Huge Capital Investment Required for U.S. Electric Industry’s Decarbonization

Yves here. The thesis of this article should not come as any surprise, that US electric utilities need to make very large investments in order to stop generating power from coal and other fossil fuel sources. The authors usefully make estimates of how large “very large” will probably wind up being. They predict foot-dragging and a death spiral as more consumers go off the grid as possible outcomes.

By Leonard S. Hyman, an economist and financial analyst specializing in the energy sector who headed utility equity research at a major brokerage house and has provided advice on industry organization, regulation, privatization, risk management and finance to investment bankers, governments and private firms and William I. Tilles, a senior industry advisor and speaker on energy and finance who headed utility equity research at two major brokerage houses and then became a portfolio manager investing in long/short global utility equities. Originally published at OilPrice

  • The industry needs to significantly increase capital spending, by 50% to 250%, depending on the resource mix.
  • Replacement of aging infrastructure and increased electricity demand necessitates this financial surge.
  • Potential outcomes include continued reliance on fossil fuels, forced action due to climate change, or a shift towards off-grid solutions by consumers.

When the COP 28 conference adjourned a short while ago, we concluded that the world’s electric industry would not decarbonize fast enough to make electrification the sure route to decarbonization. The principal culprits? Fast-growing Asian economies continue to build coal-fired power plants. Don’t get the idea, though, that American electricity producers will win any environmental stewardship prizes soon, either. They hope to get fossil fuel-produced electricity down from around 57% in 2022 to around 27% in 2050. All the S&P 500-type corporations are boasting of their goals to reach 0% carbon by 2050, however, they define it and whether anyone believes it. The electric industry, however, has to put capital-intensive assets into place ahead of demand, big projects that are hard to fake. Furthermore, electricity suppliers face two expensive challenges at the same time: rapid growth in electricity demand while operating in an increasingly hostile physical environment. Three challenges if we include decarbonization. The real question, then, is from a capital allocation perspective: what are utilities spending in relation to what they will soon need to spend (and likely finance). Also, when will they come to that realization and share it with the public? If they are not prepared, the whole decarbonization via electrification strategy goes out the window.

We estimate below the required spending to reach or not reach decarbonization goals over 20 years (rather than the standard 2050 endpoint) because of our expectation that climate events will accelerate the need for faster action. Due to the age of existing utility assets, virtually all assets will require replacement by 2042.  Demand is assumed to grow 2% per year, up from the current rate of 1%, due to accelerated electrification of the economy. All generation and storage costs are based on government estimates (from EIA and NREL). Transmission and distribution expenditures are calculated using current levels adjusted for 2% growth. Ranges bracket low and high-cost estimates and assume that all power sources simultaneously register high (or low) costs. Relative capital spending does not indicate the relative price of output because non-fossil fuels have minimal or no fuel purchase expense. (See Table 1.)

Table 1.  Average Annual Capital Spending over 20 Years Required to Finance Scenario (2023 $ Billions)

Scenarios Comment Average annual capital spending  ($ billions)
1. No change in generating capacity mix. Nuclear plants  built using current technology. $290-381
2. No change in generating capacity mix Nuclear plants built as small modular reactors $262-318
3. All fossil fuels replaced by renewables, balance of capacity nuclear. Nuclear plants built as small modular reactors. $316- 481

Twenty-year projections, at best, provide clues to problems and opportunities, based on current knowledge. Technology and the market will change over time, eventually rendering the projections obsolete. But large capital intensive businesses like utilities have to make plans, finance them, and ultimately build them. In short, putting infrastructure in place requires a long lead time. They often have to start work and modify plans as they go along rather than sit tight until everything becomes certain.

In 2023, we estimate that US capital spending to supply electricity came in at around $202 billion. Our numbers in the above table imply that electricity suppliers’ future capital spending will need to increase by either 50%, 100%, or 250%, depending on the resource mix chosen. The point here is that all of these percentages are big. And this current $202 billion annual capital expenditure level is way below our number to reach any key goals, whether to replace old plants without eliminating carbon emissions (scenarios 1 and 2) or to reach carbon zero (scenario 3). Note, too, that cost estimates are affected by whether nuclear power will be produced by conventional gigawatt-scale plants or by small modular reactors. Note, too, that no small modular reactors operate commercially in this country so their estimated costs may be shaky. (On the other hand, we do know the costs of renewables and storage, which appear competitive to or better than the small modular nuclear capital costs, so substituting one for another may not make a big difference on a capital basis.) As for transmission and distribution spending, we would expect that preventative or remedial spending will skyrocket from the levels used in the projections to improve the survival rate of facilities facing more violent weather and rising sea levels. (We also think the concept of coastal facility abandonment due to rising sea levels will soon be incorporated into future financial forecasts, perhaps as a footnoted contingency.)

The numbers tell us that the existing electricity suppliers will not provide the infrastructure needed to achieve decarbonization through electrification based on what they are spending now. Basically, the managers and owners of the US’s electricity supply have decided to talk the talk but not walk the walk. Does anyone really believe a promise like “I wIll be really good (carbon wise) by 2050, but in the meantime, please be patient?” What would Santa Claus think after getting a note like that? From this pathetic picture of lost opportunity, declining reliability, and timid incrementalism, we would draw one of three conclusions:

1. Electricity suppliers will continue to prattle on about an orderly transition to net zero and concern for consumer electric bills while working to preserve fossil assets, praying for the return of a Republican President like Donald Trump and, in effect, sabotaging decarbonization via electrification—their biggest opportunity of the century. This is the likely outcome over the short term.

2. Climate change and severe weather force action, which leads to a precipitously sharp rise in utility capital spending. This also means long-lasting boom times for businesses that produce and sell services and equipment to electricity customers and suppliers. This is event-driven, so the timing is uncertain.

3. Consumers (both residential and commercial) who can afford and want clean, reliable electricity turn to sources off the conventional grid. This exodus of the best (i.e. biggest) customers weakens the grid, creates energy inequity, and a two-tiered service that turns the public grid into a supplier of the last resource—possibly government-owned. This “death spiral” scenario is more likely to play out if industry capital investment stalls while reliability erodes.

Our advice? Ignore the PR, the smooth talk and the confidence that all will go smoothly. A relatively complacent utility industry has to gear up big time to confront both rapid growth and an increasingly hostile operating environment. Something big or something bad will eventually happen to disrupt continuity. And when it does, some people will make a lot of money.

Print Friendly, PDF & Email


  1. Mikerw0

    As Yves says, to NC readers (and again refer to the great series “Energy Destinies”) this is well summarized but not new news.

    More importantly, this is a recipe for significant inflation and a further separation of the elites from the masses. Why inflation? There is no way that the metals and mining industry is positioned to meet the global surge in demand that this would require. This can not be said strongly enough. Even putting aside the environmental impacts of increased mining, refining and smelting, we just don’t know where the ores are to accomplish this. And, using things like EVs as an example, it is not just rare earths like lithium that they need. They also need nickel (scarce), for example.

    One has to assume that the elites will continue to ensure that as we have power droughts that their lights, Acs, etc., will stay on. In periods of high demand it is not wealthy, politically connected neighborhoods (upper East Side in NYC) that have their power cut, hmm.

  2. Giordano Bruno

    There are some industries that function well under the profit motive and there are some industries that should be owned and controlled collectively. There is no reason energy should be oriented around profit. The Tennessee Valley Authority built and operates a massive energy system without ever having consumed a single cent in taxes. Capital for publicly owned utilities is raised with utility bonds that are tax-free and paid with future revenue, making them virtually risk free. There is no need for the profit motive to be anywhere close to this equation.

    1. Ben Joseph

      Basically, the managers and owners of the US’s electricity supply have decided to talk the talk but not walk the walk.

      Exactly. Why would capitalism invading utilities produce any improvements? There’s more profit in lack of grid maintenance than in modernization.

    2. Mark Dempsey

      The Sacramento Municipal Utility District (SMUD) is publicly owned. PG&E serves neighboring areas and is privately owned. SMUD electricity is 30% cheaper (at least) than PG&E. Also: no SMUD execs are consulting with criminal attorneys about facing charges of negligent homicide (for fires and gas main explosion); PG&E execs, paid orders-of-magnitude more, are.

      Socialism. It’s cheaper and works better. The temptation to skimp on maintenance and boost profits is simply too great when profit excuses all bad behavior (Thanks Milton Friedman!). Silicon Valley Bank fired its risk management people for the same reason.

  3. The Rev Kev

    ‘They predict foot-dragging and a death spiral as more consumers go off the grid as possible outcomes.’

    In this sentence, is there not the key to solving so many of these problems? If you had a massive campaign to have as many consumers go off the grid as possible, suddenly a lot of these other problems are not so tough or urgent. With less electricity demand, replacement of aging infrastructure could be done at a more reasonable pace and some areas, having gone off grid, may not need their infrastructure upgraded. Fossil fuel-produced electricity would drop automatically and would amount to a decarbonization to a large extent after the initial costs of upgrading all those homes. If you did this for everybody across the board, you wouldn’t have energy inequity nor a two-tiered service that turns the public grid into a supplier of the last resource. Of course if this happened, all those electricity suppliers would try to sue the Federal government for loss of profits, even though they had no intention in changing or investing in new infrastructure unless the Feds picked up the tab for it.

    1. Jams O'Donnell

      I’m not clear about what exactly is meant by ‘going off the grid’. Is this – using personal solar panel power plus heat-pump and battery technology, or using fossil-fuel supplies, or abandoning electricity use altogether, or some combination of all of those, or – what?

      1. The Rev Kev

        The former – making power generation local either by house or local neighbourhood. If the whole grid falls over the people in those homes will still have at least power to survive.

        1. bob

          You’re giving Governor Newsom way too much credit. Rooftop solar is a problem that everyone has solved and no one has deployed at any scale. “I would have had it all except for that meddling inbred governor!”

          There are lots of other states in the western US. Which one is doing it right? These systems are so simple and so cheap that they should be everywhere. Where are they? Show an example! And no, not a one off. You’re talking about the largest state in the US. The example should be something that can scale that large, even with the super powers of the Governor nearby. A small city in Texas even? Whattayagot?

    2. ocop

      If you shift the location of electricity generation to the customer-side (I assume this is what is meant by “going off the grid” and not a radical reduction in energy use) you are merely changing the type of the infrastructure challenge. That would still require massive spending on electrical distribution infrastructure/microgrids, (distributed) solar, (distributed) batteries, heat pumps etc. All without the economies of scale present in utility-scale infrastructure, so in the end potentially even less successful and more expensive.

      Somebody still has to spend the $$ (or not…). I think you’d see general deindustrialization rather than major industrial customers build their own power supply infrastructure to keep manufacturing in-place–what’s happening in Germany right now is probably partially illustrative.

    3. ISL

      In a country where 40% live paycheck to paycheck (or whatever % it is), and something like 60% cannot scrounge up a few hundred bucks for an emergency, and where a third do not own their own home*, off grid means the creation of great reservations of Americans with no electric power. A recipe for societal instability in a country with more arms than people.

      *solar conundrum – if you do not own your property, tenant-installed renewables are the owner’s property, whereas the landlord has zero economic incentive to invest in solar.

    4. bob

      Going “off grid” is more expensive in at least two ways, money and energy. More energy is going to be required to make all of the ‘stuff’ that will be required for each house. Instead of each house having 100 energy units delivered, they will need to produce 100 energy units, which will require equipment and more energy to produce that equipment.

      It’s much more inefficient. You can gain some efficiencies at scale. You’re going to lose all of those efficiencies going the other way, which then requires more equipment and energy to make that equipment.

      It’s just like batteries. When you add “battery” to the system, the battery itself uses energy that then needs to be produced, over and above what was required before the battery. And then there is the energy required to make the battery.

      This is not a solution for using less energy. It requires a lot more energy to run systems like this, even before you get into the energy costs of building these systems, which do not exist and would have to be produced, using….more energy.

  4. DFWCom

    I don’t begrudge financial analysis but it has its limits.

    1) time is of the essence – it takes 15 years or more to build out infrastructure like energy. We may not have that long.
    2) the whole point of a fiat currency is to pay for public purpose. There is no financial cost. It is simply the sovereign government marshaling resources when private capital is insufficient or unwilling to bear risk at an affordable price.
    3) the essential point, though, is that this is not a profit optimizing problem, it is an existential threat. Profit calculations are irrelevant.

    A national mobilization is needed that includes all the components listed above. Some will pan out, others will not. The future is never predictable. Double down on what works, course correct and find the path. A national effort like this is the purpose of a national government. Or what’s a national government for?

    1. i just don't like the gravy

      I regret to inform you that national governments are for enriching a small oligarchic sect.

  5. Piotr Berman

    Spending requirements are only a part of the picture. Currently, one of the largest nuclear projects is El Dabaa in Egypt with projected net generating capacity 4.4 GW and cost ca. 28 billions. The cost and time projections will probably be realized, given the track record of the main project manager, RosAtom.

    The largest (only?) American nuclear projects are two units of Vogtle nuclear power plant in Georgia, size of the project exactly half of El Dabaa. “By 2021 they were estimated to be over $28.5 billion.[14] In 2023 costs had increased to $34 billion, with work still to be completed on Vogtle 4.[15]” (Wikipedia). There are many structural problems here. Decades of hostility to nuclear power led to the atrophy of capacity to build and expertise. Engineers who know what can go wrong and how to avoid it retired, the main company with expertise, Westinghouse, got bankrupt, there are some newbies on the scene talking big game about small modular reactors, but they lack experience of reliability (essential in nuclear business, wind turbines falling down or shedding wings is one thing…) and now face big interest rates that double initial cost projections (nuclear has low cost if real interest rates are close to zero, as they are used for 40-60 years with low current costs).

    In another world, that would not be a problem, lacking domestic expertise USA could use foreign, like Egypt, Turkey or Finland. People who are building Vogtle units surely got expertise in what may go wrong and future projects can be better. Who could be the foreign expert? China, Russia and (perhaps) Korea. To replace current nuclear capacity and expand it in this world, many basic questions beg for answers.

    1. Bsn

      What you say here is true: “many basic questions beg for answers.” The biggest question to answer is regarding nuclear (including Vogtle units), what does one do with the waste? This sentence is interesting as well: “Engineers who know what can go wrong and how to avoid it retired”. Who were these engineers who knew how to avoid problems (such as the waste problem) and did they at least leave some notes to clue us in on their answers? Hmmm, perhaps they didn’t know what to do with waste or other problems.

      1. Piotr Berman

        Waste is a political problem. Technology how to package waste exists, potential sites exist, even Yukka mountain is OK. Perhaps the money wasted for fines related to Yukka Mt. should be re-directed for building a nuclear power plant in Nevada for reliable supplies of Las Vegas and Phoenix, Nevada complained that a state with a nuclear power plant should host such facility.

        1. Alex Cox

          Yucca Mountain is not OK. It has been rejected by the state of Nevada, and there is no remotely sane plan for long term storage there. The ‘official’ plan was to dump the waste under the mountain, and, in future decades, to send fleets of robots down to ‘make it safe’.

    2. bob

      ” talking big game about small modular reactors”

      Can anyone point to ONE of these? Just ONE that exists, in reality. It has been built and is producing electricity.

      At this point it’s as real as “carbon capture”, from what I can tell. DOES NOT EXIST.

      1. Piotr Berman

        RosAtom has one floating near the city of Pevek (a port on Arctic Ocean) and builds one in Sasha Republic on land, also in Arctic and also for a new mining complex. Reactors use a design originally for nuclear icebreakers. Pevek nuclear power plant works since 2020. There is also 3rd similar project.

        Basically, you can use designs for submarines or icebreakers, or try something larger. That said, according to RosAtom, it makes sense only in remote areas, thermal efficiency is worse that for reactors of 400 MW or larger, and it is not cheaper per MW. Actually, marine applications should have large potentials because there are many humungous cargo ships, and there could be economies of scale, replicating designs with proven record and installing on w_e_l_l maintained cargo ships (which is a large (?) minority).

        1. bob

          I think Fukushima was based on a ship design.

          Everyone speaks like this is just a matter of time until the new designs catch up and small scale modular will take over. They’re not “new” designs and they’re not about to take over.

          It’s used in the same way Carbon Capture is. An answer that people making lots of money right now bring up to stop people from looking at other options.

          Personally, I’m a fan of nuclear. I was saying 15 years ago, for the “shovel ready projects” that they should be planning to build 200 full sized nukes. Maybe 100 would have moved from design to construction and a few would be built by now.

          But, if they can’t do small scale nuclear, even on a small scale, it’s probably not going to work and it’ll just serve as another distraction. It’s amazing how little progress has been made on this front.

      2. c_heale

        Can the high level waste be dumped in those bunkers where the politicians go to hide when there is a nuclear war. You can bet these are in geologically stable rock formations.

    3. Felix_47

      I assume RosAtom is Russian. I thought Newsome was tearing down atom plants that still had a lot of life in them.

    1. Bsn

      I wonder how much was spent on campaign adds by either the for or the against proponents? Actually, I don’t wonder.

  6. Pym of Nantucket

    So impossible in the US. I guess it’s a bit of a wait-and-see on whether the West can escape the strangle hold that financial speculation has on literally everything. Wait and see how much decline is needed to open people’s eyes. Without being needlessly pessimistic, I’d say the pain of denial will have to also increase about 250%.

  7. tegnost

    Data centers are the invisible elephant in the room, and you know the owners of the data centers expect uncle sugar to pony up. I did some searching for a decent citation but all I found was data center pr.(data centers in 2014 used only 1% of global generation,, which hides the US consumption where most? of the data centers are), maybe someone else can be more successful than I.

  8. ISL

    A useful assessment for its (we are screwed) conclusion, in particular since it seems to make a vast amount of sunny suppositions based on current conditions, not the conditions for which the world it posits – as in raw material prices rising due to the demand to upgrade the grid, and I am suspect that it neglects the storage problem – a kwatt of added to the grid is not the same as a kwatt of nuclear (as if it is an option in the next 20 years – how long does it take a US nuclear plant to be built these days? 30 yrs? 40 yrs? infinite yrs?).

  9. NYMutza

    Because regulated utilities are guaranteed a certain return on capital the cost of electricity to the consumer will soar to stratospheric levels. Who will pay for this? It sounds great to de-carbon everything, but when the cost becomes unbearable it simply can’t be done. Unless government will provide substantial subsidies. In the neoliberal capitalist world we live in that isn’t going to happen.

  10. JonnyJames

    As the article outlines, things don’t look good. If PG&E (Pacific Gas and Electric, aka: Pacific Graft and Extortion Co.) can be used as an example, the opposite is true of what is needed. Instead of maintaining and improving the infrastructure over the last decades, they have asset-stripped and allowed it to become hazardous. The San Bruno, CA pipeline explosion destroyed a whole neighborhood. Although plenty of evidence of willful criminal negligence exists, PG&E was not held to account. The numerous fires caused by their antiquated equipment burned down thousands of homes, killed and displaced many people. Again, no one was held to account. One could say, the crime was rewarded, rather than punished.

    Now, PG&E charges the highest rates in the US, and it’s going up another 15% after the 1st, while asking for an ADDITIONAL rate hike later in the year. The so-called PUC is just a rubber-stamp for monopoly price-gouging. PG&E awards their senior management 10s of millions, spends 10s of millions on advertising and PR etc. The CEO a couple of years ago was paid 53 million for one year. High Crime pays, the law and taxes are only for the “little people”. Of course, both D and R politricksters are fully bribed, so it’s gonna get worse. The incentive for PG&E is to continue to extort the public, and asset-strip the system. This is but one glaring example of the institutional corruption and rot.

    1. Calexpat

      Please note: All five See Puke, California Public Utilities Commisioners who approved those rate skyrockets, are Newsom appointees. Like the non-profits that recieve contracts from politicans and whose executives then donate to those politicians, Newsom gets plenty of baksheesh via his wife’s “educational film company” getting PG&E money. Furthermore, California schools then buy her brainwashing videos as part of the new currriculum mandated by Newsom.

      “The Sacramento Bee reported the nonprofit, called The Representation Project, has received at least $800,000 in political donations from corporations that lobby state government in recent years. Jennifer Siebel Newsom founded the nonprofit in 2011 and drew a $150,000 salary in 2019, according to the couple’s most recently released tax returns. She left her job as chief executive officer that year, when Newsom became governor, and now serves as chief creative officer.

      The nonprofit focuses on gender equity and women’s representation in media, themes explored in three documentary films by Siebel Newsom. She runs a production company called Girls Club Entertainment, which paid her $50,000 in 2019. Her third film, “The Great American Lie,” came out that year and explores economic inequality and gender.”

      1. JonnyJames

        Yes, it’s institutional corruption. But to think it’s just the Gaviner, or a few bad apples is not accurate. This has gone on for decades, with R governors as well as D. Let’s keep it real. There is no way to “vote” against the interests of the oligarchy. We just have to pay the extortion, or leave the country. No matter who is “elected”, conditions will continue to worsen, as they have been for DECADES

  11. Rip Van Winkle

    Utilities are not required by law to remove PCB-containing transformers and capacitors until they reach the end of their useful life. So, for example, when they blow, that will be the end of their useful life. If the end of their useful life comes sooner than anticipated, then what will be the increased cost? Where is the new equipment going to come from?

  12. Karl

    Using nukes to balance intermittent renewables is a difficult proposition given the technical limitations of light water reactors (LWRs) and the swings in power output of renewables.

    LWR ramp rates have to be kept to around 20% per hour, and once the ramp is completed, it needs to hold steady for 3-6 hours for the reactor core to safely equilibrate. Also, the safe size of the ramp (relative to nameplate output) declines over the fuel re-load cycle, from 80% of nameplate early in the cycle (lots of flexiblility) to 20% at the end of the fuel cycle. There are other technical limitations.

    One may ask: doesn’t an LWR nuclear powered ship in the U.S. Navy require variable output for its propulsian? The answer is yes. But it’s a completely different design. It is operated typically at 15% rated power maximum and meets other stringent specifications that make them much different from commercial power plants. Needless to say, Naval nukes would be prohibitively expensive for commercial power production.

    Economic and technical limitations have tended to dictate that commercial LWRs be run essentially at constant output most of the time. The more the output fluctuates, the more thermal stresses are placed on the fuel assemblies, and over time, increases the risk of fuel cladding failure. All of this means still higher costs and risks of running nuclear plants.

    Other advanced fuel cycles (e.g. molten salt thorium) are inherently better at varying output with load. We should stop the magical thinking around LWRs and pursue alternatives, as Alvin Weinberg urged in the early 70’s. He was Oak Ridge’s main reactor designer during the Manhattan Project, Rickhover’s mentor, and fired as first Director of the new ERDA by Nixon because he opposed the magical thinking around LWRs back in the 70’s. Read The First Nuclear Era for that story.

    1. Penny

      THE question: Who will write the liability insurance policies?

      No insurance company in the world will insure a nuke for liability. Thus it falls on the taxpayers to cover almost unlimited damages should one blow upwind of a city.

      The Price Anderson Act is a joke with a tiny deductable. You and I cover the rest of the costs, or, their profits, our cancer, is our problem.

      1. Karl

        Even with this hidden subsidy (cheap insurance) nukes are still uneconomic in the USA. Nukes are more economic where design, construction and operation are rationalized as a matter of necessity, as in Korea, Finland, China and France. The high costs of big infrastructure projects in the U.S., compared to other countries, is pretty scandalous. High interest rates make such projects even more uneconomic. By contrast, solar and wind benefit from standardized designs, economies of scale, modularization, and mass production efficiencies, and as a result, their costs keep coming down.

        1. Felix"47

          Aren’t the Chinese building 22 large reactors per year and somewhere I read that next year it will be 40? They see the coal plants they are building as transitional to carry them over until they go fully nuclear. Building so many plants builds expertise and safety. We probably could contract with the Chinese and have them come over here with large construction teams and build nuclear plants. It would be sort of the reverse of what we did when Kaiser Steel went bankrupt in Fontana Ca. The Chinese wanted the basic oxygen furnace that was essentially new and they disassembled the whole plant and shipped it in their own ships and reassembled it in China. They used Chinese labor. Right before the whole thing vanished we were able to get the molten iron scene for the Terminator 2 done. I remember the project manager telling me (and this was in 1993) “Americans need their rest.”

Comments are closed.