When all is said and done, the one big takeaway from Texas’ week of freezing agony amid a power blackout is that recovery will be time-consuming and horribly expensive — ironically so since the culprit was a system designed to be price competitive. There are some opportunities to accelerate creation of true resilience, including microgrid concepts capable of serving small islands of load when the greater grid fails or is stressed, and encouragement of local storage with emerging battery technologies. The burdens, meanwhile, are certain to fall on the poorest and most vulnerable, particularly communities of color who often lift in neighborhoods with the oldest infrastructure to begin with.
But first, let’s back up to Valentines’ Day 2021, which will not be remembered in Texas for hearts and chocolates.
In short, a near catastrophic meltdown. As temperatures started to freeze on the evening of Feb. 14, consumer demand went up as heating elements came online. This caused a rather unusual winter load peaking situation since the Electric Reliability Council of Texas, the grid manager suddenly well known by its acronym ERCOT, is used to traditionally mild winters. ERCOT, which does not generate power itself but acts as manager of the balancing act between scores of Texas utilities, started losing natural gas-fueled generation late into the evening that day as winter conditions impacted fuel supply and operations.
Then, at around 2 a.m. CST the morning of Feb. 15, a “grid event” took place — essentially a collapse beneath that gap created by dropping supply and accelerating demand — and knocked around 10 gigawatts of generation, from all fuel types, offline. That is a lot of juice, roughly the power on a normal day for 3 million homes.
The effects of the power imbalance, aggravated by the winter conditions, led to even further losses as safety mechanisms tripped additional generation for safe operations.
ERCOT went into emergency operations. This, as is now well known, was when the uniquely isolated grid, with limited connection to other state systems, began to shed load to accommodate the loss of power. Load shedding events can be thought of as pulling weights off a dumbbell to relieve the lifter mid-press. In this case, ERCOT is relieving itself of stress because otherwise the whole structure could collapse. ERCOT shed load to prevent the weight of electrical demand from collapsing on the weightlifter, in this case virtually the entire state system. Recovery from a statewide blackout would have taken the isolated grid offline much longer than the four days of outages it experienced.
The “rolling blackouts” that utilities were reporting were not shared across all power users for a couple of fundamental reasons. First, switching on and off the power to distribution systems is a significant risk. They are designed for stable, constant power flows. Surging them off and on can cause permanent damage. Second, as power supply is a balancing act of supply and demand, switching on and off the wrong loads can risk disrupting the greater balance. Thus we saw a lot of extremes, no power or constant power, as distribution systems — the local utilities — prioritized critical loads to areas with institutions such as hospitals and fire stations.
Who was most affected?
The consequences of this decision were the trade-offs we have seen in the recovery. Burst pipes, downed infrastructure, and lost lives will be the stories that make headlines. However, these quick headlines will miss an underlying theme. Texas’ most vulnerable communities will almost certainly have been hit hardest. Many low-income communities correlate with inefficient housing and older power delivery systems, making them the first sections likely to be cut to conserve and optimize power delivery.
Much evidence from the Texas blackout is currently anecdotal. However, the stark contrast between the prolonged blackout in Austin’s East Side, a historically redlined district for minorities recently gentrified, and its glowing commercial downtown, suggest the legacy of a dated power system in a traditionally low-income, minority area.
We can also expect these vulnerable communities to be most affected by the costs of recovery. Insufficient power and poor insulation will result in extensive housing damage. Mold will form if not quickly addressed and the structure of some may be questionable after freezing and then thawing.
However, these physical dimensions are just the tip of the iceberg. A dynamic market, ERCOT’s prices change with the balance of supply and demand (More demand, higher prices, high supply, lower prices etc…). These power prices average around $22 per megawatt under normal conditions. The energy crisis sent ERCOT wholesale power prices soaring at over $9,000 per megawatt for nearly four days.
Cases of consumers suddenly hit with six-figure electrical bills have captured national headlines. Less explored is the wave of multi-million dollar charges that have swept through local utilities in Texas and that will be far more difficult for officials to simply waive or forgive.
Utility providers are facing a financial crisis as they are saddled with hundreds of millions in electric charges from ERCOT. Already, Denton Municipal Electric, which serves about 150,000 people 40 miles north of Dallas, is seeking debt financing for a $207 million power bill after the events of this week — roughly 89 percent of its $231 million budget and far more than it has on hand in cash reserves. The financial costs of this will ultimately be borne by ratepayers as utilities implement additional fees to cover their cash flow shortcomings. And that’s one case.
This will, again, most affect vulnerable populations with low and fixed incomes. Despite generally lower consumption, energy bills take up a larger portion of their budgets. According to a study conducted by the University of Texas at Austin’s Energy Institute, households earning less than $25,000 annually spend approximately 12.5 percent of their incomes on home energy as compared to the 4 percent expenditure of those earning above.
These vulnerable populations could face a real budget crisis as debt riders, added to pay for the energy crisis, increase their costs of energy, forcing them to choose between energy bills and other essential needs.
Why did it happen?
I need to dispel two major falsehoods. First, though charged with enabling the reliable operation of the system, ERCOT does not have the authority to force weatherization or require additional installed capacity. There may have been some things ERCOT could have done in advance, but natural gas pipeline resiliency, turbine winterization, or even the command-and-control construction of power plants for reserves are generally beyond its authority.
ERCOT has three primary tools at its disposal: transmission operation and development, price signaling for power generation, and load shed. Setting the standards for broader grid resiliency, and capacity has and always will be the domain of the governor-appointed, three-member Public Utilities Commission of Texas, or PUCT, and the Texas Legislature.
Second, there is no one event or resource that caused the trip of resources and subsequent load shed. There will be fingers pointed at wind power, some will blame natural gas, still others will blame the nuclear reactor that tripped off in south Texas. However, none of them are categorically to blame. The issue is systemic.
The heart of this problem is a power policy problem. Power policy in ERCOT has been focused on competitive power prices, but competitively priced power does not categorically incorporate reliability or resource adequacy. Utilities are not required to winterize because it has not been deemed cost effective against the low likelihood of a winter such as the one we’ve seen in recent days. In a similar vein, resource adequacy was designed to be economic, not necessarily sufficient. ERCOT’s Operating Reliability Demand Curve (ORDC) price adder and $9,000 price cap were supposed to incentivize the construction of new generation via price signaling as opposed to policy. Put simply, instead of regulated utilities justifying their power plants, the policy design assumed lucrative prices would attract new generation development.
How do we prevent this from happening again?
Before it can change, policy makers and rate payers need to ask themselves, did the competitive market design of ERCOT fail? At a theoretical level, the market mostly worked as designed. Prices rose with demand and demand was reduced through outages when it exceeded supply. The economics may be a bit abstract, but the design of ERCOT and its price caps and valuation of load assumes that when resources cannot meet demand, shares of the load are shed accordingly. ERCOT may not have intended the duration of outages as events played out, but it was the inevitable result. That said, at an intuitive level, the competitive market failed in its current design. On a day-to-day basis the grid may be fine, but decades of state policies driven by the goal of cheap power have stripped ERCOT of its resiliency and reliability for emergency events.
There is no quick fix and there is no cheap fix. The traditional method is to require generators and natural gas pipelines to winterize. However, these retrofits will be costly, since operators will have to open up existing wells and plants and design custom solutions. It would be cheaper and more effective to build new, but that would mean lost revenues from retiring resources early. Moreover, whether newly built or retrofitted, sufficient system wide winterization would take at least a couple of years.
Legislators and the PUCT could also implement capacity constructs, converting ERCOT’s energy only-market to something akin to the capacity markets of its power transmission peers PJM on the East Coast and Midcontinent System Operator, or MISO, which operates in parts of the Midwest and South. This could incentivize the construction of generation for the purposes of having additional capacity. However, doing so would take years of policy work and would significantly increase the cost of power to consumers.
There are some “non-traditional” methods that policy makers could consider for more rapid resiliency. The first is the development of more distributed generation, particularly microgrid concepts, capable of serving small islands of load when the greater grid fails or is stressed. These could be prioritized towards emergency centers and dense urban developments where the most people could be serviced off a single electrical feeder.
Another alternative is the broader integration of battery technology at the housing level. Though temporary, as batteries would run out of power and still need to charge via the grid, batteries would be capable of load shifting.This is effectively moving the consumption of power to a different hour. Consider the below graph, the yellow line with the small hills represents solar production. The problem with these hills is that they were completely off peak from load, which, during the events of this last week,was greatest in the cold night. Batteries could have supplied stored power during peaks and consumed solar power during sunny hours with lower demand.
Still, none of these proposals will provide the resilience ERCOT needs within a year. The reason for this is simple. Power planning is a long, complicated process designed to look at the consumer needs on 10-, 20-, and 30-year horizons. The grid today is the consequence of nearly 20 years of decision making shaped around goals cheap, price competitive, summer-peaking power.
Just as fixes to this system won’t come cheaply, they also will not come fast.
If you like what you’ve been reading, please click here to subscribe and we will send you updates and our newsletter.