EXPLAINING THE GRID PART FIVE
RTO Capacity Markets (plus a bit of history)
In this post, I discuss RTO capacity markets. What is the capacity product sold in an RTO capacity market? It is not a physical product, like the electricity that is sold in the RTO energy markets. Instead, a generator selling capacity in an RTO capacity market is obligated to offer energy into the RTO energy market in every hour, subject to specified exceptions in which the generator is permitted to be offline for maintenance or other authorized exceptions. The capacity obligation does not specify a price at which the energy must be offered into the energy market. The point is that the generation owner is obligated offer a certain amount of energy into the energy market on a 24/7 basis, which the RTO then can purchase if needed to serve its customer load.
To understand why RTOs have capacity markets, we first have to return to RTO energy markets. As I explained in my last post, the pricing in the RTO energy markets is intended to replicate the prices in competitive markets. One important aspect of this pricing mechanism is that, as demand increases, the market price increases. In theory, these higher prices when demand is higher provide the necessary incentive for generation owners to construct and keep in service enough capacity to meet that demand. Even if a generation facility’s offer into the energy market is accepted only a few hours a year during the period of the highest demand on the hottest day of the year, the price paid to the facility’s owner will be high enough to make it profitable to keep that facility in service. Based on this theory, when RTO markets were first being developed, it was widely believed that no other pricing mechanism was needed to ensure that there always is enough generation capacity in an RTO to serve all customer loads.
The validity of this assumption came into question shortly after the establishment of the first RTO energy market, which was run in the California Independent System Operator (CAISO) RTO. The belief at that time was that the energy market would substantially reduce the price of energy in California. Potential supplies of energy were expected to be significantly greater than demand, and this excess of supply over demand was expected to result in relatively low energy prices. Because energy prices were expected to be low, the utilities in CAISO were required to purchase all of the electricity they needed from the market, and were prohibited from entering into long-term power purchase contracts, whose prices would be higher than the prices expected from the energy market.
For a few years, until the summer of 2000, this assumption held true. But in that year, a perfect storm of events changed the equation. Traditionally, California has relied heavily on large amounts of relatively low cost energy generated by hydroelectric dams in northern California and the Pacific Northwest. The year 2000, however, was a very hot year with low precipitation, and in the summer of 2000 the source of this cheap hydropower energy literally dried up. At the same time, there were unexpected outages of fossil-fired generation facilities. Further, the hot weather caused higher than normal demand. And there was less than the usual amount of imports of power into California from surrounding states because the generation in those states was needed to service the high in-state customer loads in those states resulting from the hot temperatures. Mix these factors together and, voila, you have the California Energy Crisis of 2000! (No, the crisis was not caused by Enron, a convenient scapegoat which profited from, but did not create, the problem)
When the expected sources of cheap supply did not show up in the market, prices in the California energy market skyrocketed. At times, the supply situation got so bad that there was not enough energy offered into the market to satisfy all anticipated customer demand. Compounding the problem was another ill-considered feature of the California market. The California utilities’ retail rates to their customers were frozen and did not reflect the cost of their purchases from the energy market. The idea was that the utilities would be able to buy low and sell high. They would then apply the difference between their expected low cost of purchases from the energy market and their higher retail rates charged to customers to pay off so-called “stranded costs,” which were the costs of the utilities’ assets that were not expected to be profitable in the new system. Once a utility’s stranded costs were recovered, then its retail rates would be tied to the cost of purchasing from the energy market, which was expected to result in a lowering of retail rates.
None of the California utilities had completely recovered their stranded costs at the time of the California Energy Crisis. This meant that they were now buying high and selling low, purchasing energy at a price that greatly exceeded the frozen rates they were charging their retail customers. That was good for customers, at least in the short run, but disastrous for the utilities that ran up huge losses in a short amount of time, threatening their economic viability. And economically weak or bankrupt utilities would be unable to purchase the high priced energy from the market, leaving them unable to serve their customers.
Ultimately, the crisis was resolved when FERC approved caps on the price that could be charged in the energy market; a California agency entered into longer term power purchase agreements at fixed prices that it then sold at cost to the utilities; and the balance between supply and demand started to return to previous levels. At the same time, accusations and blame for the crisis were cast in every direction, and litigation proliferated. This litigation lasted far longer than the crisis itself. Even today, 24 years later, a few California Energy Crisis cases remain in litigation.
That was probably more (or maybe less) than you wanted to know about the California Energy Crisis. I mention it because it was very traumatic for those of us in the energy space that lived through it, and it highlighted a feature of relying solely on energy markets that many had not focused on. Energy markets can be volatile, and with virtually no reduction in demand when prices are high (for reasons I explained in my Post Explaining the Grid Part Two—Economics), market prices can rise to very high levels when the energy markets are stressed. In addition to causing economic and operational problems, these high prices attract much public attention and, inevitably, political pressure to do something.
After the California Energy Crisis, energy market purists continued to insist that sole reliance on energy markets is the optimal approach. To them, the California Energy Crisis confirmed this view. They argued that the high prices represented appropriate price signals that more generation capacity was needed in California, and that those high prices would provide an economic incentive for developers to construct new generation facilities that could sell into the California energy market.
Others, including utility owners, customers, and state and federal regulators, weren't so sure they wanted to repeat California's experience. And RTO capacity markets represented one possible solution. By entering into contracts with generators obligating them to offer enough energy into the market to meet anticipated customer load levels plus a reserve, RTOs expect to be able to avoid the extreme imbalance between supply and demand that prevailed during the California Energy Crisis.
Capacity markets also can address what generation owners refer to as the “missing money” problem. This refers to the fact that the RTO energy markets may not generate enough revenues to support the continued economic validity of generators participating in those markets. This missing money problem potentially is worsened by the fact that FERC has imposed price caps on the energy markets. Those price caps can prevent energy market prices from reaching the level of the true marginal cost of energy, thus theoretically depriving generators from earning the revenues they need remain in service and be available to generate electricity in high demand conditions.
Unlike the RTO energy markets, however, there is no agreement among RTOs on how capacity markets should be structured, or even if there should be an RTO capacity market or any capacity obligation at all. Every RTO has a different approach, which I summarize briefly below. As with all my posts on grid operations, there are numerous complicated details that I am not going to cover, but instead am giving an overview.
Let’s start with the Electric Reliability Council of Texas (ERCOT), the last holdout of the energy market purists. ERCOT relies solely on its energy market to provide the revenues needed to support the necessary generation capacity to serve its customer load. ERCOT has no capacity market and utilities in ERCOT have no obligation to acquire capacity from generation owners independently from the ERCOT markets. If you have paid any attention to the news in the last few years, you know that ERCOT’s sole reliance on an energy market has not worked out too well recently. Lacks of supply during periods of high customer demand in both the winter and the summer has resulted in rolling blackouts and energy prices as high as $9,000/megawatt-hour. Will ERCOT change after suffering through these debacles? Stay tuned.
Two other RTOs—CAISO and the Southwest Power Pool—do not have capacity markets, but instead impose an obligation for utilities and other entities supplying retail customers to either own generation facilities or enter into capacity contracts in amounts necessary to serve anticipated customer load. This approach addresses the concerns that the RTO energy markets are too volatile and that the energy markets alone do not adequately ensure that sufficient generation capacity will be constructed.
The four other RTOs do have capacity markets in addition to energy markets. These markets have some similarities. Unlike the energy markets, where the amount of energy supplied is set equal to customer loads as they exist in real time, capacity is acquired in advance. As a result, each RTO determines the total amount of capacity it determines is necessary to ensure reliable service, based on its predictions of its peak loads and on contingencies. Each RTO posts this amount in advance of the auction, along with other technical details needed for generators to participate.
Another similarity in the RTO capacity markets requires a bit more background; something I forgot to add to my prior post on energy markets. Traditionally, almost all electricity was bought and sold by utilities, which were considered to be natural monopolies. To prevent the utilities from abusing their monopoly power, the rates for the power they bought and sold were set at cost-based levels, reviewed and approved by regulators applying an amorphous “just and reasonable” standard.
Starting in the 1980s, however, generation facilities began to be constructed or purchased by “independent” owners, i.e. entities that are not utilities or affiliated with utilities. These independent owners, which were not natural monopolies, did not sell the electricity they generated directly to retail customers, and had no ability to exercise market power to influence the rates at which they sold their electricity at wholesale. As a result, the rationale for requiring cost-based rates set by regulators did not apply to sales by these independent owners.
Starting in the mid to late 1990s, FERC—which regulates wholesale sales of electricity—determined that rates established under competitive conditions free from the exercise of market power are per se just and reasonable. FERC then established a process whereby any generation owner (including a utility) able to demonstrate it does not have market power is free to sell its electricity at wholesale without first receiving approval from FERC of the rate charged for that sale.
Today, FERC has ruled that sellers in RTO energy and capacity markets either have no market power or that RTO markets have safeguards in place that prevent any generation owner with market power from abusing their market power. Therefore, all generation owners are allowed to participate in the RTO energy and capacity markets, and the prices that are established in those auctions are deemed to be just and reasonable without any FERC review of those prices.
With that explanation, let’s explore the four different RTO capacity markets. Two of these, the PJM capacity market and the New England Independent System Operator (ISO-NE) capacity market, are somewhat similar. Each RTO conducts annual auctions for the delivery of capacity three years later. The theory for conducting the auctions three years before the capacity obligation commences is that, by conducting the auctions so far in advance, developers of potential generation projects can obtain a capacity award far enough in advance to finish construction and furnish the capacity three years later. In these RTOs, the capacity award represents a guarantee of revenues for the project that can be used to help obtain financing.
Despite this similarity, the capacity auctions themselves are very different. PJM’s auction is much like the RTO energy auctions. The various offers it receives are stacked in order from lowest to highest, with the price set equal to the price offered by the marginal offer. The ISO-NE auction, by contrast, is a declining price auction. ISO-NE starts by listing a very high price, and all generators indicate whether they would sell at that price. If more than the required among of capacity is offered at that price, then ISO-NE reduces the price, and generators respond as to whether they would sell at that reduced price. ISO-NE then keeps reducing the price until the amount of capacity offered by generators remaining in the auction equals the amount of capacity ISO-NE has determined it needs.
The New York Independent System Operator’s (NYISO) capacity market is unique among the four RTO capacity markets in several respects. First, NYISO conducts shorter-term spot and monthly capacity auctions, which it evaluates using the stacking process I have already described. Unlike PJM and ISO-NE, NYISO’s auctions are not conducted three years in advance. And unlike the other RTOs, NYISO conducts capacity auctions setting prices for capacity at four different locations in New York instead of setting a single RTO-wide price.
Finally, there is the Midcontinent Independent System Operator (MISO). MISO is different from the other three RTOs that run capacity auctions in one important respect. Most of the utilities in the other three RTOs divested ownership of their electric generation facilities in the 1990s and early 2000s. These utilities thus have no alternative for obtaining generation capacity other than purchasing it from independent generation companies. In MISO, however, most utilities retain ownership over their generation facilities. Because they own their generation facilities, these utilities do not need to purchase large amounts of generation capacity from third parties and some do not need to purchase any capacity from third parties. As a result, the main focus of the MISO capacity auction is to supplement the capacity of any utility that does not own enough generation capacity to serve its own customer load.
Like PJM and ISO-New England, MISO conducts annual auctions to provide capacity for one year, but unlike the other two RTOs, MISO’s auction is not conducted three years before the year in which the capacity is to be provided. But because the MISO capacity auction is largely a supplemental auction, the capacity auction prices in MISO typically have been significantly lower than in PJM, ISO-New England, and NYISO, due to the limited amount of capacity purchased. Having said that, capacity auction prices in MISO have increased somewhat in recent years as many large coal-fired generation facilities have been retired, reducing the total supply of generation capacity in MISO and increasing the need for utilities to purchase capacity in the auction instead of simply relying on the generation they own.
There is one last thing that all RTO capacity markets have in common. No one really likes them. Generation owners think that the capacity prices have been too low. To support this contention, they point to the large number of large coal generation facilities that have been shut down in recent years for economic reasons, and also to the retirement of a few nuclear units for the same reason. State regulators and customers complain about substantial increases in capacity auction prices as recent capacity auctions in both PJM and MISO have resulted in historically high capacity prices. Energy market purists think that capacity markets are not needed and distort the price signals that result in the most efficient generation mix. And almost everyone thinks RTO capacity market rules are too complicated.
What do I think? In my view, RTO capacity markets are far from perfect. But they are necessary to ensure that enough energy is available in high stress situations where demand is high; the recent debacles in ERCOT demonstrate that an energy only market cannot do this. And RTO capacity markets also are necessary to ensure that vital dispatchable facilities—i.e., generation facilities that can be turned on when necessary and do not depend on an intermittent fuel source such as sunlight or wind—earn the necessary revenues to be kept in service as RTO energy market prices remain low. So, yes, I think that RTO capacity markets continue to be necessary. And since I am writing this post, I get the last word.
I hope you enjoyed this post. I enjoy writing them, and will never charge for subscriptions or ask for donations. I only ask that, if you did enjoy it, you press the “like” button below. Doing so will help me evaluate interest in the book I am writing on grid operations. Of course, if you have a reaction to, or question about, this post, please leave a comment and I will be happy to respond.



A very interesting read for a shrinking planet.