25. New Zealand continues to rely heavily on hydro which accounts for 64 percent of total generation. Approximately two thirds of hydro generation is in the South Island, and most is in the lower half of the South Island. However, 63 percent of demand is in the North Island. Electricity is transported on a long, thin transmission grid. Given this geographic configuration, losses and constraints are particularly significant in New Zealand.

26. Our heavy reliance on hydro and very limited storage mean that the impact of a "dry year" on electricity supply can be quite severe. Furthermore, during a dry year, changing generation patterns can cause unusual transmission constraints (particularly when the HVDC link is reversed to run north to south), which exacerbate supply shortages.

27. The market, assisted by a government publicity campaign to reduce demand, avoided supply disruptions in 2001 despite extreme hydrological conditions (although under-hedged companies suffered severe financial consequences). This has not always been possible as the following table from the Government's 1992 Electricity Review indicates:

Period	Issue
1942	Use of space heaters and radiators prohibited during peak hours between May and August in the North Island
1943	Space heating was controlled further, and so too were indoor and outdoor lighting
1946	20% power cuts imposed in the North Island
1947	30% power cuts imposed in the South Island
1958	A 15% cut was imposed in the North Island
1973	Supply authorities were requested to save up to 6%. Broadcasting hours were reduced, ripple control was increased and in some areas there were daily blackouts.
1974, 1975, 1976, 1977	Government requested "voluntary" savings
1992	Large publicity campaign used to reduce demand. Water heating was generally cut for 18 hours/day, and Comalco closed one of its three potlines (= 5% of national demand)

Source: Appendix 3, Report of the Electricity Shortage Review Committee 1992

Weaknesses/Deficiencies in the New Zealand Electricity Market

28. Officials consider that the spot market worked much as expected during winter 2001, with very high prices signalling an increasingly tight supply situation (record low hydro inflows) and record demand.

29. A number of submissions alleged that generators had exercised market power and that as a result spot prices were excessively high. The allegations included that :

thermal generation at Huntly and New Plymouth was not run as hard as it could and should have been in April and May in response to rising spot prices and that this led to excessive draw-down of lakes: the alleged motivation was to benefit from higher than necessary spot prices and disadvantage competing retailers (who were known to be under-hedged); and
generators, and especially Meridian, bid up spot prices unnecessarily high, but waited until they knew On energy's hedges expired at the end of May, again for the purpose of squeezing On energy and other net retailers out of the market.³

30. Officials have not undertaken a detailed analysis of these allegations, but note that generators and the NZEM made a number of valid points in response, including that:

thermals did run a lot harder in autumn than normal; that the short run marginal cost (SRMC) of the last available units at Huntly and New Plymouth are much higher than for the first units (i.e. that relatively high prices are required to induce them to run); and that transmission constraints, especially out of Taranaki, limited the availability of thermal generation until Transpower acted to reduce the constraints;
that the supply curve (i.e. the cost of additional generation) in New Zealand rises very sharply after Huntly and New Plymouth because very little backup generation is held only for dry years: this means that, when normal generating capacity is fully committed and the lakes are running low, prices have to rise sharply to induce additional supply, such as diesel generators and open cycle cogeneration; and
that the Market Surveillance Panel of NZEM investigated similar allegations in detail at the beginning of winter and found them not proven, although it did point out that temporary market power existed from time to time.

31. The events of the past winter have however revealed a number of areas of concern including:

inadequate use of financial risk management instruments;
inadequate demand management;
inadequate transmission investment; and
reduced retail competition.

Inadequate Use of Financial Risk Management Instruments

32. Some major retailers and large users went into the winter period seriously under-hedged against dry-year spot prices, notwithstanding a Government warning in December 2000 that the New Zealand electricity system is inherently volatile and subject to dry-year risk and that the Government would not step in to protect parties who failed to manage risk.⁴

33. Hedges for 2001 were available to buyers over the summer period (which is the optimal season for hedging for winter). It appears however that several years of warm wet winters, surplus generating capacity and record low spot prices affected buyers' assessments about investing in hedges.

34. One of the consequences of some buyers declining to purchase hedges over summer is that net generators were incentivised to acquire their own customer base to act as a natural hedge. In the absence of hedges generators are exposed to low spot prices in warm wet winters.

35. A further concern is the relative lack of sophisticated and tradable risk management instruments in the market.⁵ Note however, that vertical integration in the market has decreased the need for these forms of risk management, as the combination of generation and retail businesses achieves the same objective.

Inadequate Demand Management

36. The lack of demand-side participation in the market is also of concern. Effective demand-side participation should be a characteristic of an efficiently performing market especially when supply is constrained. The past winter demonstrated that many retailers did not have effective systems in place to incentivise reduced demand. It also showed that the design of contracts (fixed price - variable volume) inhibited the creation of a demand-side secondary market (that is, a market which pays larger consumers to reduce demand). Although a demand-side driven auction facility was created by d-cypha (a subsidiary of Transpower) during the winter, it was ineffective for the above reasons.

Inadequate Transmission Investment

37. The existence of transmission constraints reduces the amount of generation available to meet demand nationwide and inhibits retail competition behind the constraint. It would be beneficial for Transpower to have a clearer mandate from market participants to relieve constraints by reducing security standards when the system is under stress. It may also be beneficial in many instances to upgrade the grid to relieve constraints. However the New Zealand market currently lacks an agreed approach for determining the most economically efficient level of transmission investment and how it will be funded. (The Government Policy Statement of December 2000 addresses this issue: see paragraph 48.)

38. In addition, an insurance mechanism to cover transmission losses and constraints is also desirable. Although work is proceeding on the development of financial transmission rights (FTRs), they have yet to be implemented.

Reduced Retail Competition

39. A consequence of the winter events is that there is now significantly less retail competition in the market, especially with the demise of On energy and the take-over of its customer base by Genesis and Meridian.⁶ The retail market is now dominated by five retailers, four of which have significant generation assets. Three of the four vertically integrated generators/retailers are regionally based (the exception is Contact). There is a risk that these companies will tend to focus on their own region and that only weak retail competition will result. This risk is significantly increased by the existence of transmission constraints and the lack of financial instruments to manage transmission risks. Vertical integration also means that potential new retailers may be put off by the need to buy hedges from their competitors.

Major Participants in New Zealand's Electricity Generation and Retail Markets

Market Participant	Retail Market Share (% of customers)	Generation Capacity (% of total capacity)
Meridian Energy Ltd	13	28
Genesis Power Ltd	26	19
Mighty River Power Ltd	16	13
Trust Power Ltd	16	5
Contact Energy Ltd	22	25
Natural Gas Corporation	N/A	6
Todd Energy	4	2
Other	3	2
Total	100	100

N/A - Not applicable. *Market share by retail customer numbers may differ from market share by retail load due to the size and mix of customers supplied by participants.

Source: Adapted from information provided by M-co.

Options for Strengthening Performance

Allow the Market to Self-Correct

40. A number of submissions urged caution in pursuing any fundamental changes to the market. They noted that the market, while immature, is evolving to operate more effectively (it has been fully competitive only since April 1999).

41. The last three years have also been characterised by excess supply and average demand (good hydro storage, new thermals on line and mild winters). Prior to this year, market players had not yet been through a full range of supply/demand circumstances: even though "dry-year" risk had been signalled by government policy statements it had not been experienced under market arrangements. The market in that short time has also had to adapt to extensive structural reform especially in ownership of retailing. This structural reform was still bedding down in the lead up to winter 2001.

42. Furthermore the market has had only since December 2000 to adjust to Government policies arising from the Ministerial Inquiry and since August 2001 to see the final legislative form which enforces these policies. The 2001 "dry year" occurred prior to legislative implementation of these policies.

43. Officials believe that the existing market structure will self-correct to a considerable extent. As a consequence of 2001, much more realism about the need to hedge against dry-year risk and much more sophisticated (and liquid) hedging arrangements are likely to emerge. It is also likely that larger consumers will improve their ability to reduce consumption in response to high wholesale prices.

44. The introduction of enhanced information disclosure is evidence of market response. The NZEM announced in November that it has introduced a free web-site which will provide information on wholesale electricity prices, electricity demand and hydrology for hydro storage. Consumer groups have welcomed this development, although consider it does not go far enough.

45. New Zealand is facing the need to build new generation capacity in the next few years to meet rising demand. This means that wholesale market prices on average are likely to trend towards long run marginal cost (LRMC). This is set by the cost of new generating capacity and is likely to be around 5c/kWh. Officials consider there is no evidence at this stage that wholesale market prices will rise above LRMC on a sustained basis (this would signal the existence of ongoing market power in generation). Officials also note that a number of companies have signalled their intention to build new capacity. Even with wholesale prices at LRMC, New Zealand's retail electricity prices would continue to be low compared with most other countries.

46. In addition, it appears that retail margins are very thin (or negative in some instances). This is unsustainable, and accordingly there is a high likelihood of upward movement in retail prices.

47. Officials are of the view that the market is likely to adjust to a significant extent but higher prices and reduced competition will be a feature.

Accelerate Implementation of the Government Policy Statement

48. The 2000 Ministerial Inquiry into the Electricity Industry concluded that market arrangements and industry self-governance would best deliver the outcomes sought by the Government. However it identified a number of aspects of the market which required improvement. The Inquiry's recommendations were generally accepted by the Government which issued a detailed Government Policy Statement (GPS) in December 2000.

49. The GPS spelt out the Government's overall objectives for the industry and requested the industry to set up an Electricity Governance Board (EGB) as soon as possible. The Government called on the EGB (and the industry) to make a number of significant improvements to the way the market operated, including:⁷

projections of system adequacy;
disclosure of hydro spill;
disclosure of forward hedge prices;
disclosure of generator offer prices into the market (after three months);
development of real time spot market pricing and promotion of demand-side participation;
facilitating investment in distributed generation;⁸
arrangements for setting agreed transmission prices;
arrangements for agreeing on and paying for new transmission investments to relieve constraints; and
development of financial instruments to manage transmission risk.

50. An Electricity Governance Establishment Project (EGEP) chaired by Hon David Caygill has made good progress in a number of areas, although none of the above measures are in place yet. Although EGEP is responsible for some workstreams, Transpower, M-co, the NZEM Rules Committee and the MARIA Governance Board are lead agencies for others. The Minister of Energy has met regularly with the chair of EGEP to monitor and reinforce the need for speedy progress.

51. Officials consider that the market overall would have worked better had the reforms specified in the GPS been fully implemented. It is proposed that the Minister of Energy sets specific milestones for the completion of the above objectives and monitors progress. Reports on progress should be requested from the responsible lead agency and be made publicly available. Presently, two-monthly reports are requested of EGEP only.

Earlier Disclosure of Spot Market Offers

52. Many submissions urged the Government to require early disclosure of generator offers into the market to act as a check on the exercise of market power and to better inform the demand side about price trends. As noted above, the GPS called on the industry to disclose this information after a three month period. This has not yet been implemented although it is being considered by an NZEM working group.

53. The issue arises as to whether disclosure earlier than three months would be desirable. Early disclosure risks facilitating collusive and gaming behaviour by the generators, but significantly delayed disclosure reduces the pressure on generators not to take advantage of impending supply shortages by driving up prices (by holding back capacity or offering high). Officials have a split recommendation on this issues.

54. The Ministry of Economic Development (MED) considers that, on balance, there would be net benefits from earlier disclosure particularly when there is a risk of dry years. It is generally considered that generators are already able to deduce the pricing behaviour of their competitors (offering takes place half hourly, and generators know which stations are dispatched, so it is not difficult for them to work out each others pricing behaviour). This means that early public disclosure is unlikely to increase the risk of collusion between generators. On the other hand, other parties including consumers do not have access to this information, so are unable to put pressure on generators to justify their offers if and when there are indications of gaming behaviour.

55. Accordingly, MED recommends that the GPS be amended to request disclosure of generator offers two weeks in arrears and that the Minister request NZEM to implement this as soon as possible. Any amended GPS should be transmitted by the Minister of Commerce to the Commerce Commission as a statement of Government Policy in terms of section 26 of the Commerce Act 1986.

56. The Treasury notes the timing of the release of offer information is an important factor governing the effective operation of the market. A concern that collusion might occur led to the GPS requiring offer information to be disclosed 3 months in arrears, compared to the recommendation of the Ministerial Inquiry that offer information be disclosed one month in arrears.

57. Given this risk, the Treasury is unable to support the proposal that offer information be released 2 weeks in arrears without further work to assess the possible effect on collusion and the compliance costs associated with any changed disclosure requirements. The Treasury considers that because of the importance of the timing of the release of offer information and the process surrounding the development of the GPS in the first instance,⁹ a decision should not be taken to amend the requirements in the GPS without this further work.

58. As an alternative to the MED proposal, the Treasury suggests that officials report back early in the New Year with recommendations on whether offer information disclosure requirements contained in the GPS should remain at 3 months.

Separation of Retail and Generation

59. Many submitters argued that the separation of retail and generation would significantly increase the level of retail competition. However others noted that this proposal would involve major disruption to the industry and the benefits of vertical integration would be lost.

60. The benefits of separation are as follows:

facilitates retail competition. With vertical integration, entrant retailers have to purchase hedges from integrated generator/retailers which are also competitors. The new retailers will be concerned they will have difficulty obtaining hedges at competitive prices or that they will be more vulnerable than integrated competitors to high spot prices in a dry year for any unhedged demand;
facilitates the development of risk management instruments and a more liquid trading market. The more retailers in the market the more likely it is that tradable hedge products will develop; and
counters any tendency to regionalisation of competition. The generating capacity of three of the four dominant generators/retailers is concentrated regionally (the exception is Contact), leading to incentives to focus retail activities in those regions and to weak competition (by staying out of each other's territory). Note that this tendency is exacerbated by transmission constraints and the lack of instruments to hedge against transmission risk. Full implementation of the GPS reforms will largely remove these latter factors, although this may take several years.

61. The disadvantages of separation are a loss of the benefits of vertical integration. These include:

improved stability and reduced risk in the retail market as ownership of retail customers has the characteristics of a hedge of longer duration than typical financial hedges. Significant problems were experienced this year with under-hedged retailers that got into difficulties, exposing their customers to the spot market. Vertical integration minimises this risk;¹⁰
lower costs from reduced transaction costs, scale economies and lower cost of capital (because integrated companies are less risky); and
greater certainty for investment in new generation. Integrated companies are likely to find it easier to obtain funding for new generation projects compared to stand-alone generators which will need to secure firm long-term contracts.

Types of Separation

62. There are two main types of separation: corporate and ownership.

63. Corporate separation requires separate corporate structures (boards and accounting) and arms-length transactions.¹¹ The limitation with corporate separation is that the holding company still has the incentive to operate its subsidiaries in a way which benefits both companies and which makes life difficult for competitors. Considerable regulatory effort is required to counter this incentive and to deliver the outcomes sought by separation.

64. Furthermore, to ensure corporate separation does in fact facilitate entry, it is likely to be necessary to require the generator to offer a minimum quantity of hedges to non-affiliated retailers at the same price it provides hedges to its affiliated retailer, or to tender a minimum quantity of hedges. This raises the issues covered below under compulsory hedge offers.

65. Ownership separation is likely to be much more effective in achieving the objectives of separation. Crucially, ownership separation, unlike corporate separation, changes incentives, and generators would have a strong incentive to contract widely with independent retailers to spread risk. However, ownership separation would involve implementation disruption and would breach the property rights of privately owned companies (notably Contact and TrustPower), and so would require legislation.

66. Some submitters advocated ownership separation (or possibly some form of corporate separation) be confined to the three SOE generators/retailers. This could possibly be implemented under the SOE Act without further legislation, but the option is not favoured by officials as it would give Contact and other private sector companies a considerable market advantage.

Compulsory Hedge Offers

67. Another option raised by a significant number of submitters to facilitate increased retail competition and improve risk management generally was to require generators to tender a minimum percentage of their dry-year capacity in tradable long-term hedges on an ongoing basis.

68. A requirement of this nature was placed on ECNZ in 1995/1996 when Contact was separated out to ensure initial market stability and to reduce incentives to exercise market power in the spot market.¹²

69. Key design issues of mandatory hedging arrangements include:

what quantity should be offered? Options are:

a set percentage (e.g. 10 percent) of a generator's dry-year capacity;
the net unhedged position of the generator/retailer (e.g. the difference between, say, 95% of its dry-year capacity and its contracted position including the demand of its own retail base); or
its gross position (say 95 percent of its dry-year capacity);

should all generators be obliged to tender hedges (including, for example, TrustPower and NGC) or only those which may have market power?
what tradability specifications should be imposed? Tradability is essential to obtain the maximum benefit from mandatory hedging, but complex technical issues are involved given differing demand profiles and offtake locations and the need to address counter-party risk for generators (i.e. to ensure the creditworthiness of third parties to whom hedges are sold);
should affiliated retailers be able to bid? The answer to this would probably be "no" for the 10 percent and net generator options noted above, but "yes" (probably up to a set percentage) for the gross generation option;
should there be a cap on purchases by any one party? Should large users (in addition to retailers) be allowed to bid?
what reserve prices should be set? Reserve prices should probably be set at SRMC, but this is a contentious calculation, especially for hydros (where SRMC is essentially opportunity cost and differs substantially depending on hydrology and time of year);
what is the quantity of dry-year capacity? This would involve assessing the reliability of thermal plants and the availability of fuel contracts; and
what is the best timing and frequency of auctioning? Who should undertake the auctions? Should bidders pay what they bid or the market clearing price?

70. None of these technical issues are insurmountable but they are complex and contentious, and a decision-making authority would be required. Considerable lobbying would be likely.

71. The benefits of mandatory hedging include:

facilitating entry by independent retailers (because their prospects of acquiring a profile of hedges at market rates would improve);
facilitating a more robust and liquid hedge trading market. (Note that the more liquid the hedge trading market, the more likely it is that market participants will be willing to hedge and to be non-integrated generators or retailers because they will be able to trade out of untenable positions. This in turn results in a less risky and more competitive market); and
increased confidence that wholesale (and retail) prices reflect supply and demand rather than any exercise of market power.

72. The disadvantages of mandatory hedging include:

implementation would be technically complex and contentious. Primary legislation, regulation and a regulator may be required, particularly for the more intrusive options. While the industry EGB could be asked by the government to develop such a scheme, it may be opposed by the industry (which is dominated by integrated generator/retailers); and
it may reduce the incentives on parties to develop their own risk management instruments and customised hedging arrangements. (Note however that the 10 percent option would minimise this risk.)

Conclusions on Market Design

73. Officials consider that the market will self-correct to a considerable extent. They recommend that the Government re-emphasise its commitment to the GPS, particularly policies relating to dry-year risk. Officials are split on whether a change should be made to the disclosure requirements in the GPS so that spot market offers are publicly disclosed within 2 weeks (rather than 3 months).

74. Officials consider that the current level of risk does not merit pursuing the vertical separation and mandatory hedge options at this time. However, officials consider that the retail situation should be monitored. In addition, the Government should send a strong message to generators/retailers that it wants to see continued retail competition. The Government should warn the incumbents that it will consider further measures, including mandatory tendering of hedges and separation of retail and generation businesses, if there is a little or no retail competition in significant parts of the country on an ongoing basis.

75. If monitoring suggests that further Government action is required, then the option to require all generators to tender 10 percent of their dry-year capacity on an ongoing basis may warrant further investigation. It is possible that this option could be implemented without formal government regulation. The GPS could be amended to request the industry and the EGB to implement such a scheme.

National Energy Efficiency and Conservation Strategy (NEECS)

76. Energy efficiency and conservation measures in the NEECS include:

greater demand-side participation in the wholesale electricity market; and
effective mechanisms for optimising the contributions of demand-side management, investment in transmission capacity and investment in new generation.

77. The strategy indicates that the Ministry for the Environment and EECA, in conjunction with the MED and Treasury, will work with the EGB and industry players to assist policy development and implementation in these areas.

Energy Efficiency Campaigns

78. A majority of submitters believe that the Government and EECA campaign was well run and successful. These submitters believe that as the market develops, this type of intervention should not be necessary - it should be embedded in the demand-side management of the market.

³The Commerce Commission is investigating complaints of anti-competitive behaviour under the Commerce Act.

⁴Similar government warnings had been issued on 8 June 1995 and 15 December 1998.

⁵This includes an absence of hedging contracts between hydro generators and generators with reserve thermal generation (these contracts would allow hydro generators to operate less conservatively, encourage early operation of reserve thermals, and help stabilise prices).

⁶Whether this take-over breaches the thresholds in the Commerce Act is currently being investigated by the Commerce Commission.

⁷In addition to these measures, the GPS called for a significant number of improvements to the retail and downstream markets, such as a consumer complaints scheme, model consumer contracts, pre-payment meters, model distribution pricing and use of system agreement.

⁸The Electricity Industry Reform Amendment Act 2001 also facilitates distributed generation and especially new renewables by relaxing constraints on investment by line companies.

⁹The draft GPS on which the electricity industry was consulted contained the 3 month disclosure requirement and represented a change from the Inquiry's recommendation. See CAB(00)M32/4 A and B.

¹⁰This is a material issue, since there are incentives to under-hedge in the New Zealand market. In all but dry years (i.e.. in most years) spot prices will be lower, and often much lower, than hedge prices (which build-in a "dry-year" insurance premium). This means that conservatively managed retailers, with a high level of hedge cover, are vulnerable to under-cutting by retailers buying mainly on spot, and who may exit the market in dry years.

¹¹Accounting separation is a weaker form of corporate separation involving separate financial statements and nothing else. It raises difficult issues concerning transfer pricing and allocation of common costs.

¹²An important feature of hedge contracts generally is that the higher the proportion of demand covered by hedges the lower the incentive of generators to exercise market power in the spot market, because the profitability of gaming (e.g. by restricting supply to increase the market clearing price) is reduced.

Jump to main page content.

Printed: 26/11/2009