One of the most egregious aspects of ISO-NE’s capacity payments design is to make capacity payments to wind projects. Wind cannot guarantee performance in the future. Those payments are an outright gift at ratepayers expense.
Meredith, Herman Wouk, in his play "The Caine Mutiny Court Martial", described the Navy as: "a master plan designed by geniuses for execution by idiots."
We appear to be proceeding, without a master plan, to create an electric grid designed by idiots for execution by geniuses. Unfortunately, I expect we will discover that we have a significant shortage of the required geniuses.
Ex nuclear machinist mate here. I've always believed that the navy was amazing in it ability to teach completely unskilled teenagers how to run and service some of the most complex systems in the world.
Good write up Meredith, I remember the moniker "Reliability through Markets", one of the many attempts to control the grid by CAISO. It was nearly always a train wreck.
Thanks for your continued diligent work with pick and shovel (and costly tomes)! A great addition to our understanding. (Wonderful how armchair geniuses can be so certain that they're just the ones who can redesign the world.) Significantly missing from this kind of abstraction is a basic understanding of the physical reality and engineering demands of something as complex as a reliable electrical grid.
Good article Meredith. It reminds me of the difference between planned communities and those grow organically. Milton Keynes in the UK was a joke for many years for things such as concrete cows. It has emerged as something more of a success in large part because of some of the smart urban planning elements such as the grid. Putrajaya in Malaysia is another example of a planned city. I guess some plans are better than others but “market design” is bit of an oxymoron.
The idea of designing markets using theoretical models reminds me of Elinor Ostrom’s classic, Governing the Commons. Here’s a quote:
“The intellectual trap in relying entirely on models to provide the foundation for policy analysis is that scholars then presume that they are omniscient observers able to comprehend the essentials of how complex, dynamic systems work by creating stylized descriptions of some aspects of those systems.”
Market design is wholly compatible with Ostrom’s insights—all existing RTOs/ISOs have evolved over time in response to experience, rules and rule changes are often driven by users who simply want the market to do its job.
There are problems to be sure, but the core market design is security constrained bid based economic dispatch as developed by engineers and economists in the 80s and 90s. Capacity markets have been a constant problem, but they are not part of the core energy market design.
As one of the people who made a recent post disagreeing with Meredith's take on RTO markets in her book (although I like her book), I thought I would provide a brief disagreement with this post as well.
Markets have always had rules, whether by custom or law. For example, the ancient Romans had rules about the weights and measures used in markets, much like the street market in Meredith's picture, that were intended to prevent fraud. These rules constitute the markets' design, even if not created by market designers.
For many years, the market design in the electric industry was cost-based sales, with the costs reviewed by the local regulator or by FERC, depending on the nature of the transaction. These rules benefitted the many lawyers and consultants who litigated the rate cases needed to determine the cost-based rate, some of which droned on for several months. Did the design benefit ratepayers? Not always. The rates customers paid always included the costs of the rate litigation, and in the late 1970s cost-based regulation resulted in high rates, including the costs of large nuclear and other generation facilities whose construction was abandoned because it turned out the plants were not needed. It was because the cost-based rate market design was seen as failing customers that the industry moved to one that is characterized as "market-based," although the markets continue to be highly regulated.
Which gets us to the markets run by RTOs. The mere fact that those markets have a market design does not tell us whether their markets are good or bad. Rather, you have to evaluate the design of the market. I don't know anything about the FCC market that Meredith describes, but it seems like that was a bad market design. As I explain in my post on my Substack (Explaining the Grid), the RTO markets are designed to set prices at the marginal cost of producing energy or capacity, which is what classic economic theory posits is the most efficient market price, regardless of the technology involved. I happen to think that is a good goal, others are free to disagree. The RTO markets are struggling with how to deal with highly subsidized wind and solar generation. No good solutions have yet been found, and for a while at least it was political poison to even try. So that is a problem with RTO market design.
But if we want to abandon the RTO markets, then we will have to replace them with other markets with their own designs. They won't necessarily be better or worse. It will depend on the design.
Matt, I appreciate your detailed posts about RTOs. I don't agree with everything in the posts, but I think that conversations about RTOs and utility governance are important. For a long time, it was just "I love solar" or "I hate renewables" or "I love nuclear" or "I hate nuclear." Nothing about how the grid itself was managed!
Thanks Meredith. I enjoyed your post, and all of your posts. I always learn something from them, even when I don't agree with your ultimate conclusions. Having these conversations, and thinking about issues anew when someone makes a point you don't agree with is the only way to increase your understanding of the issues.
Do you view market design being the same thing as market regulation? Game theory suggests that a free market would result in market power concentration that would be unfavourable to most citizens. Market regulation is intended to limit unfavourable results. What more does market design achieve?
In my view, market design constitutes the rules governing how a market works. These rules could be imposed by some combination of rules developed by market participants, rules imposed by laws, and rules imposed by regulators. In the case of the RTO markets, there are rules designed to prevent the exercise of market power, as well as manipulation of the markets, and also regulatory oversight that involves reviewing transactions where market power or manipulation may have occurred. Market design also determines the overall structure of the market; for example you could design a farmer's market type of market where merchants gather at one location and compete for customers who come to the market; or a bilateral market where people enter into bilateral transactions with sellers; or a regulated cost-based market where a monopoly utility sells electricity to captive customers at regulated cost-based rates. The structure of the markets that were designed for RTOs involves large regional auctions where all generators who want to sell energy or capacity must submit offers to sell into the market and all entities that supply electricity to retail customers must purchase electricity or capacity at the market-clearing price. I have a more detailed discussion of markets and other aspects of RTOs and grid operations in general at my Substack--Explaining the Grid
I think market systems can be designed well but I feel that electricity has proven not to be a service that it works for. As in your book prices have not fallen in RTO areas, but actually risen a bit relatively. More importantly, even if a market can efficiently deliver a more ideal system eventually, I think that Texas has shown that the consequences of learning how to do it can easily outweigh any potential benefit
Winter Storm Uri cost Texas $200 billion dollars and over a thousand deaths. I believe recent estimates of the "levelized cost of blackouts" are around $30,000/MWh. It's totally impossible for individual market participants to pay these prices so the tragedy of the commons, which markets are supposed to avoid, means that you have to use special rules to try to get the reserve capacity and reliability built into the system that the market can't
These costs are so high that it calls into question the potential benefits of a leaner market oriented system. After reintroducing more reliability and reserve into your system after finding out that it was too far cut to the bone, your theoretical payback time for the costs of introducing the system and experiencing failures must be tremendously long
Estimates of Winter Storm Uri costs in Texas do range from $100- to $200 billion, but I have not seen any reports of over a thousand deaths from the storm. The official state count of storm related deaths was 246, and an independent researcher estimated deaths could have been around 700 (plus or minus a margin of error from the statistical approach he was using).
But in this context note that both of these quantities include many deaths due to the storm but not due to the rolling blackouts or other ERCOT related failures. For example, the reported amounts include traffic accidents from before the rolling outages started and they include deaths in Texas but outside of part of the state served by ERCOT.
The state report does not disclose enough information to identify likely electric outage related deaths, but the amount would likely fall to no more than 150 and perhaps smaller. Obviously, still a major tragedy, and by trying to be accurate in claims in no way diminishes the tragedy that resulted from the energy system failures in Texas.
Constellation has sold the output of TMI1 (830GW) for 20 years to assorted Microsoft data centers in the PJM market. How in the RTO-age can they do that? Well, at least I know the right person to ask: you. Email me at amnnr@aol.com if you need more questions. Thanks in advance.y
"Market design" is not an oxymoron, it it a subdiscipline of economics that studies how market rules affect market performance and which can be applied when choosing or revising the rules that apply to the market.
Since you used an image of a farmers' market in the post, I'll use a simple flea market example. Some flea market organizers only charge vendors and let potential consumers roam the venue for free. Other flea markets charge vendors for display space and charge potential customers an entrance fee. A basic design choice must be consciously made by the organizers, but the flea market is still a market.
Market design economics offers insights into, for example, why some auctions use rising prices (art auctions) and others use declining prices (flowers at the Aalsmeer flower auction). These insights concern choices made about how a market operates, but the auctions are still markets.
Market design also, obviously, plays a big role in RTO markets. If you dismiss market design as an oxymoron and assert that RTOs do not have markets simply because the rules of the market where chosen, then you also have to say the stock market is not a market, that futures markets are not markets, that Aalseem flower auction in the Netherlands is not a market, etc.
In fact, you'd probably have to declare the New York Farmers Markets that you used a picture of as "not a real market" because of the heavy government involvement in their development. See https://en.wikipedia.org/wiki/Farmers'_markets_in_New_York_City.
On your more general point of "market washing," I think the validity of your claim depends on just which remark you respond to. Clearly any general statement of "It’s a market, so it must be good!" is wrong. The original CAISO market design including the separate Cal PX were markets, but they were flawed (as was revealed under the stresses of the year 2000). Note the problem in that case is that they ignored the position of Wm. Hogan and other expert power market designers and instead tried to go their own way. Declaring market design an oxymoron would leave one unable to diagnose the principle failures of the original CAISO design, and similarly leave one unable to understand why the 2006 "MRTU" reforms of CAISO's market succeeded.
Market regulation is useful and limited. It is usually constrained to simple rules: safety of products sold, hours the market is open, etc.
Market DESIGN is a more far-reaching plan, meant to encourage "salutary" outcomes, whatever the definition of salutary may be at the time.
Your comment seems a bit circular to me: we can't fix market design if we don't acknowledge market design as a real thing. Well....yes....I guess so. But I don't think we can fix market design.
We should stop designing markets for specific outcomes, and go back to market regulation instead.
If you look at the quote in my main article about how society will fix its problems with market design (often with patented market designs!), you would see that the end point of market design would probably be market-design-for-world-peace.
Limited amounts of market regulation would be a better path to take.
One of the most egregious aspects of ISO-NE’s capacity payments design is to make capacity payments to wind projects. Wind cannot guarantee performance in the future. Those payments are an outright gift at ratepayers expense.
Meredith, Herman Wouk, in his play "The Caine Mutiny Court Martial", described the Navy as: "a master plan designed by geniuses for execution by idiots."
We appear to be proceeding, without a master plan, to create an electric grid designed by idiots for execution by geniuses. Unfortunately, I expect we will discover that we have a significant shortage of the required geniuses.
My husband and I both got a good chuckle out of this!
Blue Jacket’s Manual forever!
Ex nuclear machinist mate here. I've always believed that the navy was amazing in it ability to teach completely unskilled teenagers how to run and service some of the most complex systems in the world.
Like a nuclear reactor.
And all done with sliderules
I had some thoughts on slide rules and GPS a few years ago
https://youtu.be/ePlitQNMTYo?si=Ea5FgAchaaUnTUg9
Me too from 68 nuclear power school.
I tried to give it to my daughter when she went off to chemEng school in 92.
Nope. She made me buy her a $200 TI calculator
Tried to give it to my granddaughter 3 years ago when she went off to chemEng school.
Nope. I had to buy her a $700 iPad.
I still have my K&E in its leather case with the belt loop. ;-)
Love this comment - good one!
Good write up Meredith, I remember the moniker "Reliability through Markets", one of the many attempts to control the grid by CAISO. It was nearly always a train wreck.
Thanks for your continued diligent work with pick and shovel (and costly tomes)! A great addition to our understanding. (Wonderful how armchair geniuses can be so certain that they're just the ones who can redesign the world.) Significantly missing from this kind of abstraction is a basic understanding of the physical reality and engineering demands of something as complex as a reliable electrical grid.
Good article Meredith. It reminds me of the difference between planned communities and those grow organically. Milton Keynes in the UK was a joke for many years for things such as concrete cows. It has emerged as something more of a success in large part because of some of the smart urban planning elements such as the grid. Putrajaya in Malaysia is another example of a planned city. I guess some plans are better than others but “market design” is bit of an oxymoron.
Thank you for the Mirowski/Nik-Khan book reference. Reading now.
Great to see you discussing mirowski in relation to RTOs!
The idea of designing markets using theoretical models reminds me of Elinor Ostrom’s classic, Governing the Commons. Here’s a quote:
“The intellectual trap in relying entirely on models to provide the foundation for policy analysis is that scholars then presume that they are omniscient observers able to comprehend the essentials of how complex, dynamic systems work by creating stylized descriptions of some aspects of those systems.”
Market design is wholly compatible with Ostrom’s insights—all existing RTOs/ISOs have evolved over time in response to experience, rules and rule changes are often driven by users who simply want the market to do its job.
There are problems to be sure, but the core market design is security constrained bid based economic dispatch as developed by engineers and economists in the 80s and 90s. Capacity markets have been a constant problem, but they are not part of the core energy market design.
As one of the people who made a recent post disagreeing with Meredith's take on RTO markets in her book (although I like her book), I thought I would provide a brief disagreement with this post as well.
Markets have always had rules, whether by custom or law. For example, the ancient Romans had rules about the weights and measures used in markets, much like the street market in Meredith's picture, that were intended to prevent fraud. These rules constitute the markets' design, even if not created by market designers.
For many years, the market design in the electric industry was cost-based sales, with the costs reviewed by the local regulator or by FERC, depending on the nature of the transaction. These rules benefitted the many lawyers and consultants who litigated the rate cases needed to determine the cost-based rate, some of which droned on for several months. Did the design benefit ratepayers? Not always. The rates customers paid always included the costs of the rate litigation, and in the late 1970s cost-based regulation resulted in high rates, including the costs of large nuclear and other generation facilities whose construction was abandoned because it turned out the plants were not needed. It was because the cost-based rate market design was seen as failing customers that the industry moved to one that is characterized as "market-based," although the markets continue to be highly regulated.
Which gets us to the markets run by RTOs. The mere fact that those markets have a market design does not tell us whether their markets are good or bad. Rather, you have to evaluate the design of the market. I don't know anything about the FCC market that Meredith describes, but it seems like that was a bad market design. As I explain in my post on my Substack (Explaining the Grid), the RTO markets are designed to set prices at the marginal cost of producing energy or capacity, which is what classic economic theory posits is the most efficient market price, regardless of the technology involved. I happen to think that is a good goal, others are free to disagree. The RTO markets are struggling with how to deal with highly subsidized wind and solar generation. No good solutions have yet been found, and for a while at least it was political poison to even try. So that is a problem with RTO market design.
But if we want to abandon the RTO markets, then we will have to replace them with other markets with their own designs. They won't necessarily be better or worse. It will depend on the design.
Matt, I appreciate your detailed posts about RTOs. I don't agree with everything in the posts, but I think that conversations about RTOs and utility governance are important. For a long time, it was just "I love solar" or "I hate renewables" or "I love nuclear" or "I hate nuclear." Nothing about how the grid itself was managed!
Your substack is a useful part of the conversation. https://substack.com/@matt1958
Thanks Meredith. I enjoyed your post, and all of your posts. I always learn something from them, even when I don't agree with your ultimate conclusions. Having these conversations, and thinking about issues anew when someone makes a point you don't agree with is the only way to increase your understanding of the issues.
Do you view market design being the same thing as market regulation? Game theory suggests that a free market would result in market power concentration that would be unfavourable to most citizens. Market regulation is intended to limit unfavourable results. What more does market design achieve?
In my view, market design constitutes the rules governing how a market works. These rules could be imposed by some combination of rules developed by market participants, rules imposed by laws, and rules imposed by regulators. In the case of the RTO markets, there are rules designed to prevent the exercise of market power, as well as manipulation of the markets, and also regulatory oversight that involves reviewing transactions where market power or manipulation may have occurred. Market design also determines the overall structure of the market; for example you could design a farmer's market type of market where merchants gather at one location and compete for customers who come to the market; or a bilateral market where people enter into bilateral transactions with sellers; or a regulated cost-based market where a monopoly utility sells electricity to captive customers at regulated cost-based rates. The structure of the markets that were designed for RTOs involves large regional auctions where all generators who want to sell energy or capacity must submit offers to sell into the market and all entities that supply electricity to retail customers must purchase electricity or capacity at the market-clearing price. I have a more detailed discussion of markets and other aspects of RTOs and grid operations in general at my Substack--Explaining the Grid
I think market systems can be designed well but I feel that electricity has proven not to be a service that it works for. As in your book prices have not fallen in RTO areas, but actually risen a bit relatively. More importantly, even if a market can efficiently deliver a more ideal system eventually, I think that Texas has shown that the consequences of learning how to do it can easily outweigh any potential benefit
Winter Storm Uri cost Texas $200 billion dollars and over a thousand deaths. I believe recent estimates of the "levelized cost of blackouts" are around $30,000/MWh. It's totally impossible for individual market participants to pay these prices so the tragedy of the commons, which markets are supposed to avoid, means that you have to use special rules to try to get the reserve capacity and reliability built into the system that the market can't
These costs are so high that it calls into question the potential benefits of a leaner market oriented system. After reintroducing more reliability and reserve into your system after finding out that it was too far cut to the bone, your theoretical payback time for the costs of introducing the system and experiencing failures must be tremendously long
Estimates of Winter Storm Uri costs in Texas do range from $100- to $200 billion, but I have not seen any reports of over a thousand deaths from the storm. The official state count of storm related deaths was 246, and an independent researcher estimated deaths could have been around 700 (plus or minus a margin of error from the statistical approach he was using).
But in this context note that both of these quantities include many deaths due to the storm but not due to the rolling blackouts or other ERCOT related failures. For example, the reported amounts include traffic accidents from before the rolling outages started and they include deaths in Texas but outside of part of the state served by ERCOT.
The state report does not disclose enough information to identify likely electric outage related deaths, but the amount would likely fall to no more than 150 and perhaps smaller. Obviously, still a major tragedy, and by trying to be accurate in claims in no way diminishes the tragedy that resulted from the energy system failures in Texas.
This is a fine debunking of a pernicious myth, Meredith. Well done.
Constellation has sold the output of TMI1 (830GW) for 20 years to assorted Microsoft data centers in the PJM market. How in the RTO-age can they do that? Well, at least I know the right person to ask: you. Email me at amnnr@aol.com if you need more questions. Thanks in advance.y
"Market design" is not an oxymoron, it it a subdiscipline of economics that studies how market rules affect market performance and which can be applied when choosing or revising the rules that apply to the market.
Since you used an image of a farmers' market in the post, I'll use a simple flea market example. Some flea market organizers only charge vendors and let potential consumers roam the venue for free. Other flea markets charge vendors for display space and charge potential customers an entrance fee. A basic design choice must be consciously made by the organizers, but the flea market is still a market.
Market design economics offers insights into, for example, why some auctions use rising prices (art auctions) and others use declining prices (flowers at the Aalsmeer flower auction). These insights concern choices made about how a market operates, but the auctions are still markets.
Market design also, obviously, plays a big role in RTO markets. If you dismiss market design as an oxymoron and assert that RTOs do not have markets simply because the rules of the market where chosen, then you also have to say the stock market is not a market, that futures markets are not markets, that Aalseem flower auction in the Netherlands is not a market, etc.
In fact, you'd probably have to declare the New York Farmers Markets that you used a picture of as "not a real market" because of the heavy government involvement in their development. See https://en.wikipedia.org/wiki/Farmers'_markets_in_New_York_City.
On your more general point of "market washing," I think the validity of your claim depends on just which remark you respond to. Clearly any general statement of "It’s a market, so it must be good!" is wrong. The original CAISO market design including the separate Cal PX were markets, but they were flawed (as was revealed under the stresses of the year 2000). Note the problem in that case is that they ignored the position of Wm. Hogan and other expert power market designers and instead tried to go their own way. Declaring market design an oxymoron would leave one unable to diagnose the principle failures of the original CAISO design, and similarly leave one unable to understand why the 2006 "MRTU" reforms of CAISO's market succeeded.
Thank you for your comments!
Market regulation is useful and limited. It is usually constrained to simple rules: safety of products sold, hours the market is open, etc.
Market DESIGN is a more far-reaching plan, meant to encourage "salutary" outcomes, whatever the definition of salutary may be at the time.
Your comment seems a bit circular to me: we can't fix market design if we don't acknowledge market design as a real thing. Well....yes....I guess so. But I don't think we can fix market design.
We should stop designing markets for specific outcomes, and go back to market regulation instead.
If you look at the quote in my main article about how society will fix its problems with market design (often with patented market designs!), you would see that the end point of market design would probably be market-design-for-world-peace.
Limited amounts of market regulation would be a better path to take.
Great article - once again you have raised the bar for us all. Much more reading necessary to follow up on this one - thanks!
Wonderful piece, thank you. There is so much that goes into where we are as a society about what has brought us to this difficult place.
Brava!