Experience you can trust. Market Power and Market Monitoring 9th Baltic Electricity Market Mini- Forum Dr. Konstantin Petrov September 2009
Mar 30, 2015
Experience you can trust.
Market Power and Market Monitoring
9th Baltic Electricity Market Mini-Forum
Dr. Konstantin Petrov
September 2009
2
Content
Competition
Market Power
Market Monitoring
Indicators to Measure Market Power
3
Competition
4
Fundamental Price Drivers
End-User
PricesDemand
• Peak load (MW)
• Load growth (MWh)
Competition
• No straight forward indicator
Supply
• Fuel prices
• Available generation
• Available transmission
•Network access
•Regulatory uncertainty
•Market design
•Taxes/Levies/surcharges
Regulation
Competition is only one of the factors influencing electricity prices!!!
5
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”.
6
Perfect Competition
Characteristics
Each firm sets its price at the level of its
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
Under perfect competition, marginal revenue
equals price and each firm is a price taker
Assumptions:
A vast number of buyers and sellers of the
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
7
Perfectly Competitive vs. Real Markets
According to economic theory, prices in perfectly competitive markets should be
equivalent to marginal cost.
In practice, however, prices on power markets sometimes are considerably
higher…
Supply / sale
Demand /purchase
Volume (MW)
Price(€/MWh)
8
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
9
Market Power
10
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
11
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
12
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
13
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
14
Market Power Strategies - Capacity Withholding
Competitive case
Bids
Volumes
Price 1
Bids
Volumes
Price 1
Price 2
Capacity withholding
15
Market Power Strategies - Strategic Bidding
Competitive case Strategic bidding
Bids
Volumes
Price1
Bids
Volumes
Price 1
Price 2
16
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
17
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 %)
18
Market Power Examples
Daily base & peak load index APX July 2001
0
50
100
150
200
250
300
350
400
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Euro
/MW
h
Base load Peak load
Single Hour Average Price - Base
10.00
60.00
110.00
160.00
210.00
260.00
310.00
EU
R/M
Wh
APX, The Netherlands July 2001 EEX, Germany December 2001
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
19
2004
2005
31.0
7.20
0330
.09.
2003
30.1
1.20
0331
.01.
2004
31.0
3.20
0431
.05.
2004
31.0
7.20
0430
.09.
2004
30.1
1.20
0431
.01.
2005
31.0
3.20
0531
.05.
2005
31.0
7.20
0530
.09.
2005
30.1
1.20
05
100
90
80
70
60
50
40
30
20
10
0
€/M
Wh
21958 h
EEX gleitender Monatsdurchschnitt
Geschätzte Grenzkosten gleitender
Monatsdurchschnitt
Market Power Examples
Schwarz / Lang (2007) shows differences between estimated marginal
cost and EEX price in Germany
Source: Schwarz / Lang 2007
20
Market Monitoring
21
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
22
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
23
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
24
Types of Monitoring
Approved Market Rules
Compliance Monitoring Are players fulfilling their obligations?
Competition Monitoring Are players behaving competitively?
Day to day reporting
Timeliness and correct format of bids, notifications,
reporting rules etc.
Prices, volumes, constraints that became binding,
incidents
Analysis
Price trends, incidents, constraints, market structure etc.
Measures (discussions, warnings, sanctions)
Proposals for change
Market structure Market rules
Existing market participants
Qualitative and informal
monitoring
25
Indicators to Measure Market Power
26
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)
27
Concentration Ratio (CR)
m
iim WCR
1
(%)
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
28
Herfindahl-Hirschman Index (HHI)
n
i
in WCR1
2 (%)10000
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
29
Characteristics of HHI
HHI for equal shares
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
1 2 3 4 5 6 7 8 9 10 11 12 13
Number of players
HH
I
Moderately concentratedHighly concentrated
30
Characteristics of HHI
HHI for 1 dominant player
0200400600800
1000120014001600180020002200240026002800
0% 3% 6% 9% 12% 15% 18% 21% 24% 27% 30% 33%
Difference between the dominant player and the players with equal shares
HH
I
HHI with 6 players
HHI with 10 players
Moderately concentrated
31
HHI in Romania
MIN (HHI) = 2.738MAX (HHI) = 3.914
20022002
20042004
1.4551.742
1.3421.400
1.3421.488 1.439
1.5571.490
2.738
2.764
2.803
2.944
2.848
2.005
2.105 1.809
3.012
3.025
3.043
3.364
3.226
3.708
3.261
3.279
3.604
3.914
0
500
1.000
1.500
2.000
2.500
3.000
3.500
4.000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
MIN (HHI) = 1.342MAX (HHI) = 2.944
20032003
The Herfindahl-Hirschman Index calculates the sum of The Herfindahl-Hirschman Index calculates the sum of the squared market shares of all generators in the market the squared market shares of all generators in the market and it rangeand it range From 0From 0 for a perfectly competitive market for a perfectly competitive market 0 – 1.000 the market is reasonably competitive 0 – 1.000 the market is reasonably competitive 1.000 – 1.800 the market is moderately concentrated1.000 – 1.800 the market is moderately concentrated 1.800 – 10.000 the market is highly concentrated 1.800 – 10.000 the market is highly concentrated To 10.000To 10.000 for a monopoly. for a monopoly.
n
i
shareMarketHHI
1
2generator i of %
Source: OPCOM
32
Residual Supply Index
100
TD
CTCRSI i
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)
33
Price Cost Margin Index and Lerner Index
100
MC
MCPPCMI 100
P
MCPLI
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
34
Estimation of the mark-ups for the different demand segments
Simulation of strategic behaviour (Gaming)
Production Cost Curve (Estimate system marginal cost)
Build generation portfolios
Defining the Demand Segments
Simulation Models / Approach
35
Simulation Models / Techniques
Advantages
•Flexible•Tractable•Important existing literature
Cournot(Quantity)
•Little descriptive power•Not very realistic (pure quantity bids)•Weak predictive power
•Realistic (firms do not sell all their output to the spot market )•Impact on market power?
ForwardContracts
•Requires additional assumptions
•More realistic (reflects actual bidding rules)•Allow better understanding of companies’ bidding behaviours• Better predictions
SFE (Supply Function
Equilibrium) (Quantity+price)
•Little computational tractability •Multiple equilibrium
•Flexible•Tractable•“non-storability“ aspects
Bertrand(Price)
•Not very realistic (pure price bids)•Producers have limited capacity•Weak predictive power
Disadvantages
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)
Hobbs (1986), Aghion and Bolton (1987)
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)
Allaz and Vila (1993), Bolle (1993), Powell (1993),Batstone (2000), Newbery (1998) , Green (1999)
36
KEMA Market Models
Overview
PLEXOS, ProSym, SYMBAD
Short-term
dispatch and
strategic bidding
SDDP
Water optimisation
PSS/E, PowerFactory, ELEKTRA
Network Models
(Load flow
analysis)
KREMLIN
Balancing market
KEEM, PLEXOS
Long-term
planning
37
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
38
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
Experience you can trust.
Many thanks!
KEMA Consulting GmbH
Kurt-Schumacher-Str. 8, 53113 Bonn
Tel. +49 (228) 44 690 00Fax +49 (228) 44 690 99
Dr. Konstantin Petrov
Managing Consultant
Mobil +49 173 515 1946 E-mail: [email protected]