Structural Analysis of Electricity Demand and Supply Interactions*
| Author | Derek Bunn,Carlo Fezzi |
| DOI | http://doi.org/10.1111/j.1468-0084.2010.00596.x |
| Date | 01 December 2010 |
| Published date | 01 December 2010 |
827
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford, 2010. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 72, 6 (2010) 0305-9049
doi: 10.1111/j.1468-0084.2010.00596.x
Structural Analysis of Electricity Demand and
Supply InteractionsÅ
Carlo Fezzi† and Derek Bunn‡
†CSERGE, School of Environmental Sciences,University of East Anglia, Norwich NR4 7TJ,
UK (e-mail: c.fezzi@uea.ac.uk)
‡London Business School, Sussex Place, Regent’s Park, London NW1 4SA, UK
(e-mail: dbunn@london.edu)
Abstract
We specify a structural asymmetric vector error-correction model to identify and
estimate the demand and supply functions in hourly day-ahead wholesale electricity
markets. In doing so, we provide, inter alia, new insights into a well-established but
unresolved issue concerning the extent of the demand elasticity to price in these
markets. Weshow that whilst demand appears to be inelastic in the short-run, the quan-
tity traded on the market is significantly influenced by the price level and responds
to previous disequilibria in the supply curve through an asymmetric error-correction
mechanism, reacting to a positive disequilibrium but not to a negative one.
I. Introduction
Following the progressive liberalization and restructuring of the power sector, the
widespread introduction of day-ahead markets has increasingly provided the refer-
ence prices for wholesale electricity in many countries. Typically, these markets are
organized as auctions providing market-clearing prices for the subsequent daily set
of hourly, or half-hourly, trading periods. As with any other freely traded product,
the fundamental drivers of electricity production and price formation are supply and
ÅThis paper has been written during a period in which the first author was visiting the Energy Markets Group
at London Business School. The support of the University of Bologna under the Marco Polo fellowship is
gratefully acknowledged. Weare grateful to the Editor and two anonymous referees for their useful comments
which helped improving the quality of the paper. Many thanks also to Giuseppe Cavaliere, Michele Costa,
Luca Fanelli, Attilio Gardini, Francesco Lisi, Stefano Sacchetto and the participants of the 29th Conference
organized by the International Association of Energy Economists (IAEE) in Potsdam, to which an earlier
version of the paper was presented.
JEL Classification numbers: C32, Q41, L94.
828 Bulletin
demand. However, electricity presents unique features. Unlike other commodities,
electricity, once produced, cannot be economically stored. The physical laws that
determine the functioning of the power grid require, in fact, the continuous balance
between production and consumption. Electricity, therefore, has to be constantly pro-
duced in ‘real-time’, and the supply–demand balancing process is crucial both in the
physical sense of system security, and in the financial sense of price formation.
Furthermore, since demand typically varies widely throughout the day and the year,
every electricity market needs low marginal cost (high capital cost) generation units
operating baseload and flexible plant, typically with lower capital costs but higher
marginal costs, producing intermittently at the peaks (Bunn, 2003). This feature of
a steeply increasing, discontinuous supply function effectively maps high demand
volatility into even higher price volatility.1However, this daily, high-frequency price
volatility is typically not passed through to many of the end consumers (e.g. house-
holds, small companies), who, because of fixed retail tariffs, do not bear the short-term
price fluctuations. For these reasons, the extent of the electricity demand response
to price has become an important but long-standing dilemma. Borenstein (2005), for
example, observes that ‘electricity demand elasticities are subject to a nearly endless
contention’. Nevertheless, producing reliable estimates of this parameter and, more
generally, understanding the interactions of electricity supply and demand is relevant
not only for electricity market players but, considering the importance of this sec-
tor for the whole economy, has great significance from a general policy perspective.2
Furthermore, encouraging demand responsiveness to price, for example via the intro-
duction of real-time metering and pricing in the retail markets, has been proposed
as an effective way to mitigate market power and increase electricity market effi-
ciency (Borenstein and Bushnell, 1999; Rosenzweig et al., 2003; Borenstein, 2005;
Borenstein and Holland, 2005; Holland and Mansur, 2006; Spees and Lave, 2008).
Indeed, a significant body of theoretical research on electricity market be-
haviour has focused on the strategic interactions of firms, their implications for
prices and social welfare and in particular upon evaluating participants’market power.
These themes have been pursued by developing competitive models based mainly
upon Cournot (Borenstein, Bushnell and Knittel 1999; Neuhoff et al., 2005), supply
function equilibrium (Green and Newbery, 1992; Green, 1996) or other game-
theoretical formulations, whose outcomes have been evaluated via simulation or on
empirical data (Borenstein and Bushnell, 1999; Wolfram,1999; Borenstein, Bushnell
and Wolak, 2002; Mansur, 2008; Spees and Lave, 2008). In these static theoretical
1Forinstance, the coefficient of variation (CV) of electricity wholesale price is, typically, at least two times that
of demand. See, for example, the descriptive statistics in Table1 presenting the data analysed later in the article.
2The crucial importance of understanding the interactions of supply and demand, in particular for market
power mitigation, was evident in the crisis hitting the California wholesale market in the early 2000s. An un-
favourable conjuncture (low hydroelectric production and high fuel prices) and, according to most analysis
(e.g. Borenstein, 2002; Borenstein et al., 2002) a substantial abuse of market power, triggered an unprecedented
and ultimately politically unsustainable level of wholesale prices. This generated enormous profits for the
supply side and bankruptcy for many regulated utilities on the demand side, producing physical and financial
market failure, with black-outs and the collapse of the entire wholesale market.
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2010
To continue reading
Request your trial