Evolving Electricity Markets

Experts answer: What will US electricity markets look like in an emissions-free world?

Molly Robertson: Wholesale electricity markets look different across the country and across the globe. The United States has a mix of what we call deregulated electricity markets and vertically integrated electricity suppliers.

Prior to the 1990s, almost all electricity users were served by vertically integrated monopolies which owned the generation, transmission, and distribution systems that delivered electricity from power plants to households. This structure meant that wholesale markets played a very limited role in determining which generators provided power. In these monopolies, utilities were heavily regulated, and prices were set by regulators so that utilities could cover their own costs, thereby limiting the ability of utilities to charge monopoly prices.

During the 1990s, a wave of restructuring took place, which took advantage of the fact that the generation component of the electricity system could be part of a competitive market. Several regions of the United States created competitive markets to provide electric power; these markets allow generators to offer bids in an auction, and the lowest-cost generators are selected by an independent system operator to provide power. This system of competitive markets is common today, although vertically integrated utilities are still operating in some regions.

Karen Palmer: The markets that are run by the independent system operators are structured to provide an incentive for generators to bid into the competitive energy market at their marginal cost of producing electricity. At that price, the generator is indifferent to either operating and receiving the price, or not operating and saving the fuel and other operating costs.

Generators are selected for operation starting with the lowest bid price. When an auction price reaches a value that ensures equivalent supply and demand of electricity; the market has “cleared.” For electricity markets, generators make money when the clearing price is higher than their marginal cost. These monetary gains go toward paying down the fixed costs incurred by the generators and, hopefully, contribute to profits.

What happens if a generator isn’t selected to operate during the auction process?

KP: If a generator’s marginal costs are consistently higher than the marginal costs of competitors, then the generator may clear the market only at rare times when demand peaks or when other generators are out of service. In some cases, these infrequent periods of operation yield revenues that are sufficient to cover the generator’s costs. If particular generators can’t operate often enough to earn revenues that are sufficient to cover their fixed costs, then those generators are forced to retire.

MR: The energy market, in which generators are paid to provide electricity to the grid, is the main way that most generators make money.

But for the grid to operate reliably, other services are required, which generators could provide for a fee that helps cover the costs. For example, generators may sell capacity commitments, which require the generators to be available just in case their power is needed. The payments for capacity commitments can be an important source of revenue for generators that don’t expect to operate often but still may be needed in times of high demand.

Generators also may sell something called ancillary services, which refer to a suite of services that help grid operators keep their very complicated technical systems running smoothly. For example, a frequency-response service allows generators to sell the system operator very small, instantaneous adjustments in the generator’s output to ensure that the grid maintains its necessary operating frequency. Only some generators have the ability to make those very small changes in generation instantaneously. Batteries have been highly valuable providers of frequency-response services. Ancillary services generally are low-price products, but they’re still meaningful in the revenue stream for some generators.

Markets have to accommodate a lot of components in a well-functioning power system. How do wind and solar resources play into all this?

MR: Renewable energy resources like wind and solar are different from other resources, because the fuel they use to create electricity is free. When wind and solar resources offer their bids in the energy market, they typically can undercut the fossil fuel competition. In some cases, subsidies that compensate renewable generators for operating encourage renewable generators to offer negative bids into the market—the generators essentially offer to pay the market to allow them to operate and collect the subsidy.

At low quantities on the grid, renewable resources are all upsides: low-cost, zero-carbon resources.

At high levels of penetration, however, renewables challenge current market mechanisms by decreasing electricity prices and therefore decreasing the expected revenue in the day-ahead and real-time energy markets. Uncertainty over these low prices may make it more difficult for developers to determine if they should invest in new electricity generation.

KP: The other key characteristic of renewables is their intermittency. The fuel may be free, but it can’t be controlled by the operator. A solar farm doesn’t have to pay for sunshine, but it also can’t change the weather if the forecast calls for clouds and rain. Although wind and solar may offer low-cost electricity during many hours of the day, they may disappear altogether at other times. To ensure that these situations of intermittency don’t lead to outages, the grid has to find a way to ensure that enough backup resources are ready to jump in. Because these backup resources are operating infrequently, the generators have fewer hours of potential operating revenues available to recover costs.

Many market observers expect that a market with high penetration of renewable energy will face very large swings in energy prices as the generator that sets the auction clearing price switches between the low-cost renewables and the expensive backup resources. In some cases, auction-clearing energy prices may exceed the cost of the backup resources if the willingness to pay for additional energy, known as the value of lost load, exceeds the offer price of the backup resources. Prices set at this relatively higher level could signal the value of future investments, but such pricing is generally restricted by regulator-imposed price caps.

Are big price swings in wholesale markets a problem for generators?

KP: Price swings driven by variable resources pose a problem in terms of the long-term investment signals for power generators, particularly because the frequency of high-cost periods hinges largely on weather patterns that are difficult to predict.

Generators are expensive, and they take a long time to build, particularly with layers of permitting and interconnection systems that can add years to development timelines. Investing in generation becomes increasingly risky because large swaths of time could pass during which expensive generators may expect to make no revenue, and the generators would need to make sure they are available during the hours that prices spike, so they can recover their costs. The uncertainty over the frequency and duration of those price-spike periods makes it difficult for generators to obtain the necessary financing for significant project investments.

MR: These price swings could be mitigated through solutions such as building up storage technologies and allowing electricity users to respond to prices. By holding power in batteries or other technologies, energy-storage operators can buy excess energy when the energy is cheap (like when the sun is shining) and can make stored energy available for sale in the market when prices go higher (like when the sun goes down).

On the demand side, opportunities are available for electricity users to reduce their consumption during times of high prices. For example, a customer may plan to charge their electric vehicle in the middle of the day when prices are low due to the availability of solar power. Or a large industrial plant may agree to pause operation if electricity prices are sufficiently high. For demand response to deliver these benefits, electricity users must be exposed to these price swings and adjust their behavior accordingly.

For both demand response and storage, we still have questions about how much we can expect these options to contribute to solving the problem of price swings. For example, if we see a sustained shortage in renewable resources and associated high prices, on the order of weeks in duration, then storage and demand response may be insufficient to address system needs.

So, integrating renewables comes with opportunities and challenges. What kinds of adaptive measures could help address these issues?

KP: Economists are considering how redesigning electricity markets could improve investment signals and ensure that the generators which are needed on the system can make sufficient revenue to continue operating.

Current market structures rely on different mechanisms, such as capacity markets, power purchase agreements, minimum contracting obligations, or shortage prices, to facilitate enough power on the grid to serve demand. But those mechanisms may be insufficient or inefficient in procuring the right mix of resources.

Jurisdictions outside the United States are exploring different ways of signaling the need for investment in important new resources, like forward contracts for energy (in which both parties agree to buy and sell energy at a specified price on a future date) or contracts for differences (where power buyers pay generators the difference between the contracted price and the variable-energy market rate when the former is greater, and vice versa when the latter is greater), both of which can provide a stable, long-term revenue stream. Such solutions may provide more certainty for investors and help stabilize revenue streams for generators, but the details of these solutions matter for determining what types of energy projects actually get built and how much the systems cost to operate.

As reforms to the electricity market are considered, what priorities should regulators keep in mind?

MR: We think about evaluating changes to electricity markets in various ways, and whether the changes can meet the needs of the system.

First, solutions should enable the market to find low-cost approaches to meeting the needs of the grid, whether these approaches involve examples like meeting demand in real time or ensuring that the operating frequency of the grid is maintained.

Second, reforms to market structures should be built in a way that enables new and emerging technologies, which have different attributes, to provide services and compete.

Third, solutions should consider not only the ability of generators to increase supply, but also of customers to reduce demand.

Finally, as decarbonization policies at the state and federal levels continue to evolve, market-design solutions should be developed with the impact of those policies in mind. For example, if a state climate policy will require backup fossil generation to retire by a certain date, then markets should provide sufficient signals for new sources of non-intermittent clean generation to be built by that date.

What is RFF doing in this space to inform policymakers?

KP: We are spinning up work in this space and have a lot of ideas for how we can contribute to the policy dialogue. In an ongoing collaboration with Chiara Lo Prete of Penn State University, we are reviewing existing research and the international policy space to explore potential solutions to these market challenges.

The next phase of our work will involve a model of the power sector. We’ll explore a subset of proposals in greater detail with the model, to see if the proposed solutions could improve signals for investors.

We also are thinking about the role of emerging technologies in electricity markets, such as energy storage, and the challenges related to fitting them into the current structure.

Finally, we are continuing to think about the responsiveness of electricity consumers to price variation, so we can quantify the opportunity for demand-side participation in electricity markets.

This full range of work will inform conversations about designing electricity markets for a net-zero future.