A week ago I posted about electrification and particularly EV mandates, one strand of government industrial policy under which non-favored sectors of the economy must labor. This post examines a related industrial policy: manipulation of power generation by government policymakers in favor of renewable energy technologies, while fossil fuels are targeted for oblivion. These interventions are a reaction to an overwrought climate crisis narrative, but they present many obstacles, oversights and risks of their own. Chief among them is whether the power grid will be capable of meeting current and future demand for power while relying heavily on variable resources: wind and sunshine.

Like almost everything I write, this post is too long! Here is a guide to what follows. Scroll down to whatever sections might be of interest:

• Malinvestment: Idle capital
• Key Considerations to chew on
• False Premises: zero CO2? Low cost?
• Imposed Cost: what and how much?
• Supporting Growth: with renewables?
• Resource Constraints: they’re tight!
• Technological Advance: patience!
• The Presumed Elephant: CO2 costs
• Conclusion

Malinvestment

The intermittency of wind and solar power creates a fundamental problem of physically idle capital, which leaves the economy short of its production possibilities. To clarify, capital invested in wind and solar facilities is often idle in two critical ways. First, wind and solar assets have relatively low rates of utilization because of their variability, or intermittency. Second, neither provides “dispatchable” power: it is not “on call” in any sense during those idle periods, which are not entirely predictable. Wind and solar assets therefore contribute less value to the electric grid than dispatchable sources of power having equivalent capacity and utilization.

Is “idle capital” a reasonable characterization? Consider the shipping concerns that are now experimenting with sails on cargo ships. What is the economic value of such a ship without back-up power? Can you imagine them drifting in the equatorial calms for days on end? Even light winds would slow the transport of goods significantly. Idle capital might be bad enough, but a degree of idleness allows flexibility and risk mitigation in many applications. Idle, non–dispatchable capital, however, is unproductive capital.

Likewise, solar-powered signage can underperform or fail over the course of several dark, wintry days, even with battery backup. The signage is more reliable and valuable when it is backed-up by another power source. Again, idle, non-dispatchable capital is unproductive capital.

The pursuit of net-zero carbon emissions via wind and solar power creates idle capital, which increasingly lacks adequate backup power. That should be a priority, but it’s not. This misguided effort is funded from both private investment and public subsidies, but the former is very much contingent on the latter. That’s because the flood of subsidies is what allows private investors to profit from idle capital. Rent-enabled investments like these crowd out genuinely productive capital formation, which is not limited to power plants that might otherwise use fossil fuels.

Creating idle or unemployed physical capital is malinvestment, and it diminishes future economic growth. The surge in this activity began in earnest during the era of negative real interest rates. Today, in an era of higher rates, taxpayers can expect an even greater burden, as can ratepayers whose power providers are guaranteed returns on their regulated rate bases.

Key Consideration

The forced transition to net zero will be futile, but especially if wind and solar energy are the primary focus. Keep the following in mind:

• The demand for electricity is expected to soar, and soon! Policymakers have high hopes for EVs, and while adoption rates might fall well short of their goals, they’re doing their clumsy best to force EVs down our throats with mandates. But facilitating EV charging presents difficulties. Lionel Shriver states the obvious: “Going Electric Requires Electricity”. Reliable electricity!
• Perhaps more impressive than prospects for EVs is the expected growth in power demand from data centers required by the explosion of artificial intelligence applications across many industries. It’s happening now! This will be magnified with the advent of artificial general intelligence (AGI).
• Dispatchable power sources are needed to back-up unreliable wind and solar power to ensure service continuity. Maintaining backup power carries a huge “imposed cost” at the margin for wind and solar. At present, that would entail CO2 emissions, violating the net zero dictum.
• Perhaps worse than the cost of backup power would be the cost borne by users under the complete elimination of certain dispatchable power sources. An imposed cost then takes the form of outages. Users are placed at risk of losing power at home, at the office and factories, at stores, in transit, and at hospitals at peak hours or under potentially dangerous circumstances like frigid or hot weather.
• Historically, dispatchable power has allowed utilities to provide reliable electricity on-demand. Just flip the switch! This may become a thing of the past.
• Wind and solar power are sometimes available when they’re not needed, in which case the power goes unused because we lack effective power storage technology.
• Wind and solar power facilities operate at low rates of utilization, yet new facilities are always touted at their full nameplate capacity. Capacity factors for wind turbines averaged almost 36% in the U.S. in 2022, while solar facilities averaged only about 24%. This compared with nuclear power at almost 93%, natural gas (66%), and coal (48%). Obviously, the low capacity factors for wind and solar reflect their variable nature, rather than dispatchable responses to fluctuations in power demand.
• Low utilization and variability are underemphasized or omitted by those promoting wind and solar plant in the media and often in discussions of public policy, and no wonder! We hear a great deal about “additions to capacity”, which overstate the actual power-generating potential by factors of three to four times. Here is a typical example.
• Wind and solar power are far more heavily subsidized than fossil fuels. This is true in absolute terms and especially on the basis of actual power output, which reveals their overwhelmingly uneconomic nature. From the link above, here are Mitch Rolling and Isaac Orr on this point:
  • “In 2022, wind and solar generators received three and eighteen times more subsidies per MWh, respectively, than natural gas, coal, and nuclear generators combined. Solar is the clear leader, receiving anywhere from $50 to $80 per MWh over the last five years, whereas wind is a distant second at $8 to $10 per MWh …. Renewable energy sources like wind and solar are largely dependent on these subsidies, which have been ongoing for 30 years with no end in sight.”
• The first-order burden of subsidies falls on taxpayers. The second-order burdens manifest in an unstable grid and higher power costs. But just to be clear, subsidies are paid by governments to producers or consumers to reduce the cost of activity favored by policymakers. However, the International Monetary Fund frequently cites “subsidy” figures that include staff estimates of unaddressed externalities. These are based on highly-simplified models and subject to great uncertainty, of course, especially when dollar values are assigned to categories like “climate change”. Despite what alarmists would have us believe, the extent and consequences of climate change are not settled scientific issues, let alone the dollar cost.
• Wind and solar power are extremely land- and/or sea-intensive. For example, Casey Handmer estimates that a one-Gigawatt data center, if powered by solar panels, would need a footprint of 20,000 acres. 
• Solar installations are associated with a significant heat island effect: “We found temperatures over a PV plant were regularly 3–4 °C warmer than wildlands at night….”
• Wind and solar power both represent major hazards to wildlife both during and after construction.
  • In addition to the destruction of habitat both on- and offshore, turbine blades create noise, electromagnetism, and migration barriers. Wind farms have been associated with significant bird and bat fatalities. Collisions with moving blades are one thing, but changes to the winds and air pressure around turbines are also a danger to avian species.
  • There is a strong likelihood that offshore wind development is endangering whales and dolphins.
  • Solar farms present dangers to waterfowl. These creatures are tricked into diving toward what they believe to be bodies of water, only to crash into the panels.
• The production of wind and solar equipment requires the intensive use of scarce resources, including environmentally-sensitive materials. Extracting these materials often requires the excavation of massive amounts of rock subject to extensive processing. Mining and processing rely heavily on diesel fuel. Net zero? No.
• Wind and solar facilities often present major threats of toxicity at disposal, or even sooner. A recent hail storm in Texas literally destroyed a solar farm, and the smashed panels have prompted concerns not only about solar “sustainability”, but also that harsh chemicals may be leaking into the local environment.
• The transmission of power is costly, but that cost is magnified by the broad spatial distribution of wind and solar generating units. Transmission from offshore facilities is particularly complex. And high voltage lines run into tremendous local opposition and regulatory scrutiny.
• When wind turbines and solar panels are idle, so are the transmission facilities needed to reach them. Thus, low utilization and the variability of those units drives up the capital needed for power and power transmission.
• There is also an acute shortage of transformers, which presents a major bottleneck to grid development and stability.
• While zero carbon is the ostensible goal, zero carbon nuclear power has been neglected by our industrial planners. That neglect plays off exaggerated fears about safety. Fortunately, there is a growing realization that nuclear power may be surest way to carbon reductions while meeting growth in power demand. In fact, new data centers will go off-grid with their own modular reactors.
• At the Shriver link, he notes the smothering nature of power regulation, which obstructs the objective of providing reliable power and any hope of achieving net zero.
• The Biden administration has resisted the substitution of low CO2 emitting power sources for high CO2 emitting sources. For example, natural gas is more energy efficient in a variety of applications than other fuel sources. Yet policymakers seem determined to discourage the production and use of natural gas.

False Premises

Wind and solar energy are touted by the federal government as zero carbon and low-cost technologies, but both claims are false. Extracting the needed resources, fabricating, installing, connecting, and ultimately disposing of these facilities is high in carbon emissions.

The claim that wind and solar have a cost advantage over traditional power sources is based on misleading comparisons. First, putting claims about the cost of carbon aside, it goes without saying that the cost of replacing already operational coal or natural gas generating capacity with new wind and solar facilities is greater than doing nothing.

The hope among net zero advocates is that existing fossil fuel generating plant can be decommissioned as more renewables come on-line. Again, this thinking ignores the variable nature of renewable power. Dispatchable backup power is required to reliably meet power demand. Otherwise, fluctuating power supplies undermine the economy’s productive capacity, leading to declines in output, income, health, and well being. That is costly, but so is maintaining and adding back-up capacity. Costs of wind and solar should account for this necessity. It implies that wind and solar generating units carry a high cost at the margin.

Imposed Costs

A “grid report card” from the Mackinac Center for Public Policy notes the conceptual flaw in comparing the levelized cost (à la Lazard) of a variable resource with one capable of steady and dispatchable performance. From the report, here is the crux of the imposed-cost problem:

“… the more renewable generation facilities you build, the more it costs the system to make up for their variability, and the less value they provide to electricity markets.”

A committment to variable wind and solar power along with back-up capacity also implies that some capital will be idle regardless of wind and solar conditions. This is part of the imposed cost of wind and solar built into the accounting below. But while back-up power facilities will have idle periods, it is dispatchable and serves an insurance function, so it has value even when idle in preserving the stability of the grid. For that matter, sole reliance on dispatchable power sources requires excess capacity to serve an insurance function of a similar kind.

The Mackinac report card uses estimates of imposed cost from an Institute for Energy Research to construct the following comparison (expand the view or try clicking the image for a better view):

The figures shown in this table are somewhat dated, but the Mackinac authors use updated costs for Michigan from the Center of the American Experiment. These are shown below in terms of average costs per MWh through 2050, but the labels require some additional explanation.

The two bars on the left show costs for existing coal ($33/MWh) and gas-powered ($22) plants. The third and fourth bars are for new wind ($180) and solar ($278) installations. The fifth and sixth bars are for new nuclear reactors (a light water reactor ($74) and a small modular reactor($185)). Finally, the last two bars are for a new coal plant ($106) and a natural gas plant ($64), both with carbon capture and storage (CCS). It’s no surprise that existing coal and gas facilities are the most cost effective. Natural gas is by far the least costly of the new installations, followed by the light water reactor and coal.

The Mackinac “report card” is instructive in several ways. It provides a detailed analysis of different types of power generation across five dimensions, including reliability, cost, cleanliness, and market feasibility (the latter because some types of power (hydro, geothermal) have geographic limits. Natural gas comes out the clear winner on the report card because it is plentiful, energy dense, dispatchable, clean burning, and low-cost.

Supporting Growth

Growth in the demand for power cannot be met with variable resources without dispatchable backup or intolerable service interruptions. Unreliable power would seriously undermine the case for EVs, which is already tenuous at best. Data centers and other large users will go off-grid before they stand for it. This would represent a flat-out market rejection of renewable investments, ESGs be damned!

Casey Handmer makes some interesting projections of the power requirements of data centers supporting not just AI, but AGI, which he discusses in “How To Feed the AIs”. Here is his darkly humorous closing paragraph, predicated on meeting power demands from AGI via solar:

“It seems that AGI will create an irresistibly strong economic forcing function to pave the entire world with solar panels – including the oceans. We should probably think about how we want this to play out. At current rates of progress, we have about 20 years before paving is complete.”

Resource Constraints

Efforts to force a transition to wind and solar power will lead to more dramatic cost disadvantages than shown in the Mackinac report. By “forcing” a transition, I mean aggressive policies of mandates and subsidies favoring these renewables. These policies would effectuate a gross misallocation of resources. Many of the commodities needed to fabricate the components of wind and solar installations are already quite scarce, particularly on the domestic U.S. front. Inflating the demand for these commodities will result in shortages and escalating costs, magnifying the disadvantages of wind and solar power in real economic terms.

To put a finer point on the infeasibility of the net zero effort, Simon P. Michaux produced a comparative analysis in 2022 of the existing power mix versus a hypothetical power mix of renewable energy sources performing an equal amount of work, but at net-zero carbon emissions (the link is a PowerPoint summary). In the renewable energy scenario, he calculated the total quantities of various resources needed to achieve the objective over one generation of the “new” grid (to last 20 -30 years). He then calculated the numbers of years of mining or extraction needed to produce those quantities based on 2019 rates of production. Take a look at the results in the right-most column:

Those are sobering numbers. Granted, they are based on 2019 wind and solar technology. However, it’s clear that phasing out fossil fuels using today’s wind and solar technology would be out of the question within the lifetime of anyone currently living on the planet. Michaux seems to have a talent for understatement:

“Current thinking has seriously underestimated the scale of the task ahead.”

He also emphasizes the upward price pressure we’re likely to witness in the years ahead across a range of commodities.

Technological Breakthroughs

Michaux’s analysis assumes static technology, but there may come a time in the not-too-distant future when advances in wind and solar power and battery storage allow them to compete with hydrocarbons and nuclear power on a true economic basis. The best way to enable real energy breakthroughs is through market-driven economic growth. Energy production and growth is hampered, however, when governments strong-arm taxpayers, electricity buyers, and traditional energy producers while rewarding renewable developers with subsidies.

We know that improvements will come across a range of technologies. We’ve already seen reductions in the costs of solar panels themselves. Battery technology has a long way to go, but it has improved and might some day be capable of substantial smoothing in the delivery of renewable power. Collection of solar power in space is another possibility, as the feasibility of beaming power to earth has been demonstrated. This solution might also have advantages in terms of transmission depending on the locations and dispersion of collection points on earth, and it would certainly be less land intensive than solar power is today. Carbon capture and carbon conversion are advancing technologies, making net zero a more feasible possibility for traditional sources of power. Nuclear power is zero carbon, but like almost everything else, constructing plants is not. Nevertheless, fission reactors have made great strides in terms of safety and efficiency. Nuclear fusion development is still in its infancy, but there have been notable advances of late.

Some or all of these technologies will experience breakthroughs that could lead to a true, zero-carbon energy future. The timeline is highly uncertain, but it’s likely to be faster than anything like the estimates in Michaux’s analysis. Who knows? Perhaps AI will help lead us to the answers.

A Presumed Elephant

This post and my previous post have emphasized two glaring instances of government failure on their own terms: a headlong plunge into unreliable renewable energy, and forced electrification done prematurely and wrong. Some would protest that I left the veritable “elephant in the room”: the presumed external or spillover costs associated with CO2 emissions from burning fossil fuels. Renewables and electrification are both intended to prevent those costs.

External costs were not ignored, of course. Externalities were discussed explicitly in several different contexts such as the mining of new materials, EV tire wear, the substitution of “cleaner” fuels for others, toxicity at disposal, and the exaggerated reductions in CO2 from EVs when the “long tailpipe” problem is ignored. However, I noted explicitly that estimates of unaddressed externalities are often highly speculative and uncertain, and especially the costs of CO2 emissions. They should not be included in comparisons of subsidies.

Therefore, the costs of various power generating technologies shown above do not account for estimates of externalities. If you’re inclined, other SCC posts on the CO2 “elephant” can be found here.

Conclusion

Power demand is expected to soar given the coming explosion in AI applications, and especially if the heavily-subsidized and mandated transition to EVs comes to pass. But that growth in demand will not and cannot be met by relying on renewable energy sources. Their variability implies substantial idle capacity, higher costs, and service interruptions. Such a massive deployment of idle capital would represents an enormous waste of resources, but the sad fact is it’s been underway for some time.

In the years ahead, the net-zero objective will prove representative of a bumbling effort at industrial planning. Costs will be driven higher, including the cost inflicted by outages and environmental damage. Ratepayers, taxpayers, and innocents will share these burdens. Travis Fisher is spot on when he says the grid is becoming a “dangerous liability” thanks to wounds inflicted by subsidies, regulations, and mandates.

As Charles Glasser put it on Instapundit:

“The National Electrical Grid is teetering on collapse. The shift away from full-time available power (like fossil fuels, LNG, etc.) to so-called ‘green’ sources has deeply impacted reliability.”

“Also, as more whale-killing off-shore wind farms are planned, the Biden administration forgot to plan for the thousands of miles of transmission lines that will be needed. And in a perfect example of leftist autophagy, there is considerable opposition from enviro-groups who will tie up the construction of wind farms and transmission lines in court for decades.”

Meanwhile, better alternatives to wind and solar have been routinely discouraged. The substantial reductions in carbon emissions achieved in the U.S. over the past 15 years were caused primarily by the substitution of natural gas for coal in power generation. Much more of that is possible. The Biden Administration, however, wishes to prevent that substitution in favor of greater reliance on high-cost, unreliable renewables. And the Administration wishes to do so without adequately backing up those variable power sources with dispatchable capacity. Likewise, nuclear power has been shunted aside, despite its safety, low risk, and dispatchability. However, there are signs of progress in attitudes toward bringing more nuclear power on-line.

Industrial policy usually meets with failure, and net zero via wind and solar power will be no exception. Like forced electrification, unreliable power fails on its own terms. Net zero ain’t gonna happen any time soon, and not even by 2050. That is, it won’t happen unless net zero is faked through mechanisms like fraudulent carbon credits (and there might not be adequate faking capacity for that!). Full-scale net-zero investment in wind and solar power, battery capacity, and incremental transmission facilities will drive the cost of power upward, undermining economic growth. Finally, wind and solar are not the environmental panacea so often promised. Quite the contrary: mining of the necessary minerals, component fabrication, installation, and even operation all have negative environmental impacts. Disposal at the end of their useful lives might be even worse. And the presumed environmental gains … reduced atmospheric carbon concentrations and lower temperatures, are more scare story than science.

Postscript: here’s where climate alarmism has left us, and this is from a candidate for the U.S. Senate (she deleted the tweet after an avalanche of well-deserved ridicule):