Disappearing Wind and Solar Cost Texans $766 Million Extra in 17 Hours

The financial fallout of Winter Storm Fern and $1.27 billion in savings from building new natural gas plants

A quick note: Dear readers, we want to let you know that we will be moving toward a paid subscription model in the near future. We’ve seen that the growth rate on our Substack has slowed over the past year, and this is partly due to changes to Substack’s algorithm that favor paid content over free content.

Our goal is to raise our visibility by reaching the top 10 in our category, and because the leaderboard is mostly driven by revenue, moving to paid is pretty much the only way to move up.

For the time being, most of our articles will still be free. However, some, probably the extra juicy ones, will live behind a paywall for an undetermined amount of time. We’re still figuring things out on our end.

Thanks for being a loyal subscriber.

Disappearing Wind and Solar Cost Texans $766 Million Extra in 17 Hours

Everything’s bigger in Texas, including the power price spikes.

In the wake of Winter Storm Fern, the U.S. Department of Energy correctly noted that it was America’s thermal fleet that carried the day, as coal, natural gas, nuclear, and oil for our friends in New England provided 86 percent of the power generated during the peak of the storm, while wind and solar generated just 8 percent and 2 percent, respectively.

This is despite the fact that U.S. Energy Information Administration (EIA) data show wind and solar constituted roughly 23 of the total installed capacity on the grid in January, meaning they punched below their weight when the grid was stressed.

With egg on their faces, the American Clean Power Association (APCA), the top lobbying arm of the wind and solar industry, hastily started up the spin machine for damage control. As a result, it issued this website claiming that wind and solar saved consumers more than $2 billion during the winter weather event, including $200 million in savings in the Electric Reliability Council of Texas (ERCOT) Region

However, ACPA doesn’t disclose how it calculated those figures, and it didn’t respond to our emails requesting more information, so we did a little analysis of our own. We found that it is exceedingly unlikely that wind and solar saved Texas any money.

In fact, our analysis found that wind and solar disappeared just when they were needed most, costing Texans an additional $766 million in wholesale power costs in just 17 hours.

To understand why, it helps to know some background.

ERCOT’s Generation Mix

For many years, ERCOT has become increasingly dependent weather-dependent resources, particularly wind energy, for meeting peak demand.

As we have noted on several occasions, ERCOT hasn’t added any net thermal capacity since 2013, and the amount of thermal generation on the system has decreased since 2016 despite the fact that peak demand has grown by over 12,000 megawatts (MW), or by over 17 percent in the last ten years.

According to EIA Form 860M data, ERCOT had 171,542 MW of total installed capacity on its system as of December 2025, with 87,430 MW of thermal capacity and 83,542 MW of wind, solar, and battery storage, as shown in the graph below.

On paper, this should have been more than enough capacity to weather Winter Storm Fern without a massive spike in prices, and for some of the time it didn’t. However, around 5 pm on January 25th, ERCOT experienced a substantial drop in wind and solar generation. At 10 pm, the thermal fleet was delivering a maximum output of 66,087 MW of power, while wind delivered 6,025 MW, solar delivered zero, and battery storage consumed 319 MW for charging, which led to tight system conditions.

At 10 pm, the thermal fleet was generating 66,086 MW while wind, solar, and battery storage were delivering just 5,706 MW.

Wholesale Power Prices: Ready for Liftoff

These tight system conditions led to a massive spike in power prices in ERCOT.

Day-ahead and real-time wholesale price data from S&P Global data show prices spiked in concert with wind and solar’s disappearing act. For instance, notice in the graph below that while prices (dark green line for Day Ahead, green line for Real-Time) increased slightly with the rise in demand, they didn’t spike until electricity production from wind and solar both plummeted.

Day-ahead prices reached a high of $1,832 per MWh, and real-time prices reached $871 per MWh, resulting in massive power costs for Texas families and businesses.

ERCOT data indicate that approximately 62.5 percent of all MWhs purchased in January of 2026 were bought on the day-ahead market, and 37.25 percent on the real-time market. Using this split, we were able to create a blended hourly wholesale power cost for consumers and determine the additional costs to consumers compared to wholesale power costs before the variable energy sources pulled their vanishing act.

The graph below shows the combined hourly wind and solar capacity factor and the settled energy cost in ERCOT during Winter Storm Fern. It also shows the additional power costs, shaded in light red, incurred as a result of the drop in wind and solar generation, relative to a $20 million-per-hour power baseline cost, shown by the red line. The total additional cost of the light red areas is $766 million over 17 hours.

Our analysis used $20 million as the baseline because it was the approximate cost of power at 4 pm on January 25th, the hour just before wind and solar output dropped during the high-demand period.

The Anatomy of a Price Spike: Why Did They Rise So High?

Ultimately, the price increase was the result of high demand and inadequate supply.

Data from GridStatus shows outages for thermal units, wind, and solar during Winter Storm Fern. Wind and solar had lower initial outages, but outages spiked on January 24th as the cold front moved in.

In contrast, thermal facilities had relatively low outage rates, with availability ranging from 88 to 90 percent during the price spike event, which lasted from the evening of January 25th through the morning of January 26th, then falling to 87 percent on January 27th.

The outages reduced ERCOT’s reserve margin, which we define here as the amount of installed thermal capacity on the system, minus outages, plus intermittent generation, available to meet demand during any given hour during the storm. We defined the reserve margin this way because wind and solar can (and did) fail to show up due to a lack of wind or sunlight, even if they are not technically on outage.

The graph below shows the amount of thermal generation available to meet peak demand, the amount of power provided by wind, solar, and battery storage, and the hourly demand on ERCOT from January 23rd through the 27th.

The red line shows the reserve margin, which dips from over 100 percent on January 23rd, when demand was lower, and wind, solar, and storage were able to deliver more power to the grid, down to just 15 percent starting at 5 pm on January 25th.

Unsurprisingly, the falling reserve margin is highly correlated with rising electricity prices. In fact, in an energy-only market like ERCOT, where the grid operator does not make reliability payments to dispatchable generators to be available during periods of high stress, like winter storms, high prices are the primary mechanism for keeping reliable plants online, as they capture huge amounts of revenue during these “scarcity events.”

Wind, solar, and storage advocates will no doubt argue that wind and solar decreased wholesale prices when they were available, but the question isn’t whether it is possible for these resources to reduce wholesale costs at times (they absolutely can).

The questions are is the grid is properly valuing reliable generators and becoming over reliant on weather based resources, and could Texas achieve greater price stability and reliability at a fraction of the cost by adding more natural gas instead of wind, solar, and battery storage.

The graph below shows the wholesale electricity price in ERCOT by available reserve margin. Based on the data from Winter Storm Fern, the wholesale power price skyrocketed when reserve margins dipped below approximately 26 percent.

For the statisticians in the audience, this model has an R² of 0.91, meaning about 91 percent of the variation in ERCOT wholesale prices in this dataset is explained by changes in available reserve margin.

Because the high prices can be explained largely by the scarcity of the system, the solution is to add more dispatchable capacity that can boost the reserve margin, rather than rely on generation resources that could fall off the grid at any moment. For the last 25 years, ERCOT has been doing the latter.

We thought it would be fun to try and estimate what the hourly cost of power would be based on building 10,000 MW of new natural gas capacity in ERCOT, which would probably cost $25 billion at today’s prices, by matching the blended wholesale power prices at a given reserve margin percentage in the original dataset to the MWh costs after boosting the reserve margins with new natural gas capacity.

This analysis found that the new gas capacity could have saved Texas families and businesses $1.27 billion in power costs, as shown in the graph below by the difference between the dark green and light green shaded areas. If any of our readers have questions, comments, or scathing rebuttals of the methodology, we look forward to discussing them in the comments.

Why Has ERCOT Failed to Add New Thermal Plants?

While natural gas additions have mostly offset coal retirements, ERCOT hasn’t added any net dispatchable capacity since 2003. This is largely due to the energy-only market structure, which does not properly value the differences in reliability between generators like gas and coal plants and intermittent resources like wind and solar generators.

Furthermore, Texas does not require utilities to keep enough backup capacity online to make sure there are always enough reliable power plants available to meet electricity demand when the wind isn't blowing or the sun isn't shining. In fact, the market structure in Texas specifically incentivizes reliable generators to exit the market.

As we wrote in November of 2023:

This isn’t a bug of the ERCOT market; it is a feature. In a 2018 article in PVMagazine detailing a report by the WindSolar Alliance of Texas, the article author bragged about the parasitic effect [wind and solar] electricity generators have on the revenues of more reliable power plants:

This is just the beginning. Texas continues to add more wind every year, and ERCOT has estimated that the state could put online 13 GW of solar by 2030. This will ultimately mean more hours where coal and gas plants are not operating, and more retirements of conventional generation. [emphasis added]

In the case of ERCOT, the parasitic impact of wind, solar, and now battery storage has not led to retirements, but it has dissuaded new thermal generators from entering the market at a time when power demand is soaring due to population growth, industrial growth, and data centers.

[emphasis added]

Something’s gotta give.

Conclusion

The American Clean Power Association is full of hot air.

Rather than saving Texans $200 million during Winter Storm Fern, these energy sources skipped town when the chips were down, increasing wholesale power prices by $766 million compared to prices that occurred prior to their exit from the scene.

In contrast, if ERCOT had an additional 10,000 MW of gas on its system, Texans could have saved $1.27 billion in avoided scarcity prices.

Better luck next time, ACPA.