energy economics electricity markets · 2014-07-20 · 1 energy economics electricity markets prof....

© Prof. Dr. Georg Erdmann

1 Energy Economics: Electricity Markets

Prof. Dr. G. Erdmann Fachgebiet Energiesysteme / TU Berlin Einsteinufer 25 / TA 8 (Room TA 033) D-10587 Berlin

georg.erdmann@TU-Berlin.de

Internet: http://www.ensys.tu-berlin.de EPS-SIF International School of Energy 17-23 July 2014, Varenna

© Prof. Dr. Georg Erdmann

2 Content

•  Theory of Efficient Power Dispatch •  Implementation and Market Results

•  Problem 1: Possible Price Manipulation •  Problem 2: Free GHG Allowances •  Problem 3: Regulation and Balancing Power •  Problem 4: Clearing of Electricity Futures •  Problem 5: Lack of Power Plant Investments •  Problem 6: New Market Design

© Prof. Dr. Georg Erdmann

3 Theory of Efficient Power Dispatch

© Prof. Dr. Georg Erdmann

4 Generation Decision in a Competitive Market

Profit Π = p·Q – Cf– cv·Q p Price Q Generation (Quantity) cv Variable unit cost Cf Fixed costs

Profit maximizing under atomistic competition:

( ) 0f vdC d c Qd dQ dpp QdQ dQ dQ dQ dQΠ ⋅

= + − − = vdCp cdQ

= =

= 0 Sales price must exceed the variable costs, otherwise no offer

= 0

© Prof. Dr. Georg Erdmann

5

100

80

60

40

20

0 20 30 40 50 60 70

Marginal cost [Euro/MWh]

Available load without wind power [1000 MW]

Merit Order Curve Germany [without CO2 cost; Source: EU Sector Enquiry 2007, p. 260]

Demand Off-peak load

Peak load

Hard coal Lignite

Oil and gas

© Prof. Dr. Georg Erdmann

6 Implementation and Market Results

© Prof. Dr. Georg Erdmann

7 European Power Exchanges

© Prof. Dr. Georg Erdmann

8 EEX Day-ahead Auction [24 hourly products]

•  Double sided auction (bid and ask order book)

•  Multi unit auction (share auction; transaction volume and price caps can be included in the order book)

•  Sealed bid market (participants don‘t see the offers of other traders)

•  Single round auction (no continuous trade)

•  Uniform price auction (MCP instead of pay-as-bid)

•  Combinational auction (block orders are possible)

•  Gate closure 12:00 noon

•  Before 15:00 h the resulting ¼ h schedules are submitted to the TSOs

© Prof. Dr. Georg Erdmann

9 Tasks of the Balancing Group Management

The balancing group mamager (BGM) provides a ¼-h load schedule and an hourly price forecast Based on a trading strategy, bid and ask orders are submitted to the energy exchange (EPEX)

before 12:00 h

until 12:00 h

12:30 h EPEX informs traders on trading results

Day-ahead

until 15:00 h BGM submits the day ahead schedule (¼-h) to the Transmission System Operator TSO

Delivery day

00-24:00 h Deviations from the schedule are covered by the TSO in form of balancing power (or covered by intra day trading)

Following day Financial settlement of trades

© Prof. Dr. Georg Erdmann

10 Day-ahead Market [Wednesday, 31 October 2007, 18-19 h]

Day-ahead price

0 5’000 10’000 15’000 20’000 25’000 [MW] 0

50

100

150

200

250

300

350 Market Clearing

Price MCP [Euro/MWh]

Unlimited asks

Unlimited bids

Price elasticity of demand

High price volatility possible

© Prof. Dr. Georg Erdmann

11

Merit-order Effect of Wind Power [15.1.2007, 18-19 Uhr; German EEX]

0

50

100

150

200

6000 8000 10000 12000 14000 16000 Load [MW]

MCP [Euro/MWh]

Ask

Bid

Bid + 1000 MW

MCP

MCP + 1‘000 MW

250

© Prof. Dr. Georg Erdmann

12

-200

-175

-150

-125

-100

-75

-50

-25

0

25

50

75

100

125

150

175

200

5000 10000 15000 20000 25000 30000

[Euro

/MW

h]

[MW]

2008-12-22 4-5 Uhr Monday

bid ask

-101,50 Euro/MWh

© Prof. Dr. Georg Erdmann

13 Day-ahead Bid and Ask Curves

Ask curve Bid curve

Negative offerings prevent MCP to collapse during windy periods

Traded volume [MW]

Market Clearing

Price MCP

© Prof. Dr. Georg Erdmann

14 Day-ahead Bid and Ask Curves

Ask curve Bid curve

Negative offerings prevent MCP to collapse during windy periods

Traded volume [MW]

Market Clearing

Price MCP

© Prof. Dr. Georg Erdmann

15 Day-ahead Power Prices [Source: EPEX]

-­‐80

-­‐40

0

40

80

Januar Februar März April Mai Juni

2014

Phelix sowie Maximum  und  Minimum  der Stundenprodukte [Euro/MWh]

© Prof. Dr. Georg Erdmann

16 Day-ahead Electricity Market Products

Uhrzeit Sonntag Montag Dienstag Mittwoch Donnerst Freitag Samstag

0-2

2-4

4-6

6-8

8-10

10-12

12-14

14-16

16-18

18-20

20-22

22-24

© Prof. Dr. Georg Erdmann

17 Sample Power Purchase Schedule

01.01.2008 08.01.2008 15.01.2008 22.01.2008 29.01.2008

MW

© Prof. Dr. Georg Erdmann

18

Load [MW]

Portfolio Management of a Power Retailer

Mo Tu We Th Fr Sa Su

© Prof. Dr. Georg Erdmann

19

Load [MW]

Day-ahead Contracts for Singular Hours

Mo Tu We Th Fr Sa Su

Base

Peak

Remaining imbalances: balancing power

© Prof. Dr. Georg Erdmann

20 Connected Chain of Power Markets

OTC Forwards (continuous)

EEX Futures (continuous)

OTC spot trade (continuous)

Day-ahead auction

(uniform price)

OTC Intraday

(continuous)

EEX Intraday

(continuous)

(Tertiary) Reserve

(Pay-as-Bid)

years, quarters, months ahead of delivery

up to three days ahead

delivery

up to 45 min ahead

delivery

day ahead

delivery

OTC: Over the Counter

© Prof. Dr. Georg Erdmann

21 Bilateral Power Trade (OTC)

Power Producer 1

Power Producer 2

Power Producer 3

Power Producer 5

Power Producer 4

Power Retailer 1

Power Retailer 2

Power Retailer 3

Power Retailer 4

Day- ahead Power Market

The two systems “power exchange” and “OTC trade” can coexist

© Prof. Dr. Georg Erdmann

22 Power Trade in a Pool Model

Power Producer 1

Power Producer 2

Power Producer 3

Power Producer 4

Power Producer 4

Power Retailer 1

Power Retailer 2

Power Retailer 3

Power Retailer 4

Day- ahead Power Pool

© Prof. Dr. Georg Erdmann

23 Problem 1: Possible Price Manipulation

© Prof. Dr. Georg Erdmann

24 „Cournot-Game“ under Merit Order

Price  before  capacity  withdraw  

Marginal  Cost  

Price  [EURO/MWh]  

Load  [MWh]  

Demand  Price  aBer  capacity  withdraw  

© Prof. Dr. Georg Erdmann

25 Problem 2: Free GHG Allowances

© Prof. Dr. Georg Erdmann

26 EUA Allocation and Power Prices

Em Total emissions of a facility [t CO2] g(Em) Free allocation of EUA [t CO2] C Cost function [Euro] p Power price [Euro/MW] pem Emission price [Euro/t CO2] Π Profit [Euro] Q Output [MWh]

( )( , ) ( , ) ( )emQ Em p Q C Q Em p Em g EmΠ = ⋅ − − ⋅ −

emC pEm∂

= −∂ em

C Emp pQ Q∂

= + ⋅∂

First order optimality conditions

© Prof. Dr. Georg Erdmann

27

(Clean) Dark Spread in the Day-ahead Market [Quarterly average; Germany; Source: BAFA; EEX]

0

10

20

30

40

50

60

70

2000 2002 2004 2006 2008

Coal cost [38% WG] Coal cost [38% WG] + 0,88 EUA

Phelix Offpeak

Marginal cost; electricity price [EUR/MWh]

© Prof. Dr. Georg Erdmann

28

(Clean) Spark Spread in the EEX Market [Quarterly average; German Day-ahead-market; Source: BAFA; EEX]

0

20

40

60

80

2000 2002 2004 2006 2008

Phelix-Peak Natural gas cost [η=55%] + 0,36 EUA Natural gas cost [η=42%] + 0,48 EUA

Marginal cost; electricity price [EUR/MWh]

© Prof. Dr. Georg Erdmann

29 Problem 3: Regulation and Balancing Power

Power Generator TSO Balancing Group Managers

Regulation Balancing

Power Power Pay-as-bid

auction

© Prof. Dr. Georg Erdmann

30 Types of Regulation Power in Germany

Pprim Primary reserve Psek Secondary reserve Pmin Tertiary reserve (Minute reserve)

Capacity

6 s 30 s

15 min

Psek Pmin

time

Psoll

Pprim

2 min 5 min

© Prof. Dr. Georg Erdmann

31 Allocation of Regulation Cost

CP included in the grid access fee

CP

EP

CP

EP

Charges to Balancing group managers M

inut

e re

serv

e

Sec

onda

ry re

serv

e

Prim

ary

rese

rve

Capacity price

Energy price

© Prof. Dr. Georg Erdmann

32 Problem 4: Clearing of Electricity Futures

© Prof. Dr. Georg Erdmann

33 Continuous Trading of Power Futures

Year Futures

Quarter Futures

Month Futures

Week Futures Day hour

Rolling

Rolling: When delivery period of a year ahead future begins, the contract splits into

•  3 monthly futures M0, M1 and M2, •  3 quarterly futures Q1, Q2 and Q3

Spot markets Forward / Futures markets

Standardized Futures don’t match with typical load schedules. Open positions must be closed on the day-ahead market or with balancing power (supplied by the TSO)

¼ hour

Rolling

© Prof. Dr. Georg Erdmann

34 Wholesale Power Prices [Source EEX]

0

20

40

60

80

2002 2004 2006 2008 2010 2012 2014

Year-­‐ahead

Phelix

Baseload  Strompreis-­‐Indizes [Euro/MWh]

I II III IV V

© Prof. Dr. Georg Erdmann

35 Types of Bilateral Trade Contracts

•  Spot trade: Delivery and payment follow immediately after the conclusion of the trade contract

•  Forwards: Conditions of the deal (quantity, quality, price, delivery and payment conditions, …) are fixed in the contract, but the contract is physically settled at a later date

•  Futures: like Forwards, but financial settlement with

Clearing (Interest → Future prices exceed Forward prices)

•  Call-Option: like Forward, but the buyer can withdraw from the contract

•  Put-Option: like Forward, but the seller can withdraw

Cash trade

Forward trade

Deriva-tive trade

© Prof. Dr. Georg Erdmann

36 Clearing of open positions

To cover the counterparty risk, each trader is obliged to sustain a margin account at the energy exchange (or the clearing house). In addition, the trader has to open a maintenance account and to transfer an initial margin that depends, among others, on the expected volume of his open positions

If a trader holds an open position and the futures price exceeds the contracted price, the clearing house transfers the price difference times contract size from the maintenance account to the margin account of the seller and from the margin account to the maintenance account of the buyer.

Debits and credits are adjusted at the end of each trading day according to the market price at closure

If the maintenance account drops below a maintenance margin, the trader has to remargin the maintenance accounts within short time. Otherwise all open positions of the trader are closed by the clearing house at market conditions

© Prof. Dr. Georg Erdmann

37 Types of Trades

•  Hedging: the aim of hedging is to ensure price risks. Hedgers are prepared to pay a risk premium

•  Speculation: Speculation is based on price expectations that may turn to be right or wrong. Speculators ensure the hedgers and thereby earn the risk premium

•  Arbitrage: the aim is to use price differences for achieving a risk free trade profit

•  Markets that offer no arbitrage are said to be in equilibrium

© Prof. Dr. Georg Erdmann

38 Problem 5: Lack of Power Plant Investments

© Prof. Dr. Georg Erdmann

39

100

80

60

40

20

0 20 30 40 50 60 70

Marginal costs of power generation [Euro/MWh]

December 2005

January 2003

Available capacity in Germany without wind power [1000 MW]

Estimated Merit Order for Germany [without CO2 costs; Source: London Economics 2007]

© Prof. Dr. Georg Erdmann

40

Are these returns sufficient for risk adequate financing of the power plant investment?

Power Plant Investments on Competitive Markets

0

20

40

60

80

0 1000 2000 3000 4000 5000 6000

Hou

rly d

ay-a

head

pric

es [E

uro/

MW

h]

Marginal costs of a new plant

Expected annual operating hours

Ordered price duration curve for the planning horizon

hours of the year

© Prof. Dr. Georg Erdmann

41 Ordered Day-ahead Prices in 2012

-30

10

50

90

130

Day-ahead price 2012: Ø = 42.60 €/MWh

Without wind: Ø = 48.06 €/MWh

without wind + PV: Ø = 51.13 €/MWh Δ = 8.53 €/MWh

Day-ahead price [EUR/MWh]

Hours 1000 2000 3000 4000 5000 6000 7000 8000

© Prof. Dr. Georg Erdmann

42

Ordered Day-ahead Price Level Necessary for Financing new GT

Day-­‐ahead  price  [EUR/MWh]

Hours -­‐30

30

90

150

210

42.60  €/MWh

51.13  €/MWh  at  500  h/a

52.50  €/MWh  at  1000  h/a 54.07  €/MWh  at  1500  h/a

55.80  €/MWh  at  2000  h/a

1000   2000   3000   4000   5000   6000   7000   8000  

© Prof. Dr. Georg Erdmann

43 Problem 6: New Market Design

© Prof. Dr. Georg Erdmann

44

Capacity surcharge

MCP with Capacity Shortages

Market Clearing Price [Euro/MWh]

Load [MW]

short-term marginal cost

Available capacity

Demand

© Prof. Dr. Georg Erdmann

45 Strategic Reserve

•  Similar to the strategic petroleum reserve: Some (new or old) gas turbines or coal plants are taken out of the normal market supply and scheduled according to the order of the regulator

•  Power capacity surplus shrinks in normal periods and increases in critical periods. Who decides about what periods are critical?

•  The plants in the strategic reserve receive a capacity payment financed through the grid fees

•  Who decides and according to what rules about which plants should be allocated to the strategic reserve?

© Prof. Dr. Georg Erdmann

46 Decentralized Capacity Market

•  No market participant shall withdraw power from the grid without a capacity allowance which has to be purchased

•  PJM Market: Distributed power (such as wind power) can be fed-in and taken out from the distribution grid so that no capacity allowance is necessary

•  Alternative: Any power generator that feeds-in electricity must be in the possession of the appropriated number of fed-in allowances

•  How to prevent the capacity market from not being manipulated (market power issue)?

•  How to control?

© Prof. Dr. Georg Erdmann

47 Other Capacity Market Design Options

•  Comprehensive capacity markets: All plants that can be dispatched according to the demand receive a capacity premium (cash flow to existent and new plants)

•  Focused capacity markets: Certain new plants that have been selected through a competitive bidding receive a capacity premium (cash flow to new plants)

© Prof. Dr. Georg Erdmann

48 Price Elasticity of Demand

0

10

20

30

40

50

0 10′000 20′000 30′000 40′000 50′000

Market Clearing Price MCP [Euro/MWh]

Available generation capacity [MW]