Antitrust & Competition Policy Blog

Editor: D. Daniel Sokol
University of Florida
Levin College of Law

A Member of the Law Professor Blogs Network

Tuesday, July 28, 2009

A Fundamental Power Price Model with Oligopolistic Competition Representation

Posted by D. Daniel Sokol

Miguel Vazquez, Universidad Pontificia Comillas and Julian Barquan, Pontifical University Comillas of Madrid describe A Fundamental Power Price Model with Oligopolistic Competition Representation.

ABSTRACT: Most popular approaches for modeling electricity prices rely at present on microeconomics rationale. They aim to study the interaction between decisions of agents in the market, and usually represent the impact of uncertainty in such decisions in a simplified way. The usual methodology of microeconomics models is the study of the interaction between the profit-maximization problems faced by each of the firms. On the other hand, there is a growing literature that describes the power price dynamics from the financial standpoint, through the statement of a more or less complex stochastic process. However, this theoretical framework is based on the assumption of perfect competition, and therefore the stochastic process may not capture important features of price dynamics. In this paper, we suggest a mixed approach, in the sense that the price is thought of as the composition of a long-term component, where the strategic behavior is represented, and a short-term source of uncertainty that agents cannot take into account when deciding their strategies. The complex distributional implications of the oligopolistic behavior of market players are then given by the long-term-component dynamics, whereas the short-term component captures the uncertainty related to the operation of power systems. In addition, this modeling approach allows for a direct description of the long-term volatility of power markets, which is usually hard to estimate through statistical models.

http://lawprofessors.typepad.com/antitrustprof_blog/2009/07/a-fundamental-power-price-model-with-oligopolistic-competition-representation.html

| Permalink

TrackBack URL for this entry:

http://www.typepad.com/services/trackback/6a00d8341bfae553ef01157146854e970c

Listed below are links to weblogs that reference A Fundamental Power Price Model with Oligopolistic Competition Representation:

Comments

Interesting abstract. I don't know about European power markets, but I have found that US power markets are driven by the cost of the cost of fuel for the generating unit on the margin. Which unit is on the margin depends on demand levels, which vary widely across the day and across seasons. The reason that power prices move, then, is largely a function of fuel price movement in conjunction with demand movement.

In most US markets, natural gas fires the marginal unit during the majority of peak hours. In some regions, coal is relatively heavy, while in others, fuel oil is important. Congestion barriers hinder the long distance movement of power across regional boundaries.

In most US markets, demand varies mostly with temperature and humidity conditions. Hot-Warm-Mild-Cool-Cold.

The 'microeconomics' type of model described is a dispatch model. I have not found that approach to be particularly useful in producing anything more than the roughest prognostications.

My company, RisQuant Energy, models US power markets in ISO/RTO areas. That would be the midwest, mid-Atlantic, northeast, and (soon) Texas. We use these factor relationships to develop synthetic forward curves in illiquid zones out to thirty six months. We also report volatility and power/natgas delta sensitivity for that time span.

We have found that volatility is best partitioned into two parts, forward and delivery. Forward volatility is mostly about fuel price volatility while delivered volatility is mostly about weather with spot fuel volatility playing a secondary role. We do this because the uncertainty associated with weather does not dissipate like that of fuel costs. We know very little more about the weather one month out than we knew nine months out.

Posted by: James Carson | Aug 14, 2009 2:46:25 PM

Post a comment