certain things are ridiculous or unlikely, but just look at companies like Blockbuster, Nokia, and Kodak, which didnt anticipate what was coming for them in the future. Or, if they did, put them in the realm of near impossibility. That how our brains work. Thats why when we watch things like magic and illusions, they fill us with wonder. Imagining what might happen in the future and building a sufficiently broad number of scenarios is extremely difficult. 5. Dont Futurism. Dont Futurism is focused on what might go wrong or things that we dont want to do. We may refer to these things as dystopian. Fear is a potent storytelling technique, which is why fairy tales and rhymes often focus on what might go wrong if we choose the wrong path. We also see this in oppositional democratic politics, where you have a position of power and a party thats in opposition, who are there to point out the mistakes that might happen if we follow the rules of whats been put in place by the leading power. Its also found at the center of things like protests. There is a more nuanced form of Dont Futurism emerging today, which focuses on the externalities and implications of the things were bringing about in the world, such as how new services and technologies we embed in society will age. The problems with Dont Futurism are numerous. They force us into oppositional, divided factions. Theyre often polemic and call for immediate and often impossible action. And they dont integrate well with the people or industries they want to change. Dont Futurism is difficult because it often wants to be. Finding a balance between dont ideas and actionable change can be tricky. This article originally appeared in Next Big Idea Club magazine and is reprinted with permission.
Category:
E-Commerce
The return-to-office debate sees no end in sight. Workers still want flexible workand drag their feet complying with RTO, it was reported this week. Some workers have suspected such policies have been a way of companies saying: Dont like it? Quit. Turns out, maybe they are. A recent Fortune article, citing a 2024 survey of more than 1,500 U.S. managers, found that a quarter of C-suite executives hoped for some voluntary turnover after introducing an RTO policy. One in five HR leaders went further, admitting their stricter office requirements were designed to push staff out. So when the article started making the rounds on Reddit last week, the general lack of surprise was telling, and renewed discussion around worker suspicion that RTO goes beyond fostering collaboration. This belongs on the no shit sherlock subreddit, one user wrote. This should have been pretty obvious to any person with the ability to think objectively, another added. One suggested, The rest just arent admitting it yet. Their skepticism isnt misplaced. In fact, business leaders across the U.S. told the Federal Reserves Beige Book theyre banking on in-office requirements to quietly and cheaply trim headcount, all without having to play the bad guy. More than half of Fortune 100 companies now have a full-time office requirement, and research shows nearly 3 in 10 companies will demand five days a week in the office by the end of 2025. Thats despite almost half of workers warning theyd quit if remote work disappeared. To some, the ability to work from home is a perk equivalent to 8% of their salary, and not something they are prepared to give up without a fight. But those threatening to quit may have less bargaining power than they believe. A mass exodus triggered by RTO might seem like it wouldnt be in companies best interests, yet in fact, the opposite may be true. Forcing disgruntled employees to quit provides companies looking to reduce their workforce with an easy out, all without the need to foot the severance packages and bad press tied to layoffs. Rather than cleverly killing two birds with one stone, however, RTO mandates, as a workforce reduction tactic, often simply drains talent along with morale among remaining employees. At a time where employees are already disengaged at work, theres something to be said for a business strategy thats all stickand no carrot.
Category:
E-Commerce
All news |
||||||||||||||||||
|