Bill Hogan, Unbundled


A candid commentary on current topics in electric restructuring.

Fortnightly Magazine - November 2012

Energy planners and utility owners are hearing rumblings about tsunamis of change in the electricity market, triggered by earthquakes of uncertainty. Where does the uncertainty arise? The growth of renewables, the emergence of regional electricity markets, innovations in energy technology, changes in distribution systems with smart metering, demand management, and global economic upheavals all contribute to this tectonic shifting.

The lack of a clear national energy policy hasn’t helped to add clarity, as Fortnightly Editor-in-Chief Michael Burr pointed out in a recent “Frontlines” column (see “Mitt Romney and You,” September 2012). The Federal Energy Regulatory Commission (FERC) has contributed to uncertainty with controversial regulations that some view as confiscatory. While investors try to preserve and protect traditional energy sources and maintain profitability, they oppose plans for investment in new renewable innovations.

Since technology developments will dictate the future, resistance to change isn’t productive. Yet, keeping electricity on all the time is much more complex with a stream of less-reliable renewables.

In the ever-changing world of energy markets there are still several constants. One is the consistently reliable counsel of William W. Hogan, the chief architect of wholesale electric market design in the United States. His thoughts and models have shaped energy policy for decades. Currently the Raymond Plank Professor of Energy Policy at the John F. Kennedy School of Government, Hogan recently spoke with Fortnightly about electric market reform and its effect on electric industry restructuring. He discussed the successes and challenges of deregulation, market models, and the problems encountered on the path toward organized wholesale power markets.

Of course, this path hasn’t been altogether smooth. Along the way, a number of different industry sectors and institutional actors have questioned whether Hogan’s ideas on electric market reform have aided overall industry efficiency or produced notable consumer benefits. The American Public Power Association, for example, has for some time now maintained a section on its website ( that offers analytical studies by industry or academic economists that seek to answer the question of whether ratepayers are better off now—what with electric restructuring, regional grid operators, and centralized wholesale power markets—than they would have been under a parallel, but different universe.

Anyone with more than a passing interest in these subjects should take a look at the APPA site, starting with the APPA’s own Electric Market Reform Initiative (EMRI), and APPA’s web link to various investigative studies of the nation’s restructured RTO-operated wholesale electricity markets. The student also should note the APPA’s own proprietary study: APPA’s Competitive Market Plan: A Roadmap for Reforming Wholesale Electricity Markets (2011 Update).

In that context, Hogan started by addressing some of the ideas he presented in his article, “Electricity Market Reform: APPA’s Journey Down the Wrong Path,” co-authored with John Chandley in 2009.

Fortnightly: What are the two or three biggest successes of deregulation, after a decade of experience, and the two or three biggest challenges still facing deregulation?

William Hogan: This is addressed in the article I wrote with John Chandley in 2009. There are three arguments in there: What’s wrong with what the APPA is suggesting? Do you want to have open access and competitive markets or not? And do you want to have regulation or deregulation?

As an example of the success of deregulation, look at what happened to nuclear power. You have nuclear power plants, where the average capacity factor was in the low 60s. Now it’s in the 90s. That was driven by the competitive pressures of privatization in some sense—the investors had to make a profit from it. The nuclear plants weren’t living off the cost-plus world that was driving the cost up and performance down. That’s been a terrific example of the benefit of deregulation.

Fortnightly: That’s a good example. It’s the equivalent of a 50 percent increase in capacity, without building a single plant. What about other examples of investments in an era of deregulation?

Hogan: You have to take a look at some anecdotal information, like the famous graphic when AEP joined PJM, and what happened to their inter-regional trading. (See page 34 of “Electricity Market Reform,” 2009).

You have to think about this. Here’s AEP, one of the most sophisticated utilities in the country, extremely confident about its ability to trade in markets and capture all the economic benefits. They didn’t think the regional transmission organizations [RTO] were important. And then they joined PJM. Their trading in exports [from] Midwest to East went up dramatically when they joined PJM and got access to efficient dispatch.

The picture is stunning when you look at it. You don’t have to be an econometrician to figure this one out.

Fortnightly: What are the implications of deregulation on investment practices and behavior? Who wins and who loses?

Hogan: There are examples of things that aren’t very pleasant, but which are also part of the design. When we had the dash to gas, and were building all these [gas-fired] plants. Investors overestimated how much they were going to make, and that led to a big round of bankruptcies among new independent power producers. Well, in the old days, this would’ve been a stranded-asset problem, and the utility would’ve gotten recovery of the cost for all the things it had built. All of this would have been passed onto the consumer, and there would have been a great amount of regulatory proceedings. They would’ve said the utility “wasn’t prudent” and “should have known”—all this kind of stuff. None of that happened, right? They just went bankrupt. And their shareholders hated it. And that was the idea!

Most importantly—and this is the dog that didn’t bark—they stopped investing in that stuff—unlike [regulated utilities at] Shoreham. Think about the Shoreham (New York) nuclear power plant, where at about every $1 billion [of investment] they stopped and did an assessment. After they got to $6 billion, the governor intervened, the work was stopped. And all the ultimate generation was zero megawatt hours. So $6 billion divided by zero megawatt hours is the price to beat.

Fortnightly: When and how can you direct your investments to ensure long-term benefits?

Hogan: The truth of the matter is that the real benefits are going to be measured in long-term investment profiles. Even now it’s a little early to tell how this is going to unfold. We have all the complications of the renewable subsidies and things like that obfuscating what’s going on here. Think about, for example, the green agenda. If you want innovation, you have to get the incentives and the structure right. Vertically integrated monopolies are not the way to get that done.

I think there’s a case to be made that the benefits of renewables are positive and that they’re growing.

Fortnightly: In California, we saw a complete failure in the management of energy markets. Why the collapse? And how do you organize the market correctly?

Hogan: Now my explanation about California is that this was caused by bad regulations, not deregulation, and the principal problems were due to the interventions by the governor and by the Federal Energy Regulatory Commission. The governor lost his job, appropriately. It wasn’t caused by markets, but, nonetheless, that was a real catastrophe. So [what caused the crisis] is an interesting question. But whether or not you want to have open competition is another.

There’s a third question, though, which APPA proposed (see “Energy Market,” 2009), and that is, “Do you want to have regulation or deregulation?” I think this is a misleading way of asking whether you want to “restructure” deregulation. There’s a lot of regulation [in competitive markets]. The question that people get confused about all the time, and which is extremely important, is: “If you want a competitive open access market without discrimination, what is the appropriate organization of that market? What is the appropriate market design?”

It’s extremely important that you get that right, and it’s evident there’s only one way to do it. If you look at the cost-benefit analysis, comparing doing it the wrong way, like California, versus doing it the right way, like the current version of PJM, for example, then you find enormous differences in approach to the appropriate market design. And so the market design question is a much easier question to answer. It’s clear that it’s important. It’s clear that if you don’t get it right you can cause a lot of trouble. And it’s also clear what the answer is.

PUF Books - New Regulatory Finance

Fortnightly: So, what is the answer? What is the appropriate market design for the electrical energy market?

Hogan: We’re now in a situation where the core feature of the market design is a “mid-base, security constrained, economic dispatch with locational prices.” This I gave as a name to the model.

That model emerged after some experimentation—trying other ways that were supposedly simpler or easier. It turns out there aren’t any other ways that actually work. There are a lot of ways that look simpler on paper, but they don’t actually work because they’re inconsistent with reality.

Every organized market in the United States now uses that model. New England started out with a different approach, and they now use it. New York has always used it. PJM started out a different way and they now use it. SPP [Southwest Power Pool] tried something else, that didn’t work, and it’s now using it. California tried something else and it didn’t work, and they’re using it. Texas tried something different, and it didn’t work and they’re now using it. So that’s a pretty strong empirical record.

The core principle is that you’re accepting the bids and the offers that people make from different generators. You’re doing security constraints, so that reliability constraints are imposed. You do economic dispatch. So, subject to those constraints, you find the least-bid cost based on the bids of the lowest-cost point of operating the system. And then you price locationally at the nodes, which is where the difference in demand is.

You don’t try to aggregate those nodes when figuring out what these prices are. You aggregate them for billing purposes later on. When you’re actually doing the model, and you’re charging generators and paying everybody and you charge for transmission between locations, well, that’s the difference in price. That’s the basic core model, and everybody does that.

Fortnightly: Are there variations on the core model that RTOs use?

Hogan: Now they do different things, and there can be variations in certain circumstances:

  • When they’re dealing with financial transmission rights;
  • When they’re aggregating prices for customers, the average, and so forth;
  • When they have different ways of dealing with the capacity markets, which I didn’t mention because they’re not part of the design;
  • When they do different things for transmission rights, and;
  • When they do different things for transmission cost allocation.

Those are all important issues, and some are good and some are bad. But the core—which is a big idea and extremely important—is that there’s basically only one way to do it.

Fortnightly: That sounds simple, but everyone knows it isn’t. What do you see as the details that need attention?

Hogan: The model I just described exists in all the organized markets. It doesn’t exist outside the organized markets, the ISOs and RTOs. If open access and non-discrimination is supposedly our [national] policy, then every place that hasn’t adopted this isn’t achieving open access and non-discrimination.

So extending the model of the ISO or RTO to the rest of the country should be a high priority for FERC. They’re long overdue in dealing with this application. They’ve had opportunities to do it, but there are political reasons they don’t want to pick up that rock. And I understand those political reasons. They’re very powerful, and there’s a lot of pressure. But in the end you can’t have it both ways. If you say “We have open access and non-discrimination as our policy in the country and we have a market design that’s necessary if we’re going to do that,” you can’t also say “If we don’t have the market design in the country, then everything is OK.” Both can’t be true.

It’s not actually hard to implement technically. It may be that the political problems are overwhelming, but then [FERC] should stand up and say: “We can’t implement open access here because of these political obstacles.” But don’t pretend that you’re doing otherwise. There’s a lot of mumbling these days.

Editor’s Note: While FERC has chosen not to mandate RTO formation across the country, it has long promised to develop industry performance metrics for non-RTO areas, to complement the performance metrics its staff adopted for RTOs in 2010. And on October 15, as this issue was going to press, the FERC staff in fact released a report describing a final set of 39 non-RTO performance metrics (Docket AD12-8-000, “Performance Metrics in Regions Outside ISOs and RTOs”). A possible next step, the report states, will be to “establish common metrics between ISOs/RTOs and non-ISO/RTO regions.” The implication is clear: at some future date FERC may use these common metrics to draw direct comparisons between RTOs and non-RTO regions—on operational efficiency, as well as price.–Ed.

Fortnightly: Has FERC moved in any way to adopt the core model, in spite of political obstacles?

Hogan: FERC had an opportunity several years ago, when they were reviewing their transmission open access policies. John Chandley and I wrote what I thought was a brilliant paper at the time, and submitted it pro bono.

If you want to say you have open access and non-discrimination, one of the most critical services for transmission is balancing. You want to have efficient balancing. And if you have efficient balancing, then that means economic dispatch of the balancing market over several groups. And then you want to price that in a way that’s consistent with economic balancing in locational prices. So you could have a balancing market with economic dispatch, bids, security constraints, and locational prices. Just require that. Don’t say anything else. You have to do that. Then everything else follows.

And that actually is what SPP is [doing]. So it works just fine. FERC hasn’t done this, but they could do that anywhere if they had any spine.

Fortnightly: Where should FERC be focusing its efforts, or what initiative should they follow to help the markets, if political considerations prevent them from imposing market design?

Hogan: There’s a series of possible initiatives with various degrees of urgency. One of them falls under the general heading of market manipulation. There are a lot of enforcement actions that have been starting to come out of the enforcement office of FERC with interpretations of the rules that are new, and quite threatening to markets. This is a very serious problem. Here’s a situation where basically nobody knows what the rules are. FERC enforcers are making them up as they go along.

I don’t think the Commission knows this, and I don’t think they’re getting told the straight story. I think the people in the enforcement office don’t know what they’re doing, and, in many ways, don’t care. So, for example, I believe they think that if they had a strong enforcement case, even though it would completely cripple the operation of competitive markets, that would be OK. They don’t think that’s in their remit. It’s not their problem.

It’s a serious situation. With a court proceeding, all information could become public, you could begin making these arguments, and you could expose the nonsense that’s going on. And then maybe we’ll get some reaction from the courts and FERC. But this is actually a time bomb that’s out there, and I would worry about that.

Fortnightly: What are some of the other problems you see FERC reluctant to address?

Hogan: The second issue, which is very public, and which you can read about in the August 2012 issue of Public Utilities Fortnightly, addresses the payload and demand response story. There’s a great piece by Bruce Radford (“Making Demand More Dynamic”) that summarizes the debate. I agree with what he said, at the same time that I don’t agree with what Audrey Zibelman said in the same issue (“Load as a Resource”). But this is a continuing problem. It’s a mess. And it’s also beginning to become entangled in the enforcement actions. People who are following the incentives of the demand response program are being brought up for enforcement actions because they’re doing things that are uneconomic. And you say “No kidding!”

Another problem is scarcity pricing. We’ve had this festering problem for years and years. The prices in the energy market are too low, even though people are always complaining about prices being too high. You can go through the analysis, and see that they’re too low. This is the “missing money” problem, as it’s called.

There’s a list of 25 things that system operators do that cause prices to get depressed. Paul Joskow wrote a nice paper about this a few years ago. [See “Capacity Payments in Imperfect Electricity Markets: Need and Design,” MIT Department of Economics, Dec. 5, 2007.] The shortlist includes things like price caps. They bring out emergency reserves, which are real expensive, versus the price; they do demand response which is real expensive, and that depresses the price. So there are all sorts of reasons the prices are too low and the system is tight. They’re trying to fix it. Scarcity pricing has been, I think, a critical item for a long time, and I’ve been teaching about it. I keep going down to make speeches at FERC. They keep saying “You’re right in theory.” But in practice, this isn’t high on their list, and they say they’ll do it later. But I want to tell them “Now is later.”

Fortnightly: What’s your view of FERC’s approach to transmission cost allocation?

Hogan: FERC wants to build more transmission, particularly to get [access to] renewables. They want to socialize the cost, which would make their life easier. They end up with situations with people who aren’t benefiting or maybe even losing because of the transmission, and have to pay for it. They take them to court, and [the courts] say, “You know you can’t do this. You can’t charge people if they get no benefit.”

So you have to do the cost allocation so that it’s roughly commensurate with the benefits. The courts said “We’re not assigning precision, but you have to have something like this.” And then FERC comes back and says “Of course, that’s what we meant.” And then they issued a notice of proposed rulemaking [NOPR], which took a year. And during that time they kept approving cost socialization.

PUF Books - New Regulatory Finance

At the end of the NOPR, they issued an order that said, “When a beneficiary pays, that’s what we meant.” And now we’re going to ask all of the ISOs to figure out over the next 18 months the rules for how to do this, with absolutely no guidance as to how to do it. When they asked me what I thought when the order was passed, I said, “Well, they didn’t say anything bad.” And then I predicted what was going to happen for the first year of this stakeholder process. They’re going to thrash around, they’re going to try to come up with a rule of thumb and some simple-minded way of doing it that would end up being cost socialization again.

And that’s what FERC wants. They want to claim the rulemaking is one thing, but have it be the other. And then they’re going to wake up as they get toward the end, which is about now, and realize that the rules they’re proposing don’t make any sense, are not logical, are not connected to the benefits, and are not going to stand up in court. And then they’re going to have to adopt something completely different.

I’ve been going around and making speeches about a completely different way to do it that I consider to be completely obvious. Of course they don’t like it because it requires you to actually be clear about what you’re doing. But that problem isn’t going away.

There’s a related problem, uplift allocation, which is a big to-do at the moment. Basically, the question is, how should you treat virtual transactions in the day-ahead market. Should you treat uplift as [allocable to] virtual transactions as opposed to allocating uplift at an actual load? I’m a strong believer that you should allocate it to the actual load. All the arguments about allocating to the virtual transaction are answering the wrong questions. But that’s a more complicated issue.