Download Overview

TOWARD MORE COMPLETE ELECTRICITY MARKETS
Zhiyong Wu, Carnegie Mellon University, Phone 203-761-8355, E-mail: [email protected]
Marija Ilic, Carnegie Mellon University, Phone 412-268-9520, E-mail: [email protected]
Overview
This paper is motivated by the on-going problems with sustainable value-based investments in the evolving electricity markets.
Even the best functioning spot markets are challenged by the lack of signals for investment in generation and transmission capacity.
While the reasons for this situation are multifold, one of the obvious questions concerns the management of physical and financial
uncertainties, i.e. demand variations and fuel prices in the evolving electricity markets. Given that these risks are multi-temporal
and location specific, it is our premise that failure to systematically manage these uncertainties is one of the major shortcomings of
current market structure.
At present there are few liquid longer-term electricity markets, which are essential for ensuring both reliable service and sufficient
capacity reserve to avoid boom-and-bust cycles in generation capacity. Determining near-optimal investments for long-run
efficiency requires transparent signals for decision making under various physical and financial uncertainties. In this paper we
introduce a set of coordinated sub-markets, each defined for a specific time horizon, ranging across day-, month-, season-, year-,
and multi-year horizons, referred to as a Stratum Electricity Market (SEM). On-peak and off-peak heat-rates (mmBtu/MWh)
instead of energy or capacity are defined as the long-term market products to account for locational and inter-temporal
characterastics of electricity as well as for the impacts of uncertain fuel prices. In the SEM electric energy ($/MWh) is the product
traded in short-term day-ahead and real-time markets similarly to the existing spot electricity markets. The premise in this paper
is that long-term heat-rate would be converge to the electricity prices using the daily zonal fuel price indices in the day-ahead
markets. The sub-markets are cleared sequentially from longer-term to shorter-term through the uniform auctions. Only the
increamental position changes are settled against the price in higher stratum. All contracts are financial up until the day-ahead
markets when they have to be tied to physical generation capacity if the positions are not closed out in lower strata. We evaluate the
long-term effects of the SEM on the system reliability and efficiency. We also provide comparisons of different market structure
and regulatory rules which are essential for the long-term investments.
Methods
The newly proposed SEM structure is modelled as the generator investment alternative solution to the long-term resource adequacy
problem. The investment problem is modeled as a stochastic dynamic programming problem intended to provide a profit
maximization at the pre-chosen value-at-risk over a long time horizon by generators. The long-term growth and short-term
deviation of demand are represented as stochastic processes. The spot market is modeled as bi-level non-cooperative game and the
long-term forward market is formulated based on mean-variance criteria and market equilibrium arguments. The inter-dependencies
of dynamics in different sub-markets and their effects on investment decisions and profitability of market participants are analyzed
and compared with other market structures, spot only energy markets in particular.
Results
The optimal investment decisions under different market structures were studied here including the SEM market structure with spot
market gaming, spot-only market structure with gaming and spot-only market with marginal cost bids. The SEM market structure
induced new generation capacity while spot-only market structure discouraged significant investment decisions with or without the
consideration of gaming opportunities, maily due to the high price volatility and associated risks. The unit also obtained the highest
expected profits under SEM. This is mainly due to the fact that without new capacity the system would frequently slip into
blackouts when total demand is higher than installed capacity. The blackout penalty prices reduced the unit’s profit margin
substantially.
Conclusions
The core of generation resource adequacy problem is how to manage different long-term and short-term uncertainties across
stakeholders. In the regulated industry, consumers take all the risks of generation expanding and the producers take zero risks after
the state regulators approve the plan under the guaranteed rate-of-return regulation. In the liberalized electricity markets, the risks
of new generation investment shift to the other end of spectrum. The investors bear all the risks. Due to the non-storability of
electricity and the instantaneous balance of supply and demand at every second, the price and revenue volatility/uncertainties under
the current spot market setup are too huge for generation companies and investors to take. In this paper we propose the SEM
structure as an enhancement to the short-term spot market. This market would eliminate the need for various installed capacity and
reliability administrative payments currently under consideration. A fundamental modeling approach is further developed to model
and simulate the SEM structure. The results demonstrate that SEM structure enables the risk distribution fairly between different
parties and aligns their economic incentives with the short-term and long-term social welfare.
References
Chao, Hung-Po and Robert Wilson. 2004. “Resource Adequacy and Market Power Mitigation via Option Contracts”, Electric
Power Research Institute, Draft. Available at http://faculty-gsb.stanford.edu/wilson/PDF/Electricity%20Markets/ChaoWilson,Resource%20Adequacy%20and%20Market%20Power%20Mitigation031804.pdf
Cramton, Peter and Steven Stoft. 2006. “The Convergence of Market Designs for Adequate Generating Capacity”, available at
http://stoft.com/p/50.html.
Energy Information Administration (EIA). 2005. Annual Energy Review. Washington, D.C.
Hogan, William. 2005. “On an Energy-Only Eletrcitiy Market Design for Resource Adequacy”, paper prepared for the California
IOS, September 23.
Skantze, Petter and Marija Ilic. 2001. Valuation, Hedging and Speculation in Competitive Electricity Markets: A Fundamental
Approach. Kluwer Academic Publishers.
A. Botterud and M. Ilic, “Optimal investments in power generation under centralized and decentralized decision making”, IEEE
Trans. on Power Systems, vol. 20-1, pp. 254-263, February 2005.