Australian Energy Market Commission Congestion Management Review and related projects

1
Australian Energy Market CommissionCongestion
Management Review and related projects

• ACCC Regulatory Conference
• Thursday, 26th July 2007

2
Context

• AEMC presently finalising
• Congestion Management Review (CMR) Draft Report
  in response to MCE ToR
• 3 participant proposals concerning network
  congestion around Snowy mountains area
• Regional boundary change criteria
• Presenting Commissions framework and approach to
  congestion issues
• Combines theoretical insights with practical
  policy and implementation considerations

3
Background

• The National Electricity Market (NEM)
• spans the interconnected electricity transmission
  system from Qld to SA
• is an energy-only market (no capacity payments)
• is a gross market (all transactions settled
  centrally)
• prices set within each region (States plus Snowy)
  at each regional reference node (RRN) (capital
  cities)
• prices are relatively volatile in this form of
  market design
• market participants need hedging instruments ?
  critical for promoting competition and investment
• AEMCs role set out in National Electricity Law
• to undertake reviews and assess Rule change
  proposals in accordance with the NEM objective
• but not to initiate changes to the Rules

4
NEM objective

• Section 7 of the NEL - incorporates
• productive, allocative and dynamic efficiency
• (broader than technical efficiency of dispatch)
• reliability, security and safety
• good regulatory practice transparency and
  predictability of change or intervention
• Distributional outcomes not a discrete component
• AEMC particularly concerned about liquidity of
  derivative trading and investment impacts

5
Network congestion

• Reliable electricity supply requires SD at every
  point in time
• NEM dispatch algorithm minimises cost of supply
  based on bids and offers submitted
• Congestion occurs when transmission limits
  prevent least-cost dispatch
• Can increase transmission limits by investment or
  more efficient network operation
• But eliminating all congestion is unlikely to be
  efficient the issue is management

6
Pricing congestion (1)

• The NEM, as a generalisation
• prices some congestion (between RRNs), providing
  locational signals to loads and generators
• does not price congestion within regions
• Dispatch algorithm compares a generators offer
  to its nodal shadow price to determine
  dispatch
• IFF generator offer priceltNSP, plant dispatched
• Where congestion is unpriced, dispatch and
  settlement are based on different prices
• dispatch algorithm based on shadow nodal prices
  (which reflect impact of congestion)
• settlement price does not reflect impact of
  congestion (based on RRPs, adjusted by losses)

7
Pricing congestion (2)

• Disjoint between dispatch and settlement
• Can lead to dispatch (volume) risk
• generators that relieve constraint may be
  constrained-on (despite their offer pricegtRRP)
• generators that worsen constraint may be
  constrained-off (despite their offer priceltRRP)
• Incentivises disorderly bidding
• constrained-on generators may bid 10,000/MWh
• constrained-off generators may bid -1,000/MWh
• Disorderly bidding
• does not necessarily lead to significant economic
  costs and welfare losses
• is not necessarily indicative of market power
  being exercised

8
Pricing congestion (3)

• More location prices for generators (and loads?)
• Overcomes mis-pricing and disorderly bidding
• But
• more price variation risk (basis risk) for
  market participants to manage
• hence, requires basis risk management instruments
  (NEM uses Settlement Residue Units)
• Deciding how many more prices are needed
• Trade off between purity/correctness of prices
  and complexity of trading and risk management
• Also need to consider incentives for the exercise
  of market power.

9
Current AEMC work program

• Rule change proposals to address congestion in
  Snowy region, including region boundary changes
• Determinations due in August
• Rule change proposal to introduce new process for
  future region boundary change
• based on economic, not technical criteria
• Draft Determination due in August
• Congestion Management Review directed by MCE
• ToR requires interim regime to enable
  participants to manage risks of material and
  persistent congestion prior to boundary change
• AEMC approach therefore based on proportionality
  between problem/solution
• All strands of work involve assessments of
  materiality

10
Materiality of congestion

• Materiality was examined in the context of both
  physical incidence and economic cost
• AEMC considered evidence from Dr Darryl Biggar,
  NEMMCO and other sources
• Update analysis by NEMMCO (June 2007)
• Distinguishes between system normal and outage
  constraints (which might involve different policy
  response)
• System normal incidence is at fairly constant
  average level
• But average hides variations in impacts and
  changing locational patterns over time (evidence
  of life cycle of constraints)
• Raises implementation questions for localised
  intervention to manage congestion

11
Economic costs of mis-pricing

• NEM objective requires AEMC to focus on economic
  impacts of congestion
• Range of views from market participants, but
  majority view issue as second-order relative to
  other issues (e.g. water, emissions)
• AER found annual total cost of congestion (TCC)
  about 30-70m
• AEMC-commissioned modelling work of dispatch
  inefficiencies in system normal conditions found
  8m annual cost
• Range of views from stakeholders on longer-term
  impact of mis-pricing on investment

12
Options for change?

• Non-pricing incremental change options
• Reforms to the Settlement Residues Auctions (e.g.
  to firm up the hedges available)
• Sharper transmission business incentives for more
  efficient operation (e.g. outage planning)
• More/better information provision to the market
  from NEMMCO and transmission businesses
• COAG reference on the need for a national
  transmission planning body might drive further
  reform

13
Pricing options?

• All involve more localised pricing of generation
  in certain circumstances (ie when constraints
  bind)
• How are market participants able to hedge the
  resultant risk?
• Established instruments for the NEMs regional
  structure
• What new instruments are needed for more
  localised pricing?
• How are they allocated/sold?
• Change raises practical implementation issues
• Threshold/trigger for initiation
• Which constraints to price?
• Duration/threshold for removal
• (Toughest!) Allocation of rental rights (and
  obligations) comes down to choice between some
  kind of grandfathering and auctioning
  potentially large and complex impacts on market
  power and barriers to entry

14
Conclusion

• Congestion management involves applying
  theoretical insights in a real-world setting
  not a straightforward process!
• Raises complex practical implementation issues
  and trade-offs
• Draft Report intended for August 2007