ERCOT and the Texas Power Grid
The Electric Reliability Counsel of Texas (ERCOT) has the authority, as mandated by the Texas Public Utility Commission (PUCT), to oversee the electric market and system dispatch operations spanning nearly 90 percent of the state. ERCOT manages a stand-alone grid system, meaning that there is very limited capability for generation resources outside of the ERCOT footprint to flow electricity into the Texas system. As a result, ERCOT’s system actually has a lot in common with island utility systems that are designed with additional resiliency measures to maintain reliability due to characteristic challenges.
Surplus electric generating capability is essential for system reliability. This is called generation reserve margins in the power sector. Required reserve margins across control areas within the Eastern and Western U.S. Interconnections range from 12–15 percent above peak demand. This level of surplus generation has proven effective for these interconnected systems because electricity can flow between control areas during emergency conditions, providing external redundancy for temporary deficiencies.
ERCOT has historically targeted system reserve margins that mirror their interconnected counterparts. If ERCOT were a geographic versus electrical island, its system reserve margin would likely reflect that of other island utilities that are nearly double the current ERCOT target. Obviously, additional reserves add more cost to the system.
Market Structures and Characteristics
As a deregulated merchant market, ERCOT employs an energy-only market construct. The market construct reflects the means by which generators are monetarily compensated for supplying electricity to the grid. Energy-only means that there are no fixed revenue streams available to generators—they essentially make money only when they generate and deliver electricity onto the system.
With no other market mechanism available for compensating generators in ERCOT, an energy-only revenue stream must cover variable production costs and fixed operating costs and provide any return on investment that is realized through investing in, and properly maintaining, a power generation fleet. Under an energy-only market, power generators face volatile earnings profiles making it more difficult to manage operations, determine optimal new construction entry, and justify investments. By comparison, generators in regulated markets are compensated for their capital investments through fixed electric utility rates that are set by the state utility commission and provide certainty of cash flows over the term of the investment horizon.
In other U.S. deregulated markets (e.g., the PJM Interconnection), there are additional market mechanisms, in addition to energy payments that generators receive, called capacity payments. Capacity payments can reflect a substantial percentage of the overall revenue stream that generators receive in the form of a fixed payment. Similar to regulated utility rates, capacity payments provide a steady cash-flow stream for building new plants, budgeting maintenance and capital improvements, and achieving a return on investment. Unlike utility rates, capacity market payments are predicated on the availability of the generating plant, meaning that there are severe financial penalties associated with nonperformance. This structure creates a strong incentive for operational integrity and reliability as capacity payments represent a significant percentage of total revenues that generators receive: they want to ensure that this revenue stream is not threatened by unreliable, inoperable equipment.
Texas adopted an energy-only market structure years ago during deregulation of the electric markets. Its attributes were extolled by regulators and politicians alike, who praised its competitive construct and the ability to allow market fundamentals to signal the need for new generating resources through increasing, and at times volatile, power prices. New generation will enter the system in response to these price signals, and ultimately prices subside as supply scarcity decreases. While theoretically viable, an energy-only market conceals flaws induced by regulatory/political externalities and technology advancements, which are usually only exposed through extreme circumstances.
Dispatchable Versus Intermittent Generating Resources
In addition to the energy market structure, another key characteristic of the electrical grid is the mix of generating resources. Dispatchable resources are generating plants that can be turned on and off at the direction of the system operator. These typically fall within the category of fossil or nuclear plants. In contrast, intermittent generating resources are only available when the wind blows or the sun shines—they provide energy to the system when they’re running, but only nature can control when they run.
Obviously, an electrical grid needs dispatchable resources to maintain system integrity as demand fluctuates constantly throughout the day. Intermittent resources are great sources of energy and environmental efficacy; however, they limit the system operator’s ability to meet ever-changing demand from sporadic generating profiles.
Currently around 25 percent of the total generating capacity in ERCOT is represented by intermittent resources; and with thousands more megawatts currently in the queue, this number could approach 50 percent in just a few years.
Generating Capacity: Comparison of System Reliability, Investments, and Costs
As outlined previously, all generating resources are not equal in function or value from a system reliability standpoint. Dispatchable resources are the most valuable from a system integrity and operational efficiency perspective as dispatchable resources are available when demanded, and electricity must be generated at the moment it’s consumed.
Again, intermittent generation resources utilize atmospheric elements to achieve low variable costs and zero emissions, but they can’t be dispatched precisely when needed. However, energy-only markets tend to see higher levels of merchant investments in intermittent resources because renewable resources aren’t financially disadvantaged in energy-only markets for not being dispatchable. When they run, they tend to realize higher prices than their counterparts under energy and capacity market constructs, making intermittent economics more favorable in energy-only markets. Dispatchable resources, on the other hand, must commit hourly power delivery schedules with the system operator, and they face significant financial losses if they fail to deliver.
The discussion to this point has omitted the topic of fixed costs attributable to fossil versus renewable generation investments. While these costs must be factored into the overall economics of any discussion of energy costs, our focus has been on reliability and system integrity. However, it should be noted that while the capital costs of renewable resources have declined significantly over the past decade, the fixed costs attributable to constructing renewable resources remain well above those of natural gas–fueled resources—hence, the need for government subsidies to justify renewable resource investments in ERCOT.
ERCOT: Bias in Favor of Intermittent Generating Resources
The energy-only market structure in ERCOT essentially puts all generating resources, both dispatchable and intermittent, on the same playing field for revenues. Both earn revenues only when they deliver electricity—yet, as we have seen, both are not equally reliable for meeting dynamic system demand. There is no differentiation from a revenue perspective for a reliable versus intermittent generating resource under an energy-only market structure.
What’s more, the market structure actually puts some dispatchable generating resources at a distinct financial disadvantage. As previously stated, in an energy-only market, generators earn revenues only when they operate and deliver electricity to the grid. Introduce federal renewable energy subsidies into the equation, and the economics can invert.
Renewable resources earn production tax credits for every megawatt hour of energy they deliver to the grid—representing a federal subsidy that is available only to renewable generating resources. During times of low demand, fewer generating resources are needed, and the costlier resources drop from the system as prices fall and economics wane. When factoring in production subsidies, a wind generator can economically operate even when prices go to zero or even become negative, meaning that the subsidy will provide a make-whole solution for the renewable generator up to the value of the subsidy. Because many fossil generating plants must remain online due to operational limitations and other factors during these negative pricing periods, they must pay to remain online. That’s correct: fossil generating plants pay to operate, while their intermittent counterparts get paid to operate.
Market Failure and Needed Changes
The structural market flaws in ERCOT can be summed up as a fundamental truth that Uri clearly exposed: Physics and finance clash at the intersection of reliability and independence. I would argue that you can’t sustain long-term reliability on an isolated grid system through an energy-only market design alone. And market designs are ultimately the responsibility of the PUCT and lawmakers. Nevertheless, during his testimony to the Texas legislature in February, ERCOT president Bill Magness said something that I believe few focused on at the time: “[W]e manage the grid that is handed to us.”
Reliability costs money. Period. Reliability on an isolated grid system costs even more money. The market structure in ERCOT must include mechanisms and rules that advance reliability through appropriate economic incentives to ensure investment across a mix of generating technologies that are necessary to reliably operate an isolated grid system. It’s up to legislators to make the choices required to remedy the situation—kicking the can down the road has resulted in power failure.
A market redesign needs to reflect a revenue structure that financially differentiates dispatchable, reliable generating resources from intermittent resources. It also needs to factor subsidies into the operational equation, ensuring that those that are intended to benefit from subsidies do so while not systematically penalizing those that are not. Such changes would facilitate optimal generation technology investments and timing, and reward those generators that maintain reliable equipment that is available and dispatchable under any situation. Not all generating technology is equal—and any market construct that continues to put them on equal economic footing will only serve to perpetuate the circumstances leading up to the market failure experienced this past winter.
A market redesign must also reward the attributes and benefits that renewable resources bring to the system. For example, ERCOT’s Responsive Reserve Service (RRS) market would be more economic if market rules rewarded the attributes and reduced the risks associated with battery storage facilities as they participate in this ancillary services market. Currently, RRS market rules unnecessarily put battery storage facilities at extremely high financial risk even though the technology they afford is highly efficient for this market application.
No market structure is perfect. There are trade-offs and compromises required to achieve all interests. However, in reflecting on the ultimate cost of Winter Storm Uri on the Texas grid, consider this: The excess costs incurred as a result of the storm on the power grid alone would have funded a fleet of brand-new natural gas–fired generators sufficient to meet the entire peak demand of the ERCOT system by itself. That’s right—ERCOT could have achieved well over a 100 percent reserve margin for the same cost that this storm imposed on its system.
Recognizing the consequences of these market design flaws will be an important lesson for regulators and politicians, who may lament the additional cost of a market overhaul in ERCOT. It seems that Texans, directly or indirectly, have already incurred the additional costs attributable to ensuring reliability on an isolated system—without realizing any improvements in reliability.