Posts Tagged ‘power market design’

h1

United States v. KeySpan Corporation antitrust case settles for paltry $12 million

February 2, 2011

Michael Giberson

The Justice Department of the United States has agreed to a $12 million settlement with KeySpan Corporation on a Sherman antitrust act claim. The allegation was that KeySpan manipulated the New York ISO capacity market price in its part of the state from May 2006 through February 2008, reaping an estimated $49 million in excess revenue. More specifically, the allegation was the KeySpan entered into a contract in restrain of trade. The $12 million settlement agreed to between Justice and KeySpan reflects the estimated excess profits that KeySpan gained by the scheme.

KeySpan held market power in the NYISO capacity market for the New York City and Long Island area, and for years they used their market power to keep the price they were paid for capacity at the highest level the market rules allowed. However, market entry by a new competitor in 2006 threatened their price-maximizing strategy.

In response, KeySpan entered into a contract with Morgan Stanley that gave KeySpan a significant economic interest in the capacity market revenues of the new competitor. Morgan Stanley agreed to the contract with KeySpan only on condition it could engage with another counterparty to offset the risk; the counterparty Morgan Stanley secured turned out to be KeySpan’s competitor. (In fact, the competitor was the only party well suited to the deal Morgan Stanley needed to balance its risk, something that KeySpan knew would be the case.)  With the deal arranged, KeySpan could continue to profit by offering its own capacity at the maximum allowed price, pushing the capacity price to its upper limit. So it did.

Complaints filed with the Federal Energy Regulatory Commission lead to rulings in KeySpan’s favor. FERC concluded that while KeySpan clearly used its market position and financial positions to maximize the capacity price it was paid, KeySpan had not violated NYISO market rules in doing so. A rule that established a maximum offer cap is not violated when a party offers capacity at the allowed maximum, even if the effect is a market price higher than it would be otherwise. FERC further concluded that the company had not violated laws against energy market manipulation.

The Justice Department claimed that KeySpan violated Section 1 of the Sherman Act, namely that the company entered into an agreement in restraint of trade. It seems a somewhat novel application of the Sherman Act, especially if KeySpan’s actions were otherwise in compliance with laws and market rules. That said, KeySpan’s actions deterred competition that would have brought benefits to consumers in the region, and it is a broader purpose of NYISO’s market rules to promote competition in New York’s power market.

$12 million seems like a too-modest remedy.

NOTES:

h1

BPA won’t pay negative prices to get wind power producers to curtail

January 18, 2011

Michael Giberson

At a December 2010 meeting, the federal Bonneville Power Agency announced that it would not pay wind power producers in its area to curtail during overgeneration events that sometimes result from the way the agency manages water flow through hydropower facilities to comply with environmental regulations.

When reservoirs are full, the BPA’s dams can either generate power or spill any excess water. High water conditions common during late spring in the Pacific Northwest sometimes put the BPA up against environmental limits on how much water it can spill, so driving it to want to produce and distribute as much power as possible. (Spilling too much water leads to high concentrations of dissolved gas in the water, a hazard to fish.)  In the past, BPA would essentially give away power in order to maximize power generation, and utilities in the area were happy to take the cheap power and shut down their thermal power plants which were costly to run.

Over the past few years, however, the growth of wind power in the BPA’s area has presented the agency with a new problem. Wind power producers who can obtain from $20 to $40 per MWh in federal and state subsidies while they are producing power don’t want to shut down for nothing. If the BPA wants to curtail them, they’d like to be compensated for their losses. The BPA says it will not pay; in a statement it explains why:

While one possible outcome would be for BPA to compensate wind generators the value of the foregone incentives, BPA does not believe that is an appropriate consequence of actions taken to protect fish. …  Currently, qualifying renewable energy receives PTCs and/or RECs when it generates, and the cost is shared broadly by taxpayers. If BPA were to pay negative prices to comply with ESA and the Clean Water Act during high runoff events, the cost burden would shift and would be narrowly focused on BPA preference customers. We do not think the law was designed to place this cost burden on a narrow class of utility ratepayers, and we are not prepared to initiate this change.

The BPA claims it has sufficient legal authority under existing generator interconnection agreements to implement its new policy of “environmental dispatch,” but to clearly articulate the authority it will unilaterally amend provisions of its standard generation interconnection agreements to reflect the policy.

In areas with RTO/ISO power markets, negative prices are now the conventional way for coordinating resource supplies during periods of potential overgenation (mostly also involving high wind power among other contributing factors).

NOTES:

h1

Designing electricity auctions

December 2, 2010

From the inbox, notice of the new Utilities Policy, this a special issue on designing electricity auctions edited by Thomas P. Tangerås, drawn from a workshop in Stockholm, September 2009 hosted by the Research Institute of Industrial Economics.  Tangerås introduces the topic:

What are the boundaries of the market in a system with strong requirements on centralized management of power flows, production and consumption? To what extent can decisions be decentralized to market participants? Increasing shares of volatile intermittent energy production place additional strain on the transmission system and on alternative production sources to absorb the fluctuations. Are the current balancing markets designed to handle extreme short-term fluctuations? Do wholesale electricity markets provide ample investment signals, or is it necessary to introduce additional capacity markets? If so, how should these markets best be designed?

In Sweden and elsewhere, a perpetual discussion revolves around whether electricity producers make excessive profits at the expense of consumers. Electricity markets are vulnerable to the exercise of short-run market power because demand is price insensitive, and production is concentrated to a small number of firms. How do electricity markets really perform? And if markets are susceptible to the exercise of market power, are there more efficient auction designs which distribute more of the surplus to consumers?

The answers to these questions are far from obvious, as witnessed by the numerous auction designs circulating in the universe of liberalized electricity markets.

List of articles in Utilities Policy, Volume 18, Issue 4, (December 2010):

  • Designing electricity auctions: Introduction and overview by Thomas P. Tangerås
  • Three-part auctions versus self-commitment in day-ahead electricity markets by Ramteen Sioshansi, Shmuel Oren, Richard O’Neill
  • Production inefficiency of electricity markets with hydro generation by Andy Philpott, Ziming Guan, Javad Khazaei, Golbon Zakeri
  • Are the British electricity trading and transmission arrangements future-proof? by Richard Green
  • Using forward markets to improve electricity market design by Lawrence M. Ausubel, Peter Cramton
  • Virtual power plant auctions by Lawrence M. Ausubel, Peter Cramton
  • The supply function equilibrium and its policy implications for wholesale electricity auctions by Pär Holmberg, David Newbery
  • Using restructured electricity supply industries to understand oligopoly industry outcomes by Frank A. Wolak

In related news, Texas power system operator ERCOT flipped the switch on its nodal market design yesterday, and so far the lights remain on.  (Actually, I couldn’t find any news stories this morning reporting that the lights remain on, but since the Houston Chronicle is giving us this story, rather than details of horrifying market failure, I’m inferring that things must be working okay.) Lots of market data is available here.

h1

The ERCOT market in Texas readies for change in market design

November 19, 2010

Michael Giberson

The ERCOT grid is about to switch fundamental market designs from its current zonal market to a nodal market. Here is a clip from an ERCOT document describing the change:

 

ERCOT_Understanding Texas Nodal Market Implementation

Excerpt from ERCOT's "Understanding: Texas Nodal Market Integration"

 

For a more readable version, check out the full brochure, which is evailable from ERCOT’s website. ERCOT’s “Texas Nodal Market Implementation” site has more information.

DallasBlog provides additional discussion in “Lawmakers prepare for electric grid overhaul.” So what’s the point?  The DallasBlog offers the following medical metaphor:

Think of a patient who complains to the doctor of pain in his right leg, but can’t finger the exact location. That’s zonal. Treatment of an entire leg will not be very precise or efficient. It will therefore cost more.

A patient who can say the pain is localized on his rightquadriceps femoris is likely to get more precise and efficient, and therefore cheaper, treatment.

Similarly, the transition from zonal to nodal will allow the ERCOT grid to pinpoint more precisely high energy demand and congestion areas (think “pain”). Therefore “treatment,” i.e., congestion management and transmission construction, will be more precise and efficient, and thus cheaper.

The next line is, “In theory, at least.”

In a sense, “in theory” is right, but we have much more than theory to go on.  Other U.S. RTO/ISO markets have started with zonal congestion management and shifted to nodal, all have resolved problems experienced under the zonal market by the shift, and no one has chosen to shift from nodal to zonal.

In fact, I’m not aware of any significant constituency within any of these markets that has argued for a shift from nodal back to zonal.  Given the diverse and often conflicting interests of the many entities that participate in RTO/ISO markets, this silence speaks loudly in favor of a nodal market design.

h1

Will faking a consumer cartel help make power markets more efficient?

September 16, 2010

Michael Giberson

Does the Federal Energy Regulatory Commission (FERC) really want to go down this path? Do they really think that faking a consumer cartel will help make wholesale power markets work more efficiently?

Consumers come to any market in pretty direct competition with each other. Suppliers are offering their goods and I would like to buy as cheaply as possible, and so would you, and our competition will drive the price to a level higher than either one of us would prefer.  It is obvious to me and my neighbors that it would be easier for us to buy cheaply if you and your neighbors stayed home. In fact, me and my neighbors might save enough from you and your neighbors staying home that we could pay you enough to stay home.  And with a little formal coordination, we would be on our way to creating a buyers’ cartel.

A well organized buyers’ cartel could, for any given set of supply offers and demand levels, figure out the quantity of consumption that maximizes the economic surplus received by consumers in any period. The cartel would have to make side payments to consumers who have their consumption reduced, but by definition their are enough consumers benefiting from the cartel that the side payments could make everyone better off than before (or rather, all consumers better off).  Sound good?

Of course effective cartels lead to inefficient outcomes; they waste resources. The cartel’s problem is that allowing willing consumers and suppliers to pursue all of the otherwise-wasted opportunities will drive prices back up for everyone. But if buyers can be roped into participation and a sufficient scheme of side-payments is enforced, buyers could be winners (at least in the short run).

To my mind, FERC seems to be pursuing a kind of ad hoc and partial cartelization of buyers with its current ideas for encouraging “demand response” participation in markets (FERC Docket RM10-17-000, Demand Response Compensation in Organized Wholesale Energy Markets, see the Supplemental NOPR for the latest and Technical Conference materials here). FERC has proposed that qualified “demand response” resources be able to bid a demand reduction into day-ahead RTO markets, have it treated sort of like a supply offer, and be paid the market price for energy for any demand reduction accepted by the market.

FERC also invited comments on whether a “net benefits test” of some sort is needed – to make sure that a particular demand reduction actually results in benefits for other consumers – and if so, how should the net benefits be calculated. In addition, the issue of cost allocation arises. Ultimately some set of consumers somewhere will be paying the demand response resource for its service of dropping out of the market, and FERC wants a rule that doesn’t accidentally end up making some consumers worse off in any obvious way.

Read enough about these demand response compensation plans and it begins to sound like a set of instructions for turning an energy market into a Rube Goldberg machine.  In one corner of the machine a cap naps too soundly (these are energy consumers), allowing mice (these are the energy suppliers) to get too much cheese. Now comes FERC to assess the situation, and they suggest if we attach a broom handle to the rocking chair which is tied by a string to a teeter-totter that the bowling ball falls onto, then the broom handle can prod that cat at the right moment and the cat will stop mice from getting too much cheese. Clever, maybe, but no way to run an energy market.

Look, Lynne and I are both big advocates for an active and engaged demand side of the market. We’ve said so several times here in the past and occasionally highlighted research that explains the great value that could be created. We believe!

But jury-rigging the market to goose a few consumers into action isn’t the same thing as enabling an active and engaged demand side of the market.

AFTERWORD: This tirade, written late and in haste, surely requires more time and thought. Admittedly, FERC is in a tight spot. Efficient wholesale power markets really do need an engaged demand side, but the demand side is heavily encased in state-jurisdictional retail rate policies, and technically speaking outside of FERC’s reach. FERC is, in essence, trying to overlay some super-incentives for wholesale-level-hence-FERC-jurisdictional  ”demand response” to make up for the bad incentives (inadvertently, but nonetheless) created by most state retail rate policies.  It is these state policies which keep the cats napping, hence the Rube Goldberg attempt at a work-around.  The trouble is that FERC will end up creating perverse incentives, and we will end up with “Demand Response machines” every bit as stupid and wasteful as the the PURPA machines incentivized by an earlier mix of state and federal energy policies.

On a more constructive note, scanning some of the comments presented for the recent Technical Conference, I’d urge FERC carefully consider the comments of Paul Centolella of the PUC of Ohio. Centolella seems spot on in his analysis.

h1

Auctioning power transmission capacity contracts of varying duration

August 12, 2010

Michael Giberson

Suppose you are a merchant transmission line operator with a DC power line linking point A to point B.  Some potential customers prefer to buy transmission capability the day before the power flow, while others would like to buy monthly, annual, or even multi-year contracts.  While some customers want 24-hour delivery, others seek “peak hour” (6 AM to 10 PM) or “off peak” contracts (10 PM to 6 AM).  The economic value of transmission capability depends on the difference in the price of power at points A and B.  While this difference is not completely unpredictable, it does vary a great deal over the course of a day and from season to season.

The transmission line operator’s problem is to determine how it will divide up its transmission capability to sell to customers: how much to long term contracts and how much to shorter duration contracts, how much to 24-hour contracts and how much to peak or off-peak packages, and so on.

Perhaps the market design for the job is a “product mix auction.”  Paul Klemperer describes it as:

My design is straightforward in concept – each bidder can make one or more bids, and each bid contains a set of mutually exclusive offers. Each offer specifies a price (or, in the Bank of England’s auction, an interest rate) for a quantity of a specific “variety”. The auctioneer looks at all the bids and then selects a price for each “variety”. From each set of offers in each bid, the auctioneer accepts the one that gives the bidder the greatest surplus evaluated at the selected prices or no offer if all the offers would give the bidder negative surplus. All accepted offers for a variety pay the same (uniform) price for that variety.

The idea is that the menu of mutually exclusive bids allows each bidder to approximate a demand function, so bidders can, in effect, decide how much of each variety to buy after seeing the prices chosen. Meanwhile the auctioneer can look at demand before choosing the prices. (Allowing the auctioneer to choose the prices ex post creates no problem here because it allocates to each bidder precisely what that bidder would have chosen given those prices in the environments for which the auction is proposed.) Importantly, offers for each variety provide a competitive discipline on the offers for the other varieties, because they are all being auctioned simultaneously….

The product-mix auction yields better “matching” between suppliers and demanders, reduced market power, greater volume and liquidity, and therefore also improved efficiency, revenue, and quality of information than feasible alternatives. Its potential applications therefore extend well beyond the financial context.

For more information see Klemperer, Paul (2009). “The Product-Mix Auction: A New Auction Design for Differentiated Goods”.

h1

How integrated are European electricity markets?

June 15, 2010

Michael Giberson

Georg Zachmann, writing at the EU Energy Policy Blog, asks, “How integrated are European electricity markets?”  At least in the case of the German market, the answer seems to be not so much.

(For comparison, here is a paper by John Bower that seeks to assess power market integration in Europe as of the end of 2001.  Bower concluded that EU competition authorities should focus more on breaking up national power monopolies and less on building bigger transmission links between countries.)

h1

More on efficient trade between power markets

May 12, 2010

Michael Giberson

A paper by Giorgia Oggioni and Yves Smeers, “Degree of coordination in market-coupling and counter-trading,” examines the value of improving coordination between separate-but-interconnected power markets. (A post here last week cited a recent Windpower Monthly article that provides a good non-technical discussion of the issue. If you are not familiar with market coupling, I recommend you first read the Windpower Monthly article linked to in the earlier post. The Oggioni and Smeers paper provides a more technical discussion.)

In brief, Oggioni and Smeers compare market coupling regimes to both an ideal market* on the one hand and separate market-to-market coordination agreements** on the other hand. Not surprisingly, they find the ideal market design is most efficient and independent market-to-market coordination is least efficient in their numerical analysis. An encompassing market coupling system (a single market coupling system able to redispatch all available energy supply resources) also achieves a high degree of efficiency. Somewhat surprising to me was that multiple independent but overlapping market coupling systems achieved similarly high degrees of efficiency so long as each supply resource is available to at least one market coupling system and each supply resource is available on the same terms (i.e. at the same price) to each market coupling system that can access it.

The paper is written to address circumstances in the European market, but has implications for trade between power systems in the United States and elsewhere as well. To put this in a U.S. context, the article suggests that if trade between the New York ISO and ISO-New England was well integrated, and if trade between the New York ISO and PJM was well integrated, then the three systems would attain a high degree of efficiency even without resorting to a single integrated dispatch across the three regional power markets.

In principle, adding efficient trade between PJM and MISO, and efficient trade between MISO and SPP, and suddenly one can obtain efficient power system arbitrage subject to the limits of the transmission system stretching from the tip of Maine down to the eastern edge of New Mexico.

In practice a few issues intrude.  Market coupling in Europe is, I think, still limited to day-ahead coordination between power systems, leaving the transmission system operators to address independently the changes in local supply and demand that arise after the day-ahead result is published.  Moving toward real-time market coupling would create additional economic value, but at the cost of a significant increase in data sharing requirements and a higher computational burden on the system operators. In considering priorities for further power market development, then, the issue is whether the benefits of moving closer to real-time market coupling are worth the costs, and this ratio then compared to the benefit-cost ratio of other identified potential market improvements.

*Ideal market = a single security-constrained economic dispatch covering the entire region, using a fully developed transmission model and accurate depictions of generation characteristics.

**Market-to-market coordination agreements = bilateral agreements between markets that govern access to the transmission capacity between the systems and setting rules to resolve congestion on the interconnecting transmission lines.

h1

The right market design for trade between power markets

May 7, 2010

Michael Giberson

Windpower Monthly has a great article describing changes in the market for transmission capacity between power systems in Europe and the benefits of the changes.  Here is a summary by way of selected quotes, but the full story is worth reading:

Most of the electricity cables connecting Europe were built when electricity systems in each country were monopolised by a single or, at most, a few companies, each operating within their individual monopoly supply areas. Each utility ran its own system and had its own generation backup for emergencies. There was no competitive pressure on the higher costs of such “island” systems since these could easily be passed on to customers who, in those days, had no alternative suppliers that they could switch to.

These interconnectors were built between neighbouring countries’ electricity grids not to enable trading and competition across borders but rather for the utilities to help each other out….  [As the Eurpopean power industry was liberalized] insufficient connection capacity between some of the national electricity networks [emerged as] one of the key problems.  [An] efficient allocation for the scarce interconnector capacity that is available is crucial to make improvements towards an integrated European electricity market.

Before the new [market coupling] system started, transmission capacity available on the two interconnectors was sold to electricity traders in tranches in annual, monthly and daily auctions, called explicit auctions. This happened completely separate from auctioning of electricity with the result that, due to the time lag in buying the transport capacity and the actual time of use, as well as other reasons, inefficiencies occurred.

Transport capacity could be bought ahead of time and hoarded, a form of market abuse. Or transmission capacity was bought for one direction, say from Germany to Denmark, which then turned out to be inappropriate because the price difference between the two was such that the electricity ought to flow in the other direction – from the low- to the high-price zone. In such instances, the electricity did then flow in the wrong direction, contradicting market forces, or not allowing extra capacity to be used – and traders and end users lost out.

Explicit auctions were implemented for interconnectors at most European borders, recounts [EMCC managing director Enno] Bšttcher. “Even though this can be considered as progress compared to the formerly applied first-come, first-served or pro-rata regimes, explicit auctions still have many disadvantages,” he says.

Today, explicit auction methods have become more sophisticated. The fundamental flaw, however, remains: that actual trade of electricity at energy exchanges in the different market areas is separate from transmission capacity, trading leading to market inefficiency. This can be reduced by combining cross-border transmission capacity allocation and electricity trade from one country or market area to another in a so-called market-coupling regime.

Market coupling uses implicit auctioning and focuses on the short term (day ahead), rather than months or a year ahead. The transmission capacity available on an interconnector the next day, as reported by the transmission system operators (TSOs), is matched with electricity bought or sold on the energy exchanges in the two countries involved for delivery the next day, creating a price for the transmission capacity and making it clear in which direction the market requires use of the transmission capacity of the cables.

In effect, market coupling is a charge placed on the power exported or imported between countries when the network interconnector capacity is optimised to reduce congestion.

The result of market coupling is that the interconnected power systems operate more efficiently, benefiting consumers and low-cost producers of power.

As Tres Amigas works out its design for the sale of transmission capacity across the proposed three-way transmission interconnection, market coupling should be among the designs contemplated.

Note that while day-ahead market coupling seems to work well between systems with relatively few interconnecting links, more complex transmission links between systems – say as exists between PJM and the Midwest ISO – may well call for still more extensive coordination. The market coupling principle seems sound, so probably forms an adequate foundation to build upon, but simple day-ahead coordination is likely insufficient. Real-time market coupling, anyone?

h1

Integrating variable energy resources to the electric power grid (cont.)

April 27, 2010

Michael Giberson

In January we noted the Federal Energy Regulatory Commission’s questions concerning the integration of “variable energy resources” to the electric power grid.  FERC asked for comments; over 120 comments have been submitted in reply (so far).  Peter Behr, of ClimateWire, characterizes some of the positions submitted in the FERC inquiry in an article available at NYTimes.com.  Behr said more than 2,800 pages worth of comments have been sent to FERC on the issue.

The fundamental issue is whether or not current industry practices unnecessarily discriminate against variable power sources such as wind and solar.  Behr said, “This debate opens another front in the continuing, behind-the-scenes struggle between the renewable power sector and some of the electricity industry’s old guard, whose historic ways of doing business are now under challenge.”

Follow

Get every new post delivered to your Inbox.

Join 50 other followers