Category: Federalism

• Guest blogger Joshua Macey is here to discuss his new paper on how electricity regulators and grid operators are responding to increased solar and wind power, and how their interventions raise old questions that were supposed to be resolved by electricity deregulation. You can also hear an Energy Tradeoffs interview with Joshua about his piece here.

In Rate Regulation Redux, forthcoming in the University of Pennsylvania Law Review, Jackson Salovaara and I consider whether the American system for compensating electric power generators can accommodate high levels of renewables. We find that the current market structure is ill-suited to a high-renewables world. Regulators and grid operators (grid operators are the utilities that manage the power grid), it seems, are aware of the challenges renewables pose. However, instead of developing a payment system that would preserve competition in the energy sector and allow renewables to enter the market, regulators have begun an ad hoc process of reregulation that raises rates, leads to excess capacity, and prevents renewables from competing with traditional energy sources.

For most of the twentieth century, FERC treated electricity as a natural monopoly. To ensure that suppliers met demand, regulators gave utilities exclusive franchises over their service territories and permitted them to charge rates sufficient to cover their costs. In exchange, generators agreed to provide electricity to customers in their territories and cap prices. For years, this system provided reliable electricity. Nonetheless, critics complained that it limited consumer choice, failed to promote innovation, rewarded utilities for overinvesting in supply, and reduced incentives to retire uneconomic generators.

In the 1990s, FERC began to encourage a “market-based” approach to promote competition and control costs. Under this “restructured” model, which has been adopted in two-thirds of the country, an independent grid operator determines demand for electricity, solicits bids from generators, and clears enough bids to meet demand. The grid operator clears bids starting with the lowest bid but ultimately pays every generator the price bid by the highest clearing bidder. In this system, generators bid at their marginal cost. If a generator bids below its marginal cost, it risks having to provide electricity even when it would lose money in doing so. Above-marginal cost bids risk failing to clear when it would be profitable for the generator to operate.

This system promotes competition and keeps short-run costs low, but it is ill-equipped to integrate significant volumes of renewables. Generators that are dispatched infrequently or operate on the margin cannot make a profit or recover their costs. These plants are known as “peaking plants” and operate a few times in a year when demand is high (often on the hottest days of the summer or the coldest days of the winter when Americans consume a lot of electricity). Without them, grid operators would not be able to meet peak demand.

In theory, peaking plants would be able to make enough money to operate. While generators bid their variable costs almost all of the time, that assumption does not apply to peaking plants that bid only when demand is high. In most circumstances, a generator risks losing out on profitable bids if it submits a bid above its marginal costs. Because peakers are the last plants to be dispatched, they do not need to worry that they will be outbid because there are no plants available to outbid them. They can therefore submit bids that significantly exceed their marginal costs. As a result, peaking plants can drive prices to levels that would allow them to recover their fixed costs and make a profit despite the fact that they operate only a few times a year

However, a system that relies entirely on energy markets can lead to rampant market manipulation and excessive price volatility. Peaking plants have market power. Because they are the last units dispatched, if they do not operate, there will not be enough electricity to meet demand (these incentives contributed to the California energy crisis in the beginning of the twenty-first century). Peaking plants can therefore drive prices to extremely high levels. To avoid these problems, every regulator in the United States sets a ceiling on its energy market’s clearing price.

Unfortunately, a system that imposes price caps on energy markets is ill-equipped to integrate significant volumes of renewables. Renewables have a different cost structure than other generators. While the marginal costs for most generators are above zero, wind and solar facilities have very low operating costs. As renewables provide an increasing percentage of electricity, they suppress revenues for all generators. Imagine if four generators had been providing electricity to a region. Two bid $20 MWh, one bid $40 MWh, and one bid $50 MWh. All four generators were paid $50 MWh. If, however, a solar plant replaces the $50 MWh generator and offers $0 MWh, it will reduce revenues for all generators. That is because it will drive out the $50 MWh generator, making the $40 MWh generator the clearing bid, which means that every generator would be paid $40 MWh.

The challenge with this system is that as renewables suppress prices, energy markets become increasingly reliant on price spikes to ensure that all generators receive sufficient revenue. But if regulators do not increase price caps, then energy prices will not increase enough during scarcity to allow generators to make enough money to continue to operate.

There are a few possible solutions to this problem. One would be to increase price caps, but no regulator (with the possible exception of Texas) has expressed a willingness to let spot market prices rise enough to allow generators to recover their costs. Regulators have been reluctant to allow prices to rise to extremely high levels out of fear that doing so would encourage market manipulation. 

Another option is to develop other markets that would ensure that crucial generators are able to survive. To date, most regulators seem inclined to adopt this approach. Unfortunately, the markets regulators are developing do not allow meaningful competition between energy sources and instead prevent renewables from competing with traditional generators. For example, grid operators in the east coast have begun to rely on capacity markets, which pay generators for being available to provide electricity instead of for actually providing electricity, to make sure that generators receive enough revenue to continue operating. The problem with this this system is that capacity markets do not actually reward generators for providing the services the grid needs. Not all electricity has the same value. Generators that can turn on and off quickly, that can provide electricity when it is most needed, and that provide electricity to areas that are resource-constrained should be rewarded for providing these services. Energy markets are uniquely effective because they reward generators that provide electricity where and when it is really needed. Capacity markets fail to do this. To make sure that the “right” generators are being compensated in capacity markets, some grid operators have taken steps to make it more difficult for some types of generators (often renewables and nuclear) to enter capacity markets.

Equally problematically, it is difficult for generators to exit the market once they clear a capacity markets. Generators that clear capacity markets commit to operating for a period of time (often three years). During that period, they are not permitted to exit the market unless they receive regulatory approval to do so. Thus, customers are often stuck paying for dirty electricity that is no longer necessary for the grid.

Worse still, grid operators have begun to rely on “reliability-must-run” (RMR) agreements to provide even more competition to the generators that are perceived to be critical to grid reliability. When capacity markets are not able to retain generators perceived to be critical to grid reliability, grid operators have simply bailed individual generators, and they have done so without any kind of competitive bidding procedure.

In our view, these interventions resurrect many of the principles of rate regulation. Under that approach, regulators gave generators a rate of return intended to make sure that the electricity companies would be able to meet all of a region’s electricity needs. In exchange, power companies provided service at agreed-upon rates. Today, regulators have begun identifying the generators that the grid needs, making sure those generators receive enough money to operate, and preventing them from retiring prematurely. Rather than rely on market forces to determine which generators operate, regulators shield preferred generators from competition in order to ensure that those generators are financially viable. And these generators are required to provide the services the grid needs.

A superior option, which we endorse in the paper, is to design a system based on long-term contracts that would impose penalties on generators that fail to perform as promised. Some of the problems with capacity markets are that they do not compensate generators that provide the services that the grid needs, they prevent generators from competing with each other, and they make it difficult for uneconomic and superfluous generators to exit the market.

Regulators and grid operators want to ensure that there is enough capacity to provide electricity to consumers throughout the year. A bidding process would allow utilities to purchase the electricity that they need. Utilities would have an incentive to keep cost down because doing so would allow them to lower their own costs. And, by penalizing generators that fail to provide services they agreed committed to, this approach would preserve short-term price signals that create incentives for generators to provide electricity where and when it is needed.