The Riverlink Transmission Line Investment and South Australian Electricity Supply Options

Energy Forum Papers

INTERCONNECTORS AS PUBLIC GOODS

SUPPLY AUGMENTATION ALTERNATIVES

In the integrated national market that is evolving in Australia, the different demand and supply positions of individual market areas means that there are alternatives to providing for market growth.  The main forms of these are either new generation or making use of surplus generation of an oversupplied region.  Augmentations of supply, or the equivalent in voluntary demand shedding, have similar effects to a link with a supply source that is presently under-used.

"EXTERNALITIES" AND THEIR EFFECT ON EFFICIENT INVESTMENT DECISIONS

London Economics, in work conducted for the NSW and SA Treasuries, has argued that the possibility of sponsors of regional interconnectors capturing the gains from them are remote. ⁽¹⁾  LE argues that the benefits from these investments are spread widely through the electricity market and that without a regulated interconnect sub optimal investments will be made.  According to LE, "It is unlikely to be possible to capture the necessary gains from interconnection through private contracts, and moreover, to exclude others from receiving these benefits at no cost." (p. 13).  If true, this would distinguish an interconnect from an augmentation.  Although augmentation also brings general benefits, fundamental to the notion of a competitive market is the ability of a new generator to harness sufficient of the benefits to be a viable investment without government support.

The spillover or public goods nature of the investment are the key reasons for supporting regulated interconnects.  But LE also considers transmission investments to have advantages over generation investments, which are to a considerable degree substitutable.  They base these views on savings fostered by the interconnect permitting a lower cost fuel source from a previously unconnected or inadequately connected supply source.

These sources of savings are not always net savings.  They are offset if the fuel cost savings are achieved at a high investment cost in interconnectors.  Such excessive costs are possible where a regulated interconnect, or entrepreneurial interconnect has been mistakenly constructed.  Fuel cost savings are only valid where the costs of delivering the lower cost fuel derived power are less than the incremental cost of building new plant rather than a new interconnect.  In this respect, some suppliers have suggested the estimated power plant costs used by LE to be excessive.

There may well be greater merit in some of the other arguments raised by LE in favour of interconnects, namely the scale benefits of the interconnected market, the system reliability of a shared enmeshed network and savings in delaying new plant.  These are, however, of a second order importance.

There are obvious disadvantages in allowing regulated interconnects.  Modern electricity markets are being developed to escape from the inefficiencies of the regulated and centrally planned industries.  Centrally planned industry policy approaches were founded on the notion of public goods and externalities.  Cost benefit analysis is the analytical tool used to construct synthetic markets where normal markets do not appear to stack up.  But the social benefits that are said to uncapturable with the provision of the designated goods and services are also present with all goods and services.

Thus, my neighbour's beautiful garden confers benefits on me as do the superior braking characteristics of very high quality cars.  But nobody other than the direct owners of these assets would ever be expected to contribute to their costs.

Under scrutiny, virtually all goods and services said to exhibit strong "externalities" have been brought within normal market provision.  In allowing their supply to be brought about by the interaction of willing buyers and willing sellers, the extremes of overbuild and other inefficiencies that characterised government owned or controlled electricity supply are avoided.

It is the disadvantages of centrally planned decision making that the present market arrangements are seeking to avoid.  The cost of a mistaken investment in an entrepreneurial interconnect is carried by the owner.  With a regulated interconnect, the cost may be made to fall on the promoter under the Optimised Deprival Value methodology but there would be less confidence that the regulatory authorities would be able to determine that an interconnect is redundant than if this were to be determined by the market.

INCENTIVE STRUCTURES RESULTING FROM GOVERNMENT PROVISION

The great weakness of a regulated interconnect is its evasion of the test of commerciality.  Moreover, in this evasion process, commercial incentives in general are blunted.

Regulated and unregulated parts of the market will not easily coexist where the two features offer competing solutions.  Thus, a firm would be reluctant to embark on an activity, say augmentation of generation, if it thought a rival approach, a regulated interlink, could be agreed to.  In the one case, the sponsors would carry all the commercial risk while in the other the regulator would largely eliminate the risk.  Selectively eliminating risks undermines the market process by distorting the incentives to efficient market operations.

In addition, the possibility of obtaining a regulated interconnect is likely to engender considerable socially wasteful lobbying expenditure (arguably this is already taking place with the Riverlink proposal).  This creates new and counterproductive incentive patterns.  It diverts business from the task of seeking to discover and meet consumer needs, and towards spending resources persuading regulators to force consumers into particular spending patterns.

THE VALUE OF THE NSW/VIC INTERLINK

Some evidence of the outcome of an interlink can be gauged from the experiences of the NSW to Victoria link.  The connection through the Snowy system has only operated as a market link since May 1987.

The full linking of the NSW and Victorian markets has brought both a price convergence in the two states and a lowering of the average price.  In the March to May period 1997, the average price in NSW on a time of day basis was $19 and that in Victoria $15.7.  In the period September to November 1997, the average NSW price fell to $14.4 and that in Victoria fell to $14.2.

In addition, the variability of price was reduced in Victoria.  The standard deviation fell from 79% of the average to 74%.  By contrast, in NSW the standard deviation increased from 25% to 85% (albeit, this was around a lower average price).

In essence, the joining of the two markets lifted off-peak prices in Victoria and reduced them in NSW.  Peak price trends were in the opposite direction.

In Victoria, prior to the interconnect operating, zero prices were common and the lowest 10% of the time periods prices saw an average price of just over $4 increase to almost $8.  The increase in these prices in Victoria was in contrast to the experience in NSW where prices for the lowest 10% of the time periods fell from $14 to $8.  Peak prices in Victoria tended to fall from $42.5 to $35.3 in the two periods, while those of NSW increased from $30.3 to $35.8.

The price profiles over the two periods are illustrated in the chart below.

Should the gains be typical of those occurring through the year, NSW customers would obtain a benefit of $280 million per annum, while Victorian customers' annual gain is $65 million.

These benefits incorporate the effects of line losses (about a net 5% of an electricity flow that comprises some 3% of the combined NSW and Victorian markets).  However they exclude the effects of South Australia which from May 1997 changed its bidding strategy from one that was unresponsive to the Victorian price.  Some of the estimated net value of $245 million would have been due to this factor.

The benefits also would be modified by the fact that the estimates are on a time weighted basis.  On an energy weighted basis the benefits to NSW customers would be reduced (and the losses to NSW generators in NSW reduced) because the price reductions were during the periods of low demand, while the increases were in periods of high demand.  In Victoria's case, the customer benefits would be increased (and generators' gains reduced) because price increases were in periods of low demand and reductions in high demand.

Victorian generators would also be net gainers from the trading since exports are overwhelmingly to NSW but the loss of NSW generators' share of their home market would be diminished by the high prices they receive from their (relatively rare) sales in Victoria.

The gains from the NSW-Victoria interlink have been made by all parties.  The issue is whether it is possible to ensure that the promoters obtain incentives in the form of an ability to capture sufficient gains for them to undertake efficient investment expenditure.

RIVERLINK

Having outlined the case for interconnectors in general and Riverlink in particular to be a regulated link rather than an entrepreneurial link, LE goes on to evaluate a raft of alternative investment decisions.  It comes to the view that Riverlink at 200 MW capacity, followed by a 150 MW capacity expansion of the existing Victoria-South Australia link (Heywood) offers the best return.

The Riverlink project is seen to be particularly well placed in view of the excess capacity in NSW.

The report on the matter prepared jointly by ETSA and Transgrid also argues that generation at 95% availability is less reliable than transmission, which it puts at 99.9% reliability.  Together with LE that report sees additional transmission as important in reducing Optima's market power in South Australia.

Neither of these reports consider that there is likely to be great scope for voluntary load shedding in the near future.

In fact the link, though it is promoted as a means of overcoming peak shortages, will operate as a base load supplier.  It will largely displace the output of the gas powered Torrens Island plant while also offering some greater surety that Loss of Load will not occur.

One irony associated with a regulated interconnect between NSW and South Australia stems from the National Market rules.  The interconnect will confer all its advantages on the South Australian consumer.  Indeed, the NSW consumer will incur losses as a result of the investment, since the NSW power will be exported thus tending to raise the price within NSW.  Yet, because the National Market rules allocate interconnect costs to the consumers in whose jurisdiction the interconnect runs, the NSW consumers will pay about half the costs of Riverlink.  The NSW consumers will therefore face a double whammy:  higher transmission charges and increased energy costs.

OTHER VIEWS OF THE INTERCONNECT

In general, firms' submissions to NEMMCO on the matter argue the case for the sort of outcomes that coincide with their commercial interests.  Thus, the NSW generators see considerable merit in a regulated interconnector, which would offer them an additional, more secure means of transport to a lucrative market.  Victorian generators are more inclined towards an upgrade of the existing link through Heywood.  This latter view is also shared by GPU/PNV, the operator of that link, while ETSA and Transgrid are sponsoring Riverlink.

Gas interests and Optima consider that additional local generation offers a better solution and have indicated that they would construct a merchant plant if permitted to do so.  (Presumably if the regulated Riverlink proposal were not overhanging the market).

Many participants take the view that the urgency of the case is inadequately demonstrated and that the risk of a loss of load is remote.  This is supported by a claim that the increase demand which is foreseen is based on the unexpected increase in 1997, an increase which resulted from a one in one hundred year summer temperature pattern.  Some have argued that the Value of Lost Load implied by the case supporting a regulated interconnect is at several hundreds of thousands of dollars, if not at millions of dollars.

The issue is further clouded by the structure of South Australian production, which is concentrated in the two major generators owned by the State owned Optima Energy.  This concentration could allow Optima to raise the price in South Australia above the competitive market price.  LE sees Riverlink as a means of reducing that market power but other submissions suggest Riverlink is too small a source of competition to be effective in doing so.

Clearly, some further disaggregation of Optima is necessary.  We have previously advised that Optima should be divided into two businesses, one centered on the TIPS gas facility and the other on the Northern Power Station. ⁽²⁾

BANKABILITY OF ENTREPRENEURIAL INTERCONNECTS

An entrepreneurial interconnect would control the flow of power between the two markets it connects, buying from the market with the lower price and selling into the market with the higher price.  In this respect it would operate as a generator.

Such a link could be financed in a variety of ways, including by simply arbitraging the prices in the two linked markets.  More likely means of financing would be forward sales of access rights for generators or retailers.

Australia has no experience of entrepreneurial interlinks and overseas examples are rare in view of the highly regulated market from which the world is only now emerging.  Even so, examples of entrepreneurial interconnects are found in North America.  These indicate the viability of such investments.

RECOMMENDATIONS

1. NEMMCO should be aware of the dangers of intervening and requiring the garnishing of consumers electricity bills to construct facilities.  It should be aware that "government failure" from such activity can be more serious than the "market failure" it purports to address.  It should further be aware that interventions in aspects of the market send signals to businesses that can reduce the effectiveness of market based incentives.
2. NEMMCO should examine the situations that have brought forward entrepreneurial interconnects elsewhere in the world and review whether the situation in South Australia is comparable.
3. NEMMCO should review the Code provisions and examine what changes may be necessary to allow entrepreneurial links to be established.
4. The proposal for a regulated interconnect known as Riverlink should be deferred, since:
• the need is not established as the peak shortfall is unlikely to occur and its resolution would imply a VoLL of many millions of dollars per MWh;
• informed views have been offered that Riverlink will not provide the system benefits promoted during critical periods;
• alternative, lower cost and non-regulatory means of meeting a possible peak shortfall have been indicated;  some of these have included offering a different mix of solutions including:
• additional peaking plant
• alternative links
• demand management options;
• the construction of an entrepreneurial link could be viable either on the Riverlink route and/or via Heywood and the market should be allowed to test such options.

---

ENDNOTES

1.  Independent review of interconnection of South Australia and New South Wales, London Economics, NEMMCO website, December 1997.

2.  South Australia:  Energy Situation and Policy Approach, Energy Forum Issues Paper No. 6, December 1997.