Enron Mail

From:william_hogan@harvard.edu
To:
Subject:California Market Design Breakthrough
Cc:
Bcc:
Date:Mon, 14 Jan 2002 00:13:45 -0800 (PST)


California Market Design Breakthrough
William W. Hogan
January 13, 2002

A major step forward in California electricity market design appear=
s in the proposal offered by the California Independent System Operator (CA=
ISO). The CAISO paper is the focus of immediate discussion with its stakeh=
olders in anticipation of a market design reform to be filed at the Federal=
Energy Regulatory Commission (FERC). This proposal has far-reaching and i=
mportant implications. The parallel discussions on standard market design =
at FERC must take this CAISO proposal as strong testimony adding support fo=
r the Northeast market model that has been shown to work. The Midwest Syst=
em Operator should be further encouraged in its market design since it alre=
ady incorporates the insights now reported in California. Texans should lo=
ok with concern at the Texas wholesale market design that bears a strong re=
semblance to the original California model now explained as built on false =
premises. The rest of the west should applaud and embrace the simple cando=
r of the CAISO and get moving with a standard market design that respects t=
he common physics of transmission systems. European regulators and market =
participants should save years of agony and expense trying to build a viabl=
e electricity market built on the same false premises that California now e=
xposes.
California's market meltdown starting in June 2000 reverberated aro=
und the world. The high prices and market collapse have been cited in many=
policy discussions as the reason to reverse course on electricity restruct=
uring. As bad as it was, the California market collapse was made much wors=
e by factors unique to California in the retail market. And this has distr=
acted many from learning the lessons about wholesale market design that hav=
e much greater relevance beyond California's borders.
California's was the most prominent market design built on the fall=
acy that the special characteristics of the electricity transmission grid w=
ere just details that could be ignored when designing the market. Built on=
a faith that decentralized trading could solve all problems, the original =
California wholesale market design embraced a philosophy that emphasized tr=
ading and minimized the role of the essential system operator.
The key bad ideas have been discussed at length elsewhere (see www =
.whogan.com ). Central elements included simplifications supposedly inten=
ded to make the market better for traders. In place of the reality of many=
locations with important impacts on transmission constraints, the design r=
elied on a few large pricing zones. In place of a realistic model of the t=
ransmission grid, the design used a so-called commercial network model that=
ignored critical features of the grid. In place of point-to-point financi=
al transmission rights, the design employed interface rights between zones.=
In place of economic dispatch coordinated by the system operator, the des=
ign relied on decentralized trading that embraced artificially separated ma=
rkets for ancillary services and ad hoc balancing rules for real-time adjus=
tments. In place of a coordinated energy spot market, the design imposed a=
separate power exchange and balanced scheduling requirements. =20
This flawed wholesale market design did not work. From the very be=
ginning, problems appeared and by the end of 1999 the FERC had rejected the=
many attempts to fix the superstructure without rebuilding the foundation.=
Congestion Management Reform soon became Comprehensive Market Redesign. =
Although the reform effort was overtaken by later events, the failure of th=
e California wholesale market design should be a warning for every other ma=
rket that has embraced its philosophy and adopted its details.
Now the CAISO has taken up the serious task of fixing what is broke=
n. No longer constrained to be politically correct in defending the ideolo=
gical errors that drove the earlier design process, the CAISO reports the s=
imple truth. "Upon reexamination of the [Congestion Management Reform] pro=
posal ? we find that some of the crucial assumptions underlying the [Locati=
onal Pricing Areas] concept break down."(CAISO, p. 13) The assumpt=
ions were crucial and flawed. Trying to make the market simpler than is po=
ssible turned out not to be possible.
"?in reality, the 'simplicity' of the zonal system only appears so becau=
se the complexity is assumed away, allowing market participants to ignore i=
t in scheduling while the CAISO must manage it through real time adjustment=
s and periodic modifications to the rules to mitigate novel gaming strategi=
es as they arrive. ... it will be far simpler, and more transparent, to de=
sign forward [congestion management] procedures to be as consistent as poss=
ible with the real-time operating needs of the grid."(CAISO, p. 14) =
Amen. The CAISO paper elaborates on the implications and outlines a new d=
irection that is a major break with its past. The same design lesson was l=
earned earlier in the Northeast markets including PJM, New York and New Eng=
land. Similar insights can be found in well-designed wholesale markets in =
other countries, from Chile to New Zealand.
Given its prominence in promoting a fallacy, its failure in practic=
e, and now its complete reversal on the basic principles, California's turn=
around should be a pivotal moment that captures attention in the arcane wor=
ld of electricity market design.
The changes now proposed for California would be fundamental and sw=
eeping. Once reality is acknowledged, the range of possible market designs=
narrows substantially. The pieces must fit together. Prices have to refl=
ect actual operations. And the system operator must perform certain functi=
ons in a way that supports the public interest in an efficient and effectiv=
e wholesale market. The centerpiece would be a coordinated spot market org=
anized as a bid-based, security-constrained, economic dispatch with nodal p=
rices. This can support bilateral transactions with transmission usage cha=
rged at the difference in nodal prices. Point-to-point financial transmiss=
ion rights would provide the economic equivalent of otherwise unworkable ph=
ysical transmission rights. This design works in theory and in practice.=
=20
The CAISO proposal is a beginning but still tentative embrace of th=
is standard market design along with consistent day-ahead markets and unit =
commitment. The CAISO needs to go further, and some of the ideas require c=
lose scrutiny. For example, the suggested introduction of an available cap=
acity market would be new to California and the concept is not fully develo=
ped in the proposal, but this is not much different than the rest of the co=
untry where design of capacity markets is problematic. There is much yet t=
o be done with the stakeholders and then at FERC.
However, the CAISO deserves recognition and support for its candor =
and the intelligence of its proposal for the basic elements of the market d=
esign. Enlightened market participants and leaders responsible for the pub=
lic interest should resist any effort to compromise this essential reform o=
r silence the truth. The CAISO has taken a long step in the right directio=
n. =20
Everyone interested in the success of electricity restructuring sho=
uld hope that Californians complete the journey, soon. We know what to do,=
and we can't afford to get it wrong again.
The CAISO proposal, "Market Design 2002 Project: Preliminary Draft Comprehe=
nsive Design Proposal," January 8, 2002, is available on the web at:http://=
www.caiso.com/docs/09003a6080/13/58/09003a6080135879.pdf=20
For further papers on standard market design, see:http://ksgwww.harvard.edu=
/hepg/standard_mkt_dsgn.htm=20
William W. Hogan
John F. Kennedy School of Government
Harvard University
79 John F. Kennedy Street
Cambridge, MA 02138
617-495-1317 (o)
617-495-1635 (f)
william_hogan@harvard.eduwww.whogan.com =20
orhttp://www.ksg.harvard.edu/whogan