Towards a Better Design of Electricity Transmission Rights - a Monte - - PowerPoint PPT Presentation

▶
towards a better design of electricity transmission
SMART_READER_LITE
LIVE PREVIEW

Towards a Better Design of Electricity Transmission Rights - a Monte - - PowerPoint PPT Presentation

Jul 20, 2023 141 likes •344 views

Towards a Better Design of Electricity Transmission Rights - a Monte Carlo based option valuation Matthias Janssen (with Thomas Niedrig and Magnus Wobben) 10 th INFRADAY Conference, Berlin, 08 Oct 2011 Imagine you operated a power plant in

slide-1
SLIDE 1

Towards a Better Design of Electricity Transmission Rights

  • a Monte Carlo based option valuation

Matthias Janssen (with Thomas Niedrig and Magnus Wobben) 10th INFRADAY Conference, Berlin, 08 Oct 2011

slide-2
SLIDE 2

2 Frontier Economics

Imagine you operated a power plant in Germany… …and wanted to provide electricity for a factory in NL

Transmission capacity is limited You‘ll need a transmission right (TR)  auction How much would you be willing to pay?

slide-3
SLIDE 3

3 Frontier Economics

Transmission rights are options on hourly price spreads

  • Your power plant operates baseload  physically long
  • You have a TR for the next month
  • You sell a base contract to a factory in NL for next month

 long position closed Use historic spot price data to simulate the spot price characteristics Use all available market data that reveal price expectations for the delivery month Lesson Value of the TR = sum of cross-border spot price spreads during next month! Spot prices in the future unknown and no traded expectations (forward products) on hourly spot prices… A B In hours where PNL ≥ PGE  use TR In hours where PNL < PGE  don‘t use TR

  • Sell in Germany
  • Buy in NL

Problem Example

slide-4
SLIDE 4

4 Frontier Economics

  • Data
  • Model
  • Results
  • Conclusion
slide-5
SLIDE 5

5 Frontier Economics

Data

11 monthly PTR auctions November 2009 to October 2010 Data sources: APX, EEX, Endex, Entso-E, Casc EU

Available IC capacity

  • Explicit auction
  • Yearly TRs
  • Explicit auction
  • Monthly TRs

Explicit /Implicit day-ahead auction

slide-6
SLIDE 6

6 Frontier Economics

  • Data
  • Model
  • Results
  • Conclusion
slide-7
SLIDE 7

7 Frontier Economics

‐1000 ‐800 ‐600 ‐400 ‐200 200 400 600

01/2007 04/2007 07/2007 10/2007 01/2008 04/2008 07/2008 10/2008 01/2009 04/2009 07/2009 10/2009

Price difference Germany ‐ Netherlands (€/MWh) ‐1000 ‐800 ‐600 ‐400 ‐200 200 400 600

01/2007 04/2007 07/2007 10/2007 01/2008 04/2008 07/2008 10/2008 01/2009 04/2009 07/2009 10/2009

Price difference Germany ‐ Netherlands (€/MWh)

Timeline of used data

Example: Auction for November 2009 transmission right

Delivery month November 2009 2 preceding years for spot price modeling

TR auction in preceding month (21 October 2009) Price difference of month- ahead futures at the time

  • f TR auction

…same procedure for all 11 monthly PTRs

A B

Monte Carlo simulation

C

slide-8
SLIDE 8

8 Frontier Economics

30 35 40 45 50 55 60 65 70 75

MO TU WE TH FR

€/MWh

Estimation of stochastic spot price process Pt

Pt = f(t) + Yt + Xt

Diffusion process Xt Jump process Yt Deterministic part

Deterministic part Jump filtration Diffusion estimation A 1 2 3 1 2 3

20 40 60 80 100 120 140

MO TU WE TH FR

€/MWh

Estimate season

‐30 ‐20 ‐10 10 20 30 40 50 60 MO TU WE TH FR

€/MWh

20 40 60 80 100 120 140 MO TU WE TH FR

€/MWh

‐25 ‐20 ‐15 ‐10 ‐5 5 10 15 20 MO TU WE TH FR

€/MWh

Entire timeseries* Estimate jump process dYt = -βYt dt + Ju dNu – Jd dNd, dXt = -κXt dt + σ dWt Deseasonalise process Substitue spikes

Monte Carlo simulation

For visualisation purposes: Exemplary spot price work week (not price spreads).

slide-9
SLIDE 9

9 Frontier Economics

Estimating the deterministic component f(t)

Now: Cross-border spreads GE-NL

A 1

‐40 ‐30 ‐20 ‐10 10 20 30 40 50

Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday

€/MWh

Exemplary: Spreads GE-NL September 2009

Systematic pattern in spreads?

  • No overall level ≠ 0
  • No trend
  • No systematic pattern (season)

□ No season of the year □ No week/weekend pattern □ No pattern during the day (e.g. peak/off-peak)

 Nothing that reveals information for spot price spreads during the forthcoming delivery month  f(t) = 0 Use in MC simulation

slide-10
SLIDE 10

10 Frontier Economics

Filtering and estimating jumps

A 2

Estimated parameter

Frequency of neg. jumps 3.38% Mean size of neg. jumps

  • 10.12 €/MWh

Frequency of pos. jumps 3.94% Mean size of pos. jumps 9.38 €/MWh Mean reversion speed β 1.03

Identify jump increments

(increments > threshold)

Substitute jump increments

Repeat procedure until no more jumps remain Calibrate threshold such that remaining diffusion process Xt is characterised by normally distributed increments

Iterative jump filter Feed into next step Use in MC simulation

‐25 ‐20 ‐15 ‐10 ‐5 5 10 15 20 MO TU WE TH FR €/MWh 20 40 60 80 100 120 140 MO TU WE TH FR €/MWh

  • Positive jumps
  • Negative jumps

Maximum-Likelihood estimation of jump processes Yt

slide-11
SLIDE 11

11 Frontier Economics

Estimating the residual diffusion process Xt

A 3 Maximum-Likelihood estimation

Estimated parameter

Mean reversion speed κ of diffusion process 0.24 Volatility σ of diffusion process (€/MWh) 3.03

Increments approximately normally distributed after filtration Use in MC simulation

‐25 ‐20 ‐15 ‐10 ‐5 5 10 15 20 MO TU WE TH FR

€/MWh

slide-12
SLIDE 12

12 Frontier Economics

Integrating information on expected price spreads

Forwards on hours Forwards on month Use in simulation

No forward products traded on hourly prices or spreads Still, spread in prices of month-ahead base forward reveal information about expected spot price spreads on average For every TR,

  • we set the deterministic part f(t) to the month-ahead base

future price spread

  • traded on the day of the TR auction

B

Example: for TR Nov 2009:

  • f(t) = 0.51 €/MWh
  • Traded on 21 Oct 2009
slide-13
SLIDE 13

13 Frontier Economics

Monte-Carlo (MC) simulation

‐200 ‐150 ‐100 ‐50 50 100

Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday Wednesday Thursday Friday Saturday Sunday Monday Tuesday

€/MWh MC‐Path 1 MC‐Path 2 MC‐Path 3 MC‐Path 4 Payoff in a MC path = sum of all positive hourly spreads

MC simulation Simulation of 10,000 simulation paths by generating random numbers according to the estimated parameters TR value Value of transmission right = average of all 10,000 payoffs C

slide-14
SLIDE 14

14 Frontier Economics

  • Data
  • Model
  • Results
  • Conclusion
slide-15
SLIDE 15

15 Frontier Economics

2 4 6 8 10 12 14 16 ‐15 ‐10 ‐5 5 10 15 Transmission right value in €/MWh Price spread month‐ahead futures in €/MWh Model results Intrinsic Value Auction results

Modelled values > auction results

Modelled TR values exceed actual auction results by far!

Day-ahead

Day-ahead auction 12 am

Delivery day

0 am 12 pm 8 am Nomination of long-term PTRs

Potential institutional reasons

Non-firmness of tranmission rights Timing of option exercise

Now we take nomination uncertainty into account

TSOs with the right to curtail transport if available transmission capacity is lower than expected

Even though: Actually realisable profit (ex-post) = 3,4 €/MWh

slide-16
SLIDE 16

16 Frontier Economics

Are PTRs underprice?

1 2 3 4 5 6 ‐5 ‐4 ‐3 ‐2 ‐1 1 2 3 4 5 Transmission right value in €/MWh Price spread month‐ahead futures in €/MWh Model results with "perfect option exercise" Model results with "best guess option exercise" Intrinsic Value Auction results

Assumption No payoff in case of non-nomination  Physical transmission right (PTR) with UIOLI* Current European market design

* UIOLI = Use-it-or-lose-it  we assume that a potential manual resell does not reveal any payoffs.

  • Non-firmness of

PTRs

Institutional reasons

  • Constant volatility
  • …

Model failures Potential reasons

  • Players don‘t value PTRs as
  • ptions  buy PTRs as

they‘re a bargain!

Market preferences PTRs with “best guess exercise” worth much less Actual PTRs still cheaper than modelled PTRs with “best guess exercise”

slide-17
SLIDE 17

17 Frontier Economics

  • Data
  • Model
  • Results
  • Conclusion
slide-18
SLIDE 18

18 Frontier Economics

In short…

What did we learn? Further research

  • Valuation of hourly exercisable transmission rights

is challenging, but possible

  • Nomination uncertainty depresses the option value

significantly  Use-it-or-sell-it or Financial TRs are preferrable

  • PTRs could well be underpriced

 in that case there are arbitrage potentials

  • Improve the model
  • Analyse other borders
slide-19
SLIDE 19

19 Frontier Economics Frontier Economics Limited in Europe is a member of the Frontier Economics network, which consists of separate companies based in Europe (Brussels, Cologne, London and Madrid) and Australia (Melbourne & Sydney). The companies are independently owned, and legal commitments entered into by any one company do not impose any obligations on other companies in the network. All views expressed in this document are the views of Frontier Economics Limited.

slide-20
SLIDE 20

20 Frontier Economics FRONTIER ECONOMICS EUROPE LTD. BRUSSELS | COLOGNE | LONDON | MADRID Frontier Economics Ltd, 71 High Holborn, London, WC1V 6DA

  • Tel. +44 (0)20 7031 7000 Fax. +44 (0)20 7031 7001 www.frontier-economics.com