Transcript Market Power and Market Monitoring

Market Power and Market Monitoring
9th Baltic Electricity Market Mini-Forum
Dr. Konstantin Petrov
September 2009
Experience you can trust.
Content
 Competition
 Market Power
 Market Monitoring
 Indicators to Measure Market Power
1
Competition
2
Fundamental Price Drivers
•Network access
Regulation •Regulatory uncertainty
•Market design
•Taxes/Levies/surcharges
Supply
• Fuel prices
End-User
Prices
Demand
• Peak load (MW)
• Load growth (MWh)
• Available generation
• Available transmission
Competition
• No straight forward indicator
Competition is only one of the factors influencing electricity prices!!!
3
Definition
 Adam Smith: "in the sense of rivalry in a race, a race to get limited
supplies or a race to be rid of excess supplies".
 Marshall: “The strict meaning of competition seems to be the racing of
one person against another, with special reference to bidding for the sale
or purchase of anything. […] In modern economic theory, a market is said
to be competitive, when the number of firms selling a homogeneous
commodity is so large, and each firm’s market share is so small, that no
individual firm finds itself able to influence appreciably the commodity price
by varying the quantity of output it sells“.
 Oxford dictionary of economics: “the situation when anybody who
wants to buy or sell has a choice of possible suppliers or customers”.
4
Perfect Competition
Characteristics
Assumptions:
 Each firm sets its price at the level of its
 A vast number of buyers and sellers of the
marginal costs to maximise its profits
 If a firm sets a price above the price of
other firms it sells nothing
 If a firm sets a price below the other firms’,
it will have to supply all of the market
demand for the product
 If a firm charges less than marginal costs,
it will fail to break even for that unit of
output
same, homogeneous product
 Perfect mobility of people and resources,
 Profit-maximising behaviour by producers,
welfare-maximising behaviour by
consumers
 Perfect knowledge by all buyers and
sellers of all relevant present and future
conditions in all markets
 Absence of externalities
 Under perfect competition, marginal
revenue equals price and each firm is a
price taker
5
Perfectly Competitive vs. Real Markets
 According to economic theory, prices in perfectly competitive markets should be
equivalent to marginal cost.
Price
(€/MWh)
Supply / sale
Demand /purchase
Volume (MW)
 In practice, however, prices on power markets sometimes are considerably
higher…
6
Workable Competition
 Perfect competition is well-recognised as unrealistic in real life
 The concept of workable competition emerged from literature (Clark,
1940):
– “a market where competition is not perfect but allows the different
buyers and sellers to choose between a sufficient number of
alternatives” or
– “workable competition can be defined as the persistent absence of
players with market power”
 Basic criteria: limited number of suppliers, no large differences between
parties, limited barriers to entry, low switching costs
7
Market Power
8
Market Power / Definition
 Definition: “Market power is the ability of a firm to profitably raise the price
of a product”
 Market power exists in nearly every product market
 Only perfectly competitive markets exhibit no market power; in all markets,
privately-owned firms continually attempt to exercise market power
 Prof. Wolak:
– Question is not whether or not firms exercise market power
– Question is when does the exercise of market power cause significant
harm to consumers
9
Technical Characteristics of Electricity Business
 Power flows follow the laws of physics
 Transport of electricity is constrained (congestion)
 Power generation is constrained (ramping rate, black start capabilities,
environmental factors)
 Storage is extremely limited and expensive
 Demand varies each second and need to be balanced instantaneously
10
Economic Characteristics of Electricity Business
 Absolute cost advantages of established firms
– Techniques
– Know how
– Research
 Consumer loyalty
 Capital requirements and funding constraints
 Economies of scales
 Irreversible commitment
11
Typical Reasons for Market Power
 Transmission constraints and market fragmentation
 High degree of concentration
 Inelastic demand
 Peak demand conditions and instantaneous balancing
 Strong national incumbents
 Joint capital control of generation and transmission capacities
 Gaps in market arrangements
12
Market Power Strategies - Capacity Withholding
Competitive case
Capacity withholding
Bids
Bids
Price 2
Price 1
Price 1
Volumes
Volumes
13
Market Power Strategies - Strategic Bidding
Competitive case
Strategic bidding
Bids
Bids
Price 2
Price
1
Price 1
Volumes
Volumes
14
Market Power Determinants
 Physical - Generation & Load
– Cost, capacities (supply curve)
– Ownership structure of generation assets (market share)
– Price elasticity of demand
– Load profile
– Network constraints
 Administrative & Regulatory Environment
– Type of market - mandatory/voluntary
– Existence of bilateral trading, forward contracts etc.
– Market entry regime, market monitoring
– Horizontal / vertical integration
15
Studies on Market Power
 Newbery/Green (1992 and 1995) observed significant exercise of market
power in the old E&W Pool (complacent duopoly)
 Frank Wolak estimated substantial welfare damage caused by the
exercise of market power in California in 2000/2001
 European Commission (Sector Study 2007) concluded unsatisfactory level
of competition on EU market due to high concentration and constrained
transmission interconnection (market fragmentation)
 Von Hirschhausen (2007) observed insufficient competition in the German
electricity market (mark-ups > 30 %)
16
Market Power Examples
APX, The Netherlands July 2001
EEX, Germany December 2001
Daily base & peak load index APX July 2001
Single Hour Average Price - Base
400
310.00
350
260.00
EUR/MWh
250
200
150
210.00
160.00
110.00
60.00
100
10.00
Tu
e
Fr
i
Su
n
Peak load
W
ed
Base load
on
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
M
8
Sa
t
7
Th
u
6
Tu
e
5
Fr
i
Su
n
4
on
3
W
ed
2
Tu
e
1
M
0
Sa
t
04
/1
2/
01
06
/1
2/
01
08
/1
2/
01
10
/1
2/
01
12
/1
2/
01
14
/1
2/
01
16
/1
2/
01
18
/1
2/
01
20
/1
2/
01
22
/1
2/
01
24
/1
2/
01
26
/1
2/
01
28
/1
2/
01
30
/1
2/
01
01
/0
1/
02
50
Th
u
Euro/MWh
300
 The Dutch regulator and TSO expressed significant concerns in 2001
when huge spikes appeared at APX (up to 1200 €/MWh in single hour)
 In Germany, large price spikes appeared at EEX in December 2001
17
100
90
80
€/MWh
Market Power Examples
EEX gleitender Monatsdurchschnitt
Geschätzte Grenzkosten gleitender
Monatsdurchschnitt
2005
70
60
50
2004
40
30
20
21958 h
10
0
Source: Schwarz / Lang 2007
 Schwarz / Lang (2007) shows differences between estimated marginal
cost and EEX price in Germany
18
Market Monitoring
19
Objectives
 Identify market design flaws or anti-competitive behaviour
 Assist the different parties in identifying improvements in market
arrangements
 Ensure corrective actions
 Prevent from future problems
 Close monitoring discourage rule violation
20
Scope
 Continuous analysis to identify potential problems requiring further study,
including design flaws and undesirable behaviour
 Investigation of any problems identified by own screening, or in response
to complaints from others
 Regular reporting on the results of analysis and investigations
 Collect and publish other relevant market information
 Take corrective actions, initiate / recommend rules changes and other
improvements
21
Principles
Main principles
 Outcome orientation
 Customer focus
 Proportionality (i.e. being aware of and limiting the regulatory burden) and cost
efficiency
Additional principles
 Consistency over time
 Public reporting subject to any confidentiality requirements
 Encourage market participants to collaborate with the monitoring activities
– Transparency allows other entities to perform their own analyses and
contribute to monitoring
 Special investigations in response to specific market outcomes and/or events
 Avoid regulatory micro-management interfering with normal market outcomes
22
Types of Monitoring
Existing market participants
Approved Market Rules
Competition Monitoring
Are players behaving competitively?
Compliance Monitoring
Are players fulfilling their obligations?
Day to day reporting
Timeliness and correct
format of bids, notifications,
reporting rules etc.
Prices, volumes, constraints
that became binding,
incidents
Qualitative and
informal
monitoring
Analysis
Price trends, incidents, constraints,
market structure etc.
Measures (discussions,
warnings, sanctions)
Proposals for change
Market structure
Market rules
23
Indicators to Measure Market Power
24
Competition Analysis
 Study whether market power is exercised or exists
 Types of analysis
– Structural Analysis
 Concentration ratio (CR)
 Market Share and Herfindahl-Hirschman Index (HHI)
 Residual Supply Index (Pivotal Supplier Indicator)
– Behavioural Analysis
 Price-Cost Margin Index (PCMI)
 Lerner Index (LI)
– Simulation Models (competitive benchmark using oligopoly models)
25
Concentration Ratio (CR)
CRm 
m
W (%)
i 1
i
 A commonly accepted measure of market concentration
 Takes into account the market share of the firms
 Measure the concentration of the one (CR1), three (CR3) or five largest
companies (CR5)
 For monitoring purposes thresholds indicating market power: CR1> 33,3%, CR3 >
50%, CR5 > 66,7 %
 Does not assess the behaviour, market power may be exercised in markets with
low concentration
26
Herfindahl-Hirschman Index (HHI)
n
CRn  10000W 2i (%)
i 1
 A commonly accepted measure of market concentration
 Takes into account the market share of the firms
 Approaches zero when a market consists of a large number of firms of relatively
equal size
 HHI increases both as the number of firms in the market decreases and as the
disparity in size between those firms increases
 For monitoring purposes usually some indicative limits are used, e.g.: low
concentration HHI < 1000, moderate concentration 1000 < HHI < 1800, high
concentration HHI > 1800
 Does not assess the behaviour, market power may be exercised in markets with
low concentration
27
Characteristics of HHI
HHI for equal shares
10000
9000
8000
7000
HHI
6000
5000
4000
3000
2000
1000
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Number of players
Highly concentrated
Moderately concentrated
28
Characteristics of HHI
HHI for 1 dominant player
HHI with 6 players
HHI
HHI with 10 players
2800
2600
2400
2200
2000
1800
1600
1400
1200
1000
800
600
400
200
0
Moderately concentrated
0%
3%
6%
9%
12% 15% 18% 21% 24% 27% 30% 33%
Difference between the dominant player
and the players with equal shares
29
HHI in Romania
The Herfindahl-Hirschman Index calculates the sum of
the squared market shares of all generators in the market
and it range
From 0
for a perfectly competitive market
0 – 1.000
the market is reasonably competitive
1.000 – 1.800 the market is moderately concentrated
1.800 – 10.000 the market is highly concentrated
To 10.000
for a monopoly.
MIN (HHI) = 2.738
MAX (HHI) = 3.914
n
HHI 
%Market share of i generator2
2002
MIN (HHI) = 1.342
MAX (HHI) = 2.944
i 1
2003
2004
4.000
3.500 3.914
3.708
3.604
3.000
3.364
3.279
2.500
2.944
3.261
2.000
2.738
3.226
1.500
2.848
2.803
1.742
2.005
3.043
1.455
2.105 1.809
3.025
1.000
3.012
500
1.557
1.490
1.488
2.764
1.400
1.439
1.342
1.342
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Source: OPCOM
30
25
26
27
Residual Supply Index
RSI
TC  Ci

100
TD
 A commonly accepted measure of market concentration
 Measures the importance of each individual company to meet demand
 If RSI higher than 100 %, the residual capacity is enough to cover demand
 For monitoring purposes usually some indicative thresholds are used, e.g.
RSI should be not less than 110 % for more than 5 % of time (438 hours
per year)
31
Price Cost Margin Index and Lerner Index
PCMI
P  MC

100
MC
LI

P  MC
100
P
 Commonly accepted measures of competitive behaviour
 Takes into account the difference between actual market prices and
competitive market price measured by marginal cost
 Significant deviations from competitive market prices indicate market
power
 Difficulties with computation of marginal cost and interpretation of results
32
Simulation Models / Approach
Defining the Demand Segments
Build generation portfolios
Production Cost Curve
(Estimate system marginal cost)
Simulation of strategic behaviour
(Gaming)
Estimation of the mark-ups for the
different demand segments
33
Simulation Models / Techniques
Advantages
Cournot
(Quantity)
•Flexible
•Tractable
•Important existing literature
Disadvantages
•Little descriptive power
•Not very realistic (pure quantity bids)
•Weak predictive power
Kahn (1998), Andersson and Bergman (1995), Hogan (1997) Borenstein and Bushnell (1999), Smeers
and Wei (1999), Hobbs et al. (2002, 2003), Younes and Ilic (1997), Berry et al. (1998), Stoft (1999),
Willems (2002)
Bertrand
(Price)
•Flexible
•Tractable
•“non-storability“ aspects
•Not very realistic (pure price bids)
•Producers have limited capacity
•Weak predictive power
Hobbs (1986), Aghion and Bolton (1987)
SFE (Supply
Function
Equilibrium)
•More realistic (reflects actual bidding rules)
•Allow better understanding of companies’
bidding behaviours
• Better predictions
•Little computational tractability
•Multiple equilibrium
(Quantity+price)
Klemperer and Meyer (1989), Green and Newbery (1992), Bolle (1992), Bohn et al. (1999), Rudkevich
et al. (1999), Day et al. (2001), Baldick and Hogan (2001)
Forward
Contracts
•Realistic (firms do not sell all their output to the
spot market )
•Impact on market power?
•Requires additional assumptions
Allaz and Vila (1993), Bolle (1993), Powell (1993),Batstone (2000), Newbery (1998) , Green (1999)
34
KEMA Market Models
Overview
Long-term planning
KEEM, PLEXOS
Network Models
(Load flow
analysis)
Short-term
dispatch and
strategic bidding
Balancing market
PSS/E, PowerFactory,
ELEKTRA
PLEXOS, ProSym,
SYMBAD
KREMLIN
Water optimisation
SDDP
35
Practical Issues
 Structural indicators possess low predictability power, market power can exist in
market with low concentration
 Estimation of marginal cost may be arbitrary:
– Large data volumes needed, low data granularity and data gaps
– Consideration of start-up cost
– CO2 emissions
– Mathematical algorithms
 It is arbitrary whether (short-term) marginal cost and perfect competition models
should be used to set competitive prices
 Definition of competitive prices using oligopolistic competition requires different
type of modelling and depends on series of assumptions
36
Conclusions
 Market power is a major threat on electricity markets
 The major question is not whether some companies have market power, the
question is whether they abuse it!
 Monitoring is necessary to:
– Identify market design flaws or anti-competitive behaviour
– Ensure corrective actions
– Prevent from future problems
 There are several good structural indicators (CR, RSI, HHI), however with low
predictability power
 Behavioural indicators (LI, PCMI) describe better market performance, however
difficulties in the computation and interpretation process
37
Many thanks!
Dr. Konstantin Petrov
Managing Consultant
Mobil +49 173 515 1946
E-mail: [email protected]
KEMA Consulting GmbH
Kurt-Schumacher-Str. 8, 53113 Bonn
Tel. +49 (228) 44 690 00
Fax +49 (228) 44 690 99
Experience you can trust.