Seven Things People Said About the Texas Blackout of 2021, Tested

Within minutes of the lights going out in Texas in February 2021, both teams had their stories ready. One said the wind turbines froze. The other said the free market failed. Each story fit on a bumper sticker, each flattered its tellers, and each was missing a large piece of the evidence. Reality, as usual, had not been consulted.

The grid operator, the Electric Reliability Council of Texas, known as ERCOT, came within minutes of a collapse that could have left the state without power for weeks. That is a serious matter, and it deserves better than a partisan food fight. So here is the method: seven claims people made, each stated at its strongest, each tested against the record, with a verdict. The best single source is the joint report of the Federal Energy Regulatory Commission and the North American Electric Reliability Corporation, together with the regional entities, issued in November 2021, along with the earlier federal report on the 2011 freeze. Readers should check the figures against those reports themselves.

What happened, in brief

Winter Storm Uri brought extreme cold across Texas in mid-February 2021, in the days around February 13 to 17. Demand for electricity surged as people heated their homes, and generating capacity fell away at the same time as coal, nuclear, wind and gas plants lost output. Rotating outages began in the early hours of February 15 and, for many, did not rotate. More than four million customers lost power at the peak, some for days, in freezing weather. The state's official count of deaths linked to the storm was 246, and other estimates have been higher.

On February 15, the grid came close to a full collapse, and ERCOT says it was 4 minutes and 37 seconds from complete failure when the partial shutdowns were put in place. A full collapse is called a black start, and restarting a grid from zero is a slow and uncertain job that can take weeks. The operators' decision to cut load in time was what prevented it.

Now the claims.

Claim 1: "The wind turbines froze, and that caused the blackout"

The claim at its strongest. Wind turbines in Texas did freeze, and some wind generation was lost in the cold. Early in the crisis, many officials and commentators pointed to frozen turbines as the cause of the outage.

The record. Some wind generation was lost to icing and cold. But natural gas output dropped more than five times as much as wind output, and coal and nuclear output fell as well, so the loss of wind was a smaller part of the shortfall than the headlines suggested. The great bulk of the lost generation was from thermal plants, not from wind.

Verdict: mostly false as an explanation. Frozen turbines were real and they were not the main cause. Blaming wind alone gets the numbers wrong.

Claim 2: "It was all the fault of natural gas, and wind and solar did fine"

The claim at its strongest. The mirror-image story says that gas and other conventional plants failed and wind and solar held up, so the lesson is to build more renewables.

The record. Natural gas generation fell by far the most, and that was the largest single reason for the shortfall. But the claim that renewables "did fine" is also wrong. Wind did lose output, and the system was short of dependable capacity of every kind. A grid that is short of firm power in a deep freeze cannot make up the difference with sources that produce when the weather allows. What the event exposed is that the plants most needed in the crisis were precisely those that could be called on when the weather was worst, and that too many of them could not run.

Verdict: half true and half false. Gas was the biggest failure, and renewables were no answer to the specific problem. Both partisan stories get one half right.

Claim 3: "Natural gas supply froze, and that is how the electric and gas systems caused each other to fail"

The claim at its strongest. The gas system and the electric system depend on each other. Gas plants need gas, and gas wells, gathering lines and processing plants need electricity to run their pumps and compressors.

The record. This claim is correct, and it is the least understood part of the story. Natural gas supply fell sharply during the freeze, and some gas compressor stations lost power. The loss of that infrastructure reduced the fuel available for power plants, which then lost output, which forced more cutting. It was a loop, and each half made the other worse. Many generators, especially gas facilities, had also not been weatherized.

Verdict: true, and it is the key to the whole event. Anyone who talks about the grid without talking about the gas system is leaving out the engine room.

Claim 4: "Most of the outages came from gas, so wind and solar were innocent"

The claim at its strongest. Counted by fuel type, gas accounted for the largest share of the lost generation.

The record. It did, and wind also lost output. The right way to read the numbers is by what each source was expected to deliver. Wind and solar produce when the weather allows; they are not, by design, dependable on demand, and the system was not counting on them for much in the coldest hours. Gas, coal and nuclear are supposed to be dependable, and they were the ones counted on. So the key statistic is not the share of units that failed, but how much dependable capacity was missing at the moment it was needed. By that measure the thermal fleet's failures were the decisive ones.

Verdict: the facts are right and the conclusion is a dodge. Counting units by fuel type tells you who failed. Counting dependable megawatts tells you why the lights went out.

Claim 5: "Texas's isolated grid caused the blackout"

The claim at its strongest. ERCOT is largely separate from the two big interconnections that cover the rest of the country. If Texas had been plugged in to its neighbors, it could have imported power.

The record. ERCOT serves about 90 percent of the state's electricity load and is connected to the rest of the country only by a small number of direct-current ties with limited capacity, a little over a thousand megawatts. That isolation has a reason: it keeps the grid outside the jurisdiction of the federal government's rate regulator for most purposes, because power that does not cross state lines is regulated by the state. That is a choice Texas made decades ago, and a federalist might admire it.

It is true that isolation limited how much Texas could import. But the neighbors were in the same storm and had their own shortages. The Southwest Power Pool, for one, ordered rolling blackouts across 14 states on February 15 and 16. More interconnection could have helped at the margin, and it would not have solved a shortage of tens of thousands of megawatts across a region freezing at the same time.

Verdict: a partial truth. Interconnection is a reasonable subject for policy, and it was not the cause. Anyone who says it would have prevented the crisis is guessing.

Claim 6: "Nobody could have seen this coming"

The claim at its strongest. The storm was extreme, and the argument goes that no reasonable plan could have anticipated it.

The record. This is where the record is most damaging, and it hurts people who ought to have known better. Texas had been through freezes before. A severe cold event in February 2011 caused rolling blackouts across much of the state, and federal regulators recommended winterizing equipment. Similar recommendations after a cold wave in December 1989 had also been ignored. The 2011 recommendations were not implemented, largely because of cost, and many critical generators, especially gas facilities, still lacked winterization in 2021.

Cold weather itself was a known risk. The scale of the 2021 storm was unusual, and some of what happened was beyond what any plan would have covered, but the need to winterize equipment was not a surprise to anyone who had read the 2011 recommendations.

Verdict: false, and the opposite is closer to the truth. The warning had been issued, in writing, ten years earlier.

Claim 7: "Deregulation and the free market caused it" versus "Regulation and green mandates caused it"

The claim at its strongest. Each side has a version. The left says Texas's deregulated market failed to provide reliable power, because nobody was paid to be ready for emergencies. The right says government mandates and subsidies for wind distorted the market and crowded out dependable plants.

The record. The honest answer is more interesting than either. Texas has what is called an energy-only market: generators are paid for the electricity they deliver, with prices that can rise to a cap, which in 2021 was $9,000 per megawatt-hour, and there is no separate payment for simply being available. In theory the high prices in a shortage pay for the investment that keeps plants ready. In practice, the market gave plant owners little reason to spend on cold-weather protection that might be used once in a decade, and the federal subsidy for wind, which paid wind operators for each unit produced, is argued to have distorted decisions on the other side, a pattern traced in The Wind Tax Credit Was Temporary in 1992 and Kept Coming Back for Decades. Both a design flaw in the market and a distortion from policy were at work, and neither side's story is the whole truth.

The market is also not free in a simple sense. The system had grown with state oversight, rules about what counts as a critical facility, and decisions about winterization standards that were largely left to the operators. It is a hybrid.

Verdict: both are partly right and both are overselling.

The strongest case for the other side

A defender of the market approach would say that the system performed as designed in most winters, that prices signaled scarcity as they should, and that adding a mandatory capacity payment would raise costs for customers every year to guard against a freak event. That is a respectable position. The same defender would add that a regulator-driven reliability standard has its own failures and can lock in expensive, politically favored plants.

Here is the reply. A market is supposed to price the product customers actually want, and what a customer in a freeze wants is for the lights to stay on. If an energy-only market does not pay for readiness, then either its prices have to do it reliably, or something else has to. The 2011 warning and the 2021 failure show that, in this case, the signal was not strong enough or not trusted enough to produce cold-weather protection. A conservative who believes in markets should want the market to be designed so that dependability is paid for, because a market that fails the customer in the worst hour is not delivering what a market is for.

The lesson: pay for what you need when you need it

The practical lesson is about dispatchable capacity, meaning power that can be turned on when needed and is not dependent on the weather. It is not a case against wind or solar, which produce cheap electricity when they produce. It is a case for being honest about what each source does. A grid needs energy over the year and it needs capacity at the worst hour. Wind and solar deliver much of the first and, in the worst hours of a winter freeze, little of the second. Gas, coal, nuclear and hydro deliver capacity, if they are protected against the conditions they will face.

So the policy questions that follow are plain, and they can be answered by people of different politics.

1. Who pays for winterization? Whether through a mandate, a payment for readiness or a penalty for failing to perform, someone has to be accountable.
2. How is gas treated in an emergency? The electric and gas systems must be planned together, and critical gas facilities must be protected against rotating cuts.
3. What price is paid for reliability, and by whom? A capacity payment costs money every year. The cost has to be weighed against the cost of a blackout, which in 2021 included lives.
4. How much do we want to rely on any one source? Diversity of firm power, including nuclear, a subject in Count Deaths per Unit of Power and Nuclear Beats Nearly Everything Else, is insurance.

For families, the point is that reliability is not free and not automatic. It is built, paid for and maintained. The same logic applies to rate cases that decide what you pay for it, covered in Your Power Bill Is Going Up, So Here Is How to Speak at the Rate Case, and to the broader story of how America stopped building quickly, in We Raised the Empire State Building in About a Year, So Why Not a Power Line. The question of who decides what the grid must withstand is a question of federalism too: the state kept control of its grid, and the state must therefore answer for it. The Texas story is a case of the principle Madison set out in Federalism Lets Fifty States Experiment, and the Constitution Planned It: when states run their own systems, they own both the credit and the blame.

The case in brief

1. Winter Storm Uri in February 2021 took ERCOT within minutes of a grid collapse, and the joint FERC and NERC report of November 2021 is the best single source for what happened.
2. Frozen wind turbines were real but were not the main cause, and natural gas losses were the largest part, but wind and solar were not the answer to the shortfall either.
3. The gas and electric systems failed together, because gas wells and plants need electricity and power plants need gas, and some gas infrastructure was cut off by the rotating outages.
4. The cold-weather failures were not a surprise: the 2011 freeze produced recommendations that were not implemented, largely because of cost, and ERCOT's isolation limited imports but did not cause the crisis.
5. The lesson is to pay for dispatchable capacity and its winter protection, so that a market built on energy alone also rewards being ready when the lights are about to go out.

What you can do: read the executive summary of the November 2021 FERC and NERC report on Winter Storm Uri, then ask your state public utility commission what standards for cold-weather readiness apply to power plants and gas facilities that serve your area.