Energy Musings - December 22, 2025

Merry Christmas and Happy Hanukkah. We highlight several studies about the overestimation of the cost of fossil fuel energy and underestimation of wind's cost. Bad policies follow.

Blowing Wind Is Expensive Working Part-Time

A new study from European researchers confirms prior examinations documenting the diminishing efficiency of offshore wind farms as more turbines were added, meaning that the output from the last turbine coming online is lower than that of the first. The study by researchers at the Delft University of Technology examined the output records from 72 European wind farms. It was published in the journal Cell Reports Sustainability and has been called “landmark” by other researchers. The significance is that the current government policy targets for offshore wind often overestimate actual energy production by up to 50%.

The study involved creating a model to estimate the energy that wind farms should generate and then comparing that theoretical output with the actual production. As the researchers stated:

“Energy extraction depends on the vertical transfer of momentum from these high-altitude winds down to the turbines, which sets a physical ceiling on how much energy can [be] harvested. A closed-form analytical model, validated against more than 420 years of operational data from 72 wind farms, defines this upper limit through a dimensionless Wind Farm Wind Factor, which condenses the key design and operational conditions of the wind farm, turbine, and site.”

Using the theoretical physical upper limit to offshore wind farm production, the researchers then examined the policies of the UK, France, Germany, the US, the Netherlands, and Belgium. The comparisons of the national policy targets showed that the expected energy production from these offshore wind farms under these various government policies was up to 50% higher than can realistically be achieved.

Why does this happen? The arrangement of wind turbines creates turbulence that impacts the output of neighboring turbines. When the wind flows through a wind farm, it is disrupted (wind wakes), making it less impactful for neighboring turbines. This phenomenon is often referred to as “stealing” the wind energy.

Why is this conclusion so important? As the researchers wrote, “Such overestimation not only hides true energy costs but also underestimates power variability, integration, and curtailment risks, and it distorts policy pathways. When projections exceed physical limits by such margins, the resulting electricity shortfall can destabilize decarbonization strategies and reach deep into society and the economy.”

In other words, the overestimation of the output of these offshore wind farms, as they are critical components of government energy policies, means that they may be inflicting undo financial harm on ratepayers and society. Moreover, the harm may endure for years and even decades, given the lifespans of these wind turbines. This point was made by the researchers when they wrote:

“Because of the long lead times to develop projects and new electricity systems, including storage, and the long operational life of these assets, errors in projections will affect multiple generations. The heavy demands on society (e.g., qualified labor), the economy, and the environment mean that corrective paths may become costly or unfeasible for a country or region.”

That is a telling message. Not only are governments overestimating the benefits of offshore wind, but they are also constructing a grid based on that input, which will continue to inflict financial harm for years and likely prevent the development of an improved alternative. Governments are indeed failing to follow the Hippocratic Oath: Do no harm.

The study includes numerous charts comparing theoretical output measures with actual data, validating the model’s results. We are not showing any of those charts; instead, we have selected ones that demonstrate the problems of offshore wind output and its impact on system costs and the policies of various governments.

The first chart shows the typical output from wind farms compared to the regular electricity demand pattern. As shown, there are times when wind farms pump out substantially more electricity than the grid needs. As the chart demonstrates, the wind farm produces 38% of its total output above the market demand, meaning it must be curtailed, stored, or demand increased. These options have a cost borne by ratepayers.

Wind power is plentiful at times but absent at others.

A second chart of the output from a wind farm shows the fraction of hours spent at various power levels as the Wind Farm Wind Factor (WFWF) increases. The WFWF is a measure of the output impacted by the wind farm layout and wake effects.

Wind energy is scarce for many hours.

The chart above shows that a larger share of total energy is produced in fewer high-output hours, leaving more hours at minimal output. This output pattern increases the risk of curtailment and also the need for more power. It suggests that more turbines must be installed to meet the same energy target, increasing both the wind farm’s total capacity and the risk of oversupply. Both factors add to the cost of wind power.

Most government offshore wind cost estimates are too low.

The final chart shows a comparison of the theoretical levelized cost of electricity (LCOE) assumptions embedded in government offshore wind policies. While we find many problems with the LCOE metric due to assumptions employed in its calculation, the comparison is telling.

The blue dashed line in the chart represents the LCOE trend assuming an inverse relationship with the capacity factor of wind farms based on a reference price of 80€/MWh ($94/MWh) at a 50% capacity factor, assuming that the capacity factor is 90% of its theoretical capacity factor limit. The 90% measure reflects the expected loss of electricity between the turbine’s generation and its use by the grid. The circular markers indicate the LCOE estimates based on the capacity factor published as goals of national policies and projects. In contrast, the star-shaped markers represent the expected LCOE values based on the model’s capacity factors.

Of the nine country policies or projects examined, two exceeded the theoretical LCOE, one matched it, and six were well below it. These results indicate that overestimating a wind farm’s capacity factor leads to an overestimation of its electricity output, and thus an underestimation of the cost of electricity.

The study’s authors noted that because the calculated LCOE is inversely proportional to the capacity factor, even modest errors in the capacity factor can significantly affect project feasibility. They further pointed out that this error factor may be particularly relevant for projects with long intervals between initial site studies by government agencies, formal project tenders, and the construction of actual wind farms.

An example of these challenges was presented in the study’s examination of two U.S. offshore wind farms – Atlantic Shores South and New York Bight. The authors wrote the following about the evolution of these projects and the potential impact on the final cost of the electricity the projects would produce.

“The US offshore wind case studies (US 1: Atlantic Shores South, US 2: New York Bight) illustrate a distinct challenge in translating policy assumptions into project realities. Both projects initially adopted an assumed capacity factor of 40%, as calculated by the Bureau of Ocean Energy Management (BOEM) during the leasing phase, based on a conservative capacity density of 3 MW/km2 (see Bureau of Ocean Energy Management,28 Lease Area Descriptions). At this lower density, our analytical model predicts capacity factors of 43.3% for US 1 and 41.6% for US 2—slightly above BOEM’s original estimate and confirming its initial reasonableness. Notably, BOEM had already acknowledged during the leasing process that actual projects would likely be developed at higher densities. Indeed, the final designs for US 1 and US 2 reached 6.9 and 6.8 MW/km2, respectively—more than double the original planning assumption. Nevertheless, the policy and publicly communicated capacity factor have remained fixed at 40%, with no adjustment for the increased aerodynamic losses associated with such dense layouts. This disconnect highlights a failure to update initial leasing-phase assumptions in response to project-level changes, resulting in unrealistic and overly optimistic expectations for the ultimate energy yield of these higher-density projects.”

This is a significant problem for offshore wind because, contrary to popular belief, its costs have not been declining in recent years. The study noted that “After a period of steady decline until around 2018-2019, offshore wind costs have reportedly seen a 30%-60% increase over the past 4 years.” That observation was footnoted to independent studies by energy consultant DNV and management expert McKinsey & Company. The authors of the report further noted that published LCOE estimates for offshore projects can differ widely, even within the same regions and time frames. The differences likely reflect different assumptions in the LCOE calculation, a point that is very germane to the rest of this article.

First, we need to show the key conclusion of the offshore wind report. The authors wrote:

“Leveraging the validated model, we evaluated offshore wind policy targets from several countries, including the UK, France, Germany, the US, the Netherlands, and Belgium. Our analysis identified substantial and systematic discrepancies between national policy projections and the realistic aerodynamic limits. Notably, the Dutch offshore wind program exhibited the most significant overestimation, predicting capacity factors nearly 50% above feasible limits. Similar, although less extreme, overestimations were observed for France (up to 22%), Belgium (24%), and the US (13%–20%). Such widespread discrepancies underscore a global risk of inflated expectations, potentially leading to misguided investments and infrastructure planning and failure of energy supply.”

The significance of the offshore wind study’s conclusion is its role in studies of the cost of renewable energy. One study was prepared by WindEurope in conjunction with Hitachi. The other was an analysis of the U.K. National Energy Operations Service (NEOS) report on the cost of the government’s energy plan for the next decade, undertaken by David Turver of Eigen Values.

Turver found, by studying the assumptions used in the annex to the National Energy System Operator (NESO) study, that the cost of renewables is understated. In contrast, the cost of energy from fossil fuels is overstated. A key assumption in the two reports is the theoretical performance of wind energy, both on- and offshore.

Turver noted that the NESO report assumed that offshore wind’s capacity cost in 2025 was £2.6 billion ($3.5 billion) per gigawatt (GW). The cost is projected to decline to just £2.0 billion ($2.7 billion)/GW in 2035. The assumption underlying this projection is that the history of declining offshore wind costs will continue over the next five years, whereas recent years, as noted above, have seen the opposite. Furthermore, there are few signs that offshore wind’s increasing costs are set to reverse.

Turver pointed out that the Planning Inspectorate forecasted that the 2.4 GW Hornsea 4 North Sea wind project would cost £5-8 billion ($6.7-10.7 billion) in 2021. Developer Ørsted recently cancelled Hornsea 4 because the costs have likely ballooned beyond even the high cost estimate. That would translate to a cost of £3,333 ($4,466) per kilowatt (kW). That cost is well above NESO’s estimates.

Turver’s analysis noted that NESO assumes “asset lives of 35, 25, 30, and 28 years for solar, onshore wind, fixed offshore wind, and floating offshore wind, respectively. Renewables Obligation subsidies last 20 years, current Contracts for Difference 15 years, and new contracts on offer for AR7 have been extended to 20 years.” The difference between governmental subsidy lives and the assumed asset lives points up two issues. First, the history of renewable energy projects is that once subsidies end, the projects are either shut down and removed or repowered, adding more costs that are then financed via new subsidies. Therefore, the assumed asset lives for offshore wind bears little relationship to the subsidy lives.

The second issue is that long asset lives allow the capital cost to be spread over a much longer period, thereby lowering the LCOE estimate. Turver noted that NESO assumes a 25-year life for gas-fired power plants, even though the industry is extending their lives to 30 and even 40 years. The shorter lifespan of gas-fired plants compared to the 30- and 28-year lives assumed for offshore wind means the LCOE of the former is higher, while the costs of offshore wind are lower.

Cost of capital estimates are too low.

Another key assumption in the economics of offshore wind is the project’s cost of capital. NESO assumes that offshore wind’s cost of capital in 2025 is 8.3%, falling to 6.3% in 2035. As Turver notes, Ørsted’s cost of capital is about 6.5%. They have stated publicly that they aim for returns on offshore wind projects that are 1.5% to 3.0% above their cost of capital. That makes their target discount rate for projects 8.0% to 9.5%, well above the NESO assumptions for 2035. The higher-than-assumed cost of capital means that the cost of the electricity produced by offshore wind farms will be higher than NESO predicts.

NESO is very optimistic about the capacity factor for offshore wind. For fixed-bottom offshore wind, NESO assumes a capacity factor of 51% in 2025. It projects the capacity factor will fall to 46.8% in 2035. Turver notes that the offshore wind fleet participating in the Contract for Difference (CfD) subsidy program achieved a 46.8% capacity factor for only one year, 2020. All other years have seen lower capacity factors, with 2024 falling to 38.8%. He further noted that NESO assumes a 30.3% to 31.3% capacity factor for onshore wind, while the CfD onshore wind fleet averaged a capacity factor of around 25% for 2021-2024.

Turver additionally points to NESO’s price estimates for offshore wind for 2025 and 2035. For fixed-bottom offshore wind, the price begins at £70.10 ($93.93) per megawatt-hour (MWh) in 2025, falling to £53.20 ($71.29)/MWh in 2035. He notes that Hornsea 4 won a contract last year at £85 ($114)/MWh, which was cancelled as being uneconomic. The latest government subsidy program for offshore wind (AR7) offered £118 ($158)/MWh in September 2025 prices for 20-year contracts, 121% above NESO’s estimate for 2025.

Offshore wind costs are projected to decline, but will they?

While David Turver’s examination of the assumptions in the NESO report shows that the cost of fossil-fuel energy was overestimated and the cost of offshore wind energy was understated, government officials, regulators, and the media tell people the opposite. Seldom, if ever, do these officials and others examine the assumptions used in the analysis and how they shape the conclusions, let alone whether the conclusions are reasonable.

That is why we were interested in the new report from WindEurope and Hitachi showing that renewables are the least costly energy option for Europe. Once again, the study relies on LCOE estimates to drive the results. The study used the LCOE estimates prepared in 2024 by the Fraunhofer Institute for Solar Energy Systems ISE. In examining the LCOE report, we find two key assumptions – asset life and weighted average cost of capital – that understate the cost of renewable energy, making it more competitive with fossil-fuel-generated energy.

An article by WindTech International summarized the key points of the study, which included four renewable energy scenarios for reaching net-zero emissions by 2050, and one scenario that fails because it relies on fossil fuels. The scenarios employ various ratios of renewable and fossil fuel energy. The key conclusion is that “a system built around high shares of wind and solar remains the lowest-cost option, even when accounting for grids, storage, and back-up generation.”

The high-renewables scenario shows €331 billion ($387 billion) in savings by 2035 compared with the slow transition scenario. Such a renewables scenario is also credited with offering strong energy security because electricity production consistently exceeds demand, thereby reducing dependency on imported energy. In the study, imported energy falls from 71% in 2030 to 22% by 2050. In contrast, the slow-transition scenario reduces the dependency on imported electricity from 78% in 2030 to just 54% in 2050. The increased energy security and reduced dependence on imported power mean Europe becomes less vulnerable to external shocks. Lastly, the European wind sector benefits from local employment growing from 440,000 today to 600,000 by 2030.

Once again, we find that European studies designed to promote the use of renewable energy, especially offshore wind power, are flawed because they rely on assumptions that understate the cost of renewables and overstate the cost of fossil fuel energy. The point of this analysis is that people should always be skeptical of studies on the economics of energy until all the critical assumptions underlying them are understood and validated as reasonable.