We make a series of recommendations concerning new pipeline authorisation and third-party access tariffs for the natural gas network in Ireland. Our recommendations take into account fundamental economic principles, the requirements of the EU Gas Directive, the goals of Irish government policy, and the specific needs of users of the Irish network. Our understanding of the current situation derives in part from extensive consultations with the Department of Public Enterprise (DPE), Bord Gais (BGE), the Irish Business and Employers Confederation (IBEC), the Electricity Supply Board (ESB) and many other interested parties including project developers and a variety of customers.

We begin by examining the tariff recommendations contained in the Bord Gais/Arthur Andersen report. We reject the Full Postalised system on grounds of efficiency and non-discrimination, and endorse the Irish Entry/Postalised Exit system. Separate entry charges are essential to ensuring cost-reflectivity and preventing discrimination, as required by the Gas Directive. Postalised exit charges can be justified as a legitimate public policy goal. If postalised exit charges are supplemented by our specific recommendations to preserve a level playing field for potentially competing pipeline projects, then they will represent a legitimate Public Service Obligation consistent with the Gas Directive.

We recommend that tariffs be calculated using a depreciation methodology that gives an annual 2% decline in real charges (on a per unit basis). This “RPI-2%” approach reduces the competitive distortions that can arise under conventional depreciation techniques such as straight-line depreciation, while protecting BGE from the danger of financial distress that might arise under a one-time shift to economic depreciation.

Cost-reflectivity should be the basis for determining entry and exit charges. Several other principles apply to tariff-setting for the Irish system. First, because BGE will play a major role in the calculation of tariffs, and occupies a unique position as the state-owned, vertically-integrated incumbent monopolist with a proposed interconnector project, we recommend a special “transparency obligation”. BGE should be required to publish in full its cost estimates and allocations, its volume forecasts, and all the calculations that produce its tariffs. The publication requirement should include the provision of spreadsheets and any specialised software.

Second, we propose that Corrib entry charges should be designed to recover in full the cost of the spur from the Ring Main to the beach where Corrib gas will be landed. Charges should be based on reasonable assumptions of future volumes from the Corrib field and from other potential West Coast gas developments. The Corrib spur will offer other benefits, such as bringing gas to some local populations who do not currently have access to the network. Reasonable estimates of these benefits should be deducted from the Corrib entry charge.

We describe two possible ways of handling entry charges: one where all the risk of future utilisation is borne by the Corrib partners (as would be the case if they themselves paid for the spur); a second where all the risk is borne by BGE. We recommend that the parties involved be free to choose either system, as long as they neither subsidise the Corrib partners nor force them to pay more than the full costs of the spur on expectation.

Third, we recommend that the EU grant be removed from the ratebase of the existing Moffat-Dublin interconnector (Interconnector 1) and moved onshore. The tariff for Interconnector 1 should then be reset to the tariff that would now apply, had tariffs for Interconnector 1 been based on “RPI-2%” consistently from the date it began service.

A number of the currently proposed projects have the potential to serve customers directly, bypassing the BGE system. In the context of a postalised tariff, this raises the prospect of cherry-picking and inefficient bypass. At the same time, the postalised rate could deter competition from potential pipeline developers in areas that are effectively subsidised by the onshore system. To ensure competitive neutrality, we therefore recommend a Public Service Levy (PSL) to be applied to all pipelines, whose proceeds would be paid into a Geographic Equalisation Fund. The postalised exit charge would then be viewed as having two components: a cost-related charge, and the PSL. The latter would be positive in the Dublin area, and negative further away. A customer in the Dublin area switching from BGE to an independent pipeline would still be liable for the PSL charge. A customer in the southwest would still receive the PSL benefit, whether or not it used the BGE network. Independent pipelines would therefore compete with BGE on a level playing field, rather than taking advantage of, or being hindered by, potential distortions of the postalised system. From the viewpoint of pipelines seeking to compete with BGE, our proposal has exactly the same effect as would the adoption of “distance-based charges for the onshore network.

We recommend that the government authorise the construction of any pipeline that can show long-term contractual commitments by customers for a specific amount of capacity. We propose that the requirement be a ten-year commitment for 4.5 TWh per year. We propose safeguards to prevent market abuse should such a pipeline first obtain authorisation but later find itself in a position to exploit market power. Authorised pipelines will be obliged to serve future customers at prices tied to those in the initial contracts that formed the basis of the government’s authorisation.

We also propose a safeguard to ensure that BGE competes on commercial grounds. Rather than show contracts, BGE should be required to wait until a deadline has passed. If no other pipeline has obtained sufficient long-term contracts to secure authorisation at that point, then BGE’s “Interconnector 2” project will be authorised by default.

We explain that this system effectively allows large customers to choose which of several competing projects will be built. We believe that large users who are willing to sign contractual commitments are in the best position to evaluate the merits of competing proposals. Even though BGE’s proposal for Interconnector 2 will have to wait and see if other projects can secure contracts before a specific deadline, large users will naturally keep the potential merits of Interconnector 2 in mind. If large users believe that Interconnector 2 is the most attractive project, then they can ensure its authorisation simply by refusing to sign contracts with any of the competing projects. We therefore conclude that our proposal allows all potential projects to compete on the merits.

Our recommended pipeline authorisation scheme has several distinct advantages for Ireland under current market conditions. We explain the advantages of our proposal over the alternatives of simply having the government pick a preferred project for authorisation, or of automatically granting authorisations to any pipeline that seeks approval. We also explain the merits of our proposal relative to two other schemes that we considered in considerable detail: authorising all projects that could commit to charging below a regulated price cap, and conducting a franchise auction where an exclusive authorisation would be granted in a bidding process.

Our proposed authorisation system is appropriate for choosing between the several projects that currently compete to bring additional gas to Ireland and that aim to serve new power stations as well as other customers. However, our proposal might not be appropriate in future circumstances if a potential project lacks competitors, or if a potential project would principally serve smaller customers who would not be willing to sign long-term contracts. We therefore recommend a specific set of guidelines for the government to determine when authorisations should be conditional on securing long-term contracts.

Background

The Department of Public Enterprise has asked The Brattle Group to recommend a policy for the authorisation of new pipelines, and to advise on tariffs for third party access to the transmission network. The Department asked us to make recommendations concerning the principles of authorisation and tariffication, on the basis of compatibility with the EU Gas Directive, Irish government policy, and fundamental economic principles.

Our recommendations follow extensive consultation with interested parties. We met on an informal basis with representatives of BGE Transmission, a number of private pipeline developers, offshore natural gas developers, IPPs, and representatives of most of the major eligible customers in Ireland. We presented our draft proposals publicly on April 4, and subsequently received written submissions from a large number of interested parties. This report includes our responses to the points raised in those submissions.

Although our recommendations concern principles rather than actual tariffs, we recognise that industry participants have a natural interest in knowing the quantitative implications of those principles. We have therefore attached indicative tariffs based on our recommendations, produced at our request by BGE (Appendix 3).

Finally, our recommendations and discussion are limited to the terms of reference provided by the Department. During our consultations, a number of issues were raised that fall outside those terms, including

• ·Possible legislative requirements for implementing the recommendations.
• ·The appropriate cost of capital for BGE Transmission.
• ·Connection charges.
• ·Electricity transmission tariffs.
• ·Guaranteeing the international competitiveness of particular Irish companies.
  

Tariff Recommendations

The Bord Gais/Arthur Andersen study examined a number of alternative tariff systems for the BGE transmission system: Full Postalised, Separated Postalised, Distance-Related, Zonal and Entry/Exit. It concluded that the Full Postalised system and a variant on Separated Postalised known as the Irish Entry/Postalised Exit system were the two most satisfactory approaches. Both were deemed acceptable to consumers, but the Full Postalised system was considered unacceptable from the perspective of producers.

Under the Full Postalised system, all users pay the same fee for use of the system. In particular, “the system” is taken to include both the onshore infrastructure and the UK interconnector, and the tariff therefore covers both onshore and offshore costs. Consequently, even users who purchase gas from Kinsale, or in the future from Corrib or other indigenous sources, contribute to the cost of the UK interconnector.

Under the Irish Entry/Postalised Exit system, users pay a tariff comprised of two components: a distinct entry charge for each entry point used and an exit charge that is common to all exit points, and is designed to recover the costs of the onshore network.[1]

We recommend a version of the Irish Entry/Postalised Exit system. Below we explain in detail our objections to the Full Postalised system, and discuss the other methodologies.

The Full Postalised System

Proponents of the Full Postalised system claim that it would produce lower gas prices than the Irish Entry/Postalised Exit system. However, such claims are flawed: a fully postalised tariff regime could be expected to raise costs in the long term.

A fully postalised system would impose a tax on indigenous production and cross-subsidise the existing interconnector, promoting inefficiency and higher prices in the long run. Full postalisation is not cost-reflective, and would deter both indigenous production and efficient expansion of Ireland’s pipeline capacity to the UK. A fully postalised system is discriminatory and would be inefficient, as it cross-subsidises imported gas. For example, gas deliveries from Kinsale would be forced to contribute to the cost of the existing interconnector, even though Kinsale gas does not use the asset. Such discrimination is clearly incompatible with the Gas Directive.

We therefore recommend against full postalisation, on grounds of economic efficiency, long-run customer benefits, and compatibility with the Gas Directive. Below we explain these arguments in detail.

Claims in Favour of the Full Postalised System

Some respondents to the Bord Gais/Arthur Anderson study[2] have argued that a fully postalised system produces lower gas prices than an “Irish Entry/Postalised Exit tariff. The Electricity Supply Board believes that the Irish Entry/Postalised Exit’ tariff creates a situation where the existing price for gas transported from Moffat is increased by about 20%[.][3] Irish Fertilizer Industries (IFI) argued that, the Postalised System previously proposed by the Department of Public Enterprise is the correct one for now and the futureThe new proposals would increase the cost of gas transportation to IFI and further disimprove our competitiveness.[4]

The claim that the fully postalised system would produce lower prices, however, is flawed. As we explain below, it assumes that indigenous production could somehow flourish under a tariff system that forced it to subsidise Interconnector 1. In the absence of indigenous gas, the fully postalised system would produce identical prices to the Irish Entry/Postalised Exit sytem. To understand the long-term implications of alternative systems on prices, it is therefore important to investigate the potential development of indigenous gas sources. It is also important to understand the impact of full postalisation on potential investments in other interconnectors, as the efficiency of new investment will be key to the absolute level of gas prices. We explain that the fully postalised system is likely to deter indigenous gas production and efficient investments, producing higher gas prices in the long run. In the long run, separate and efficient pricing of any interconnectors in an Irish Entry/Postalised Exit system can be expected to produce lower prices.

The Claims Assume Continued Indigenous Production

Claims that the fully postalised system would mean lower prices rest on two assumptions: that imported gas sets market prices, and that there would still be significant volumes of indigenous gas produced under a fully postalised system. This second assumption was made, but never explored, by the Bord Gais/Arthur Anderson study.

However, the use of a fully postalised system itself constitutes a disincentive to the development of indigenous gas. Without indigenous production, the tariff under the fully postalised system would be exactly equal to the tariff under the “Irish Entry/Postalised Exit” proposal. We demonstrate this phenomenon mathematically in Appendix 1.

Full Postalisation Would Discourage Indigenous Production

Natural gas from the Corrib gas field has to compete with imported gas on a delivered price basis. Corrib gas can expect no premium over the delivered price of imported gas: the price of Corrib’s gas at Sligo has to equal the price of imported gas at Moffat, plus the difference in transportation charges between gas from Sligo and gas from Moffat.[5]

Under a fully postalised system the difference in transportation prices from Sligo and from Moffatt would be zero. Under the “Irish Entry/Postalised Exit” proposal however, a difference in transportation prices would emerge, and Corrib’s gas could command a premium. The indicative tariffs produced by BGE suggest that the difference would be on the order of 1.2p/therm. Since this difference reflects underlying transportation costs, failure to reflect it in a fully postalised system would be like taxing Corrib to subsidise the interconnectors. The “tax” should be compared with UK gas prices, which typically lie in a 10p/therm to 15p/therm range.

This difference may be sufficient to prevent Corrib from becoming commercially viable. If under an “Irish Entry/Postalised Exit” tariff Corrib has a profit margin of 10%, then the “tax” under the fully postalised system would completely wipe it out. Even if Corrib can still get developed under a fully postalised system, the “tax” could significantly reduce Corrib’s production, for example by preventing the development of marginal parts of the field that would be economic without the “tax”.

The arguments above apply equally to any indigenous production of natural gas. Efficient indigenous production should be encouraged. Tax revenues, greater diversity and hence greater security of supply would all benefit from the development of the Corrib field, and other economically viable offshore fields.

Full Postalisation Would Distort Investment in UK Interconnectors

A fully postalised system would also deter private investment in the expansion of Ireland’s pipeline capacity to the UK. Competing interconnectors such as Premier Transmission[6] and Celtic Pipeline proposals would, like Corrib, be subsidising the original interconnector. Premier Transmission and Interconnector Energy Ireland have both affirmed this point in the consultations. Premier Transmission argued,[7]

Premier Transmission supports the “Irish Entry” approach. This offers improved cost reflectivity and thereby minimises cross-subsidy between alternative sources of supply. The approach ensures that alternative pipeline developers, such as Premier Transmission, are not prevented from offering services by tariffs that do not reflect underlying costs.

Interconnector Energy Ireland commented that,[8]

The Full Postalised Tariff would require an alternative supplier to pay for a substantial part of BGE’s network (interconnector) for which that alternative supplier would have no need, while also bearing the cost of his own infrastructure…In effect, the application of the Full Postalised Tariff to a privately funded supplier is a subsidisation of BGE. Subsidisation of this nature makes it impossible for a third party interconnector to compete with BGE and would be, therefore, anti-competitive and a barrier to Entry.

If private investment were deterred, then Ireland would face a capacity shortage and higher market prices, unless BGE assumed responsibility for expanding capacity under its legal duty to supply all of Ireland’s needs. However, this scenario tempts an inefficient outcome. If BGE does not face competition from other potential projects, then there is no market test of its Interconnector 2 project. The project may be too big, resulting in higher prices if anticipated volumes do not materialise. Alternatively, the project may be a more costly way of adding capacity than one of the proposed private pipelines.

Respondents to the consultation were concerned by this possibility. Premier Periclase argued that investment in Interconnector 2 “can only be justified by assumptions of substantial gas volume throughput increases, largely arising from IPPs. This is an area that has yet to be developed to the stage where demand volume/location can be quantified.”[9] The Electricity Supply Board commented that, “there is a concern that over-investment or stranding of transport capacity would visit undue costs on captive customers.”[10]

Suspicion surrounds projections of the volume of interconnector gas that will be required in the future. If a pipeline that is larger and more expensive than necessary for the volumes actually demanded is constructed, then the interconnector will be forced to raise prices to cover costs. As imported gas sets the market price, this means higher gas prices for everyone in Ireland. The possibility that an interconnector more expensive than necessary will be built is increased under the fully postalised tariff, which discourages other projects and pushes Interconnector 2 to be built by default.

Conceivably, BGE could be stopped from raising prices. However, as BGE is a government-owned company, forcing it to lose money on Interconnector 2 simply means that the Irish taxpayer loses money.

Full Postalisation Is Incompatible with the EU Gas Directive

The EU Gas Directive requires non-discrimination in setting tariffs and other terms of service. Article 7.2 states that:

In any event, the transmission, storage and/or LNG undertaking must not discriminate between system users or classes of system users, particularly in favour of its related undertakings.

The Brattle Group’s report for the European Commission on implementing the Gas Directive stressed that non-discrimination has to be understood in the broad economic sense.[11]

Access terms that are formally non-discriminatory, but whose economic effect is to unreasonably disadvantage certain classes of customers, are discriminatory in the economic sense…To reflect the principle of non-discrimination, tariffs should reflect costs in a broad sense.

Arguably, an inefficient pricing system can be seen as discriminatory against new gas projects. Corrib could argue that the fully postalised system discriminates against it, as the tariff regime does not allow Corrib to compete on the basis of its fundamental economics. Under the fully postalised system Corrib is forced to pay, not only for the cost of the field and the cost of the spur connecting it to BGE’s transmission network, but for the UK interconnector of which it makes no use. This could be seen as discriminating in favour of BGE and against Corrib.

In addition, Corrib may argue that BGE holds a dominant position in gas transmission, and that a fully postalised tariff constitutes an abuse of this position contrary to Irish and European competition law. The fully postalised tariff could be seen as forcing people to pay for the BGE interconnector(s) whether they use it or not, effectively constituting a ‘tying arrangement’. This ‘tying arrangement’ could be alleged to constitute an abuse of a dominant position by deterring alternative pipeline projects.

The Irish Entry/Postalised Exit System

The use of cost-reflective entry fees resolves the objections outlined above to the fully postalised system. Cost-reflectivity prevents discrimination between indigenous and imported gas, and provides efficient incentives for the development of indigenous gas sources. It ensures compatibility with the Gas Directive.

Geographic Tariff Equalisation as a Public Service Obligation

Our consultations have indicated widespread support for continued geographic equalisation of on-shore gas transportation charges (“postalisation”). Equalisation contributes to a number of environmental and social goals. It reduces the incentive to locate new gas-fired power plants and other businesses in the Dublin area, and thereby:

· Avoids additional congestion of traffic and housing.

· Reduces transmission losses that would arise from further concentration of generating capacity in the Dublin area.

· Avoids increased localisation of emissions of locally acting pollutants.

· Fosters economic development of other regions within Ireland.

These benefits of postalisation are independent of who owns or operates the network. Postalisation is intended to benefit customers and society, not pipeline companies.

It would not be appropriate for this study to offer an opinion on the desirability of equalisation. However, we do view it as a legitimate goal of public policy, and, if implemented appropriately, an acceptable Public Service Obligation (“PSO”) under the terms of the EU Gas Directive.[12] In evaluating alternative methodologies, we have therefore taken the desirability of on-shore postalisation as given.

Non-Discrimination

Some parties have suggested that tariffs should be designed to ensure the competitiveness of Irish industry with companies in continental Europe. They view the term “non-discriminatory” as requiring the same total charges for gas transportation in Ireland as continental Europe. However, “non-discrimination” permits differences in total charges where justified by cost differences. If it costs less to deliver gas to Ireland than to another Member State, then Irish industry can enjoy paying lower prices without other countries raising claims of discrimination. Conversely, if it costs more to deliver gas to Ireland than to another Member State, Irish industry cannot interpret “non-discrimination” to require a government subsidy to remove the difference. Such a subsidy would constitute “state aid” which must satisfy strict conditions that are not part of this study.

Finally, the comments that we received in connection to Irish competitiveness were made only under the belief that a fully postalised system would lower total costs for Irish industry. We explained above that this belief relies on the continued viability of indigenous gas despite a tariff methodology that would seriously discriminate against it. This belief also ignores the potential for full postalisation to raise total gas prices, by providing inefficient incentives for new pipeline infrastructure.

Other Proposed Systems

The BGE/AA report also considered a number of other proposed systems. We do not explore any of these systems, because they would not preserve geographic price equalisation of the on-shore network.

Depreciation Methodology

In our draft proposals we recommended the use of “economic depreciation”, which would give tariffs that remained constant over time in real terms, on a per unit basis (i.e., the capacity charge per peak day kWh and commodity charge per kWh would both be constant over time in real terms).

For designing entry and exit charges, we do not recommend a depreciation schedule based on accounting conventions such as a “straight line” in nominal terms or one that is “trended” for inflation. These conventions can produce significant distortions that lead to economic inefficiency. If two pipelines compete to transport gas, conventional depreciation techniques would produce much lower prices for the older pipeline than for the new one. Such a difference in prices could distort competition between the pipelines. Economic depreciation minimises such distortions, and approximates the time pattern of prices in competitive markets.

However, BGE has expressed concern that a switch to economic depreciation would lead to a dramatic reduction in earnings in the next few years. Assuming that Interconnector 2 is built, BGE provided figures demonstrating that an immediate switch to economic depreciation would cause its interest cover to fall significantly below 1, potentially leading to financial distress. We regard this as a significant cause for concern.[13] BGE is also concerned that the use of economic depreciation shifts too much capital recovery to rely on projected volume growth that might not materialise.

It can also be argued that in a competitive market, technological progress and other efficiency gains should cause the unit price to fall in real terms over time. This argument is accepted by regulators in the United Kingdom and elsewhere.

We acknowledge these arguments and conclude that BGE’s proposed depreciation method, which would produce unit prices (both capacity and commodity) that fall slightly in real terms, is reasonable. BGE has suggested that a value of 2% for X would reflect their own estimation of the rate of technological improvement over time, and would also produce a more acceptable earnings profile. This value is compatible with regulatory practice elsewhere, and we regard it as reasonable.

At our request, BGE has calculated indicative tariffs based on our recommendations, for a number of alternative demand and supply scenarios. These are attached in Appendix 3.

Further Principles for Setting Tariffs

Cost-reflectivity should be a basic principle for setting entry tariffs in Ireland. This implies that each entry charge in the recommended Irish Entry/Postalised Exit system should be set to recover the costs of all infrastructure associated with entry at that point. The postalised exit charge should recover the costs of all other infrastructure associated with gas transportation.

This general principle requires elaboration in particular cases. Below we lay out additional recommendations on the principles that should underlie tariff-setting, relating to:

· The need for transparency in the calculation of tariffs, in particular because of the involvement of BGE.

· Tariffs for the onshore Corrib “spur”.

· Tariffs for Interconnector 1.

Transparency

As a practical matter, BGE is likely to play a major role in calculating regulated tariffs in Ireland, whatever principles of tariff-setting are adopted. This naturally gives rise to concern about potential abuse. Even in the absence of actual abuse, there is a danger that the real or imagined potential for abuse will engender mistrust in the system, leading to under-investment by customers and third parties. Without rigorous oversight of the calculations, it will also be difficult for the DPE or a future Regulator to be sure of their validity.

To avoid such problems, we recommend that BGE be subject to a rigorous “transparency obligation”. This obligation would include a requirement to publish full details of all tariff calculations, including

· A full description of the transmission infrastructure (pipelines, compressors and other major items).

· For each such item, details including physical data (e.g. length, diameter etc of individual pipes), age, expected remaining life, original cost, current “regulatory book value” and the depreciation scheme used to arrive at this value, and future depreciation schedule used in the tariff calculation.

· Historical data by month on pipeline capacity utilisation by receipt and major delivery point, linepack status, and episodes of curtailment and congestion.

· Volume forecasts used in the tariff calculations, with full explanation of how they were derived. BGE should also provide a range of alternative scenarios and volume forecasts that it considers reasonable, as it has done in the Gas 2025 study. These would facilitate sensitivity analysis of future BGE tariffs.

· The forecasts of future capital expenditure used in calculating tariffs. These should be provided in full detail, including a year-by-year breakdown. BGE should also explain the basis of such forecasts. For example, if BGE expects to reinforce a certain location at a certain date, it should explain the need for the reinforcement, the determination of the extent of the reinforcement, and how the particular date is derived from its volume and/or other forecasts.

· All assessments of variable (O&M) costs used in calculating tariffs.

· The mechanics of the calculations themselves, in full detail.

In addition, BGE should be required to perform and publish sensitivity analyses showing how its tariffs would respond to alternative assumptions concerning future volumes and other relevant factors. The figures used by BGE should be subject to independent validation by the DPE/Regulator. Finally, our proposals concerning bypass, described later in this report, would require BGE to perform further calculations (to determine the “Public Service Levy”). The same obligations should apply to these calculations.

Publication should be full, i.e., BGE should provide the underlying data and all necessary computer models etc, in electronic format. It should also provide access to any specialised software used for tariff calculations.[14] As a further element of the requirement, BGE should ensure that users can follow its calculations in full. We recommend a one-day seminar where BGE would explain in detail its calculations, and answer all reasonable questions. BGE should also appoint a staff member responsible for responding to further reasonable questions in a timely and open fashion.

Finally, publication must respect customer confidentiality. The publication requirement described above appears to be compatible with customer confidentiality.

Full implementation of the “transparency obligation” should ensure that third parties can easily and confidently:

· Understand the data and assumptions underlying the BGE tariff calculations.

· Challenge particular elements of the data, or particular assumptions.

· Reproduce the calculations themselves, without excessive effort and with confidence as to each step.

· Carry out their own sensitivity analyses to see how the BGE tariffs would change in response to alternative assumptions.

In our consultations, we heard numerous complaints alleging “excessive optimism” in the BGE volume forecasts. One possible authorisation policy that we discuss below would allow the BGE Interconnector 2 project to secure authorisation by signing contracts like any other privately-funded project. Although we do not recommend this option, we note that its effective implementation would require government scrutiny of any BGE contracts for cost-reflectivity. This would require the development of a reasonable base case demand scenario. A reasonable “transparency obligation” would require that this scenario be agreed with the DPE or Regulator.

We understand that BGE has a statutory obligation to ensure adequacy of supply under all likely scenarios. However, this does not mean that potential contracts for the Interconnector 2 project should be evaluated based on excessively optimistic scenarios. The volume forecasts should be based on a central or average expected scenario. Most market participants believe that the “base case” used in the BGE/AA report (“Gas World B”) is unduly optimistic in its forecasts of future volumes. For example, it assumes that Moneypoint will be fully converted to gas, although closing down the plant might be more economical and reduce emissions even more than a conversion to gas.[15] The “base case” also assumes that all gas-fired plant will continue to run as baseload in the future, whereas it seems more likely that the construction of new plant will force older gas-fired plants “up the merit order” to run less frequently.

Unique Position of BGE

The imposition of a unique transparency obligation on BGE is not discriminatory. Rather, it is a response to the unique position that BGE holds in the Irish natural gas industry: it is the incumbent former monopolist, has a dominant position in transmission and arguably in supply, is vertically integrated and wholly state-owned.[16] Unlike private pipelines, BGE has the ability to ensure full cost recovery, effectively placing its customers (and ultimately, Irish taxpayers) at risk for under-utilised investment. Moreover, under our proposals concerning bypass (the “Public Service Levy” described later in this report), BGE’s calculations will have a direct and significant financial impact on the customers of private pipelines, creating an obvious potential for discrimination by BGE. This issue does not arise for any other transmission provider.

The Corrib Entry Charge

The Corrib entry charge should be designed to recover the cost of the pipeline “spur” that would have to be constructed from the beach to the Ring Main.[17] However, calculating the entry charge is complicated by several factors:

· Even the most optimistic projections of production from the Corrib field suggest that the field would be depleted before the spur reaches the end of its engineering life.

· Other gas fields in the vicinity of Corrib could be developed and use the same spur after the Corrib field is depleted.

· The spur offers additional benefits such as making natural gas accessible to some local towns.

These factors lead us to the following recommendations:

1. The Corrib entry charge should be designed to recover the full cost of the spur, in expected NPV terms. In other words, the expected (“probability-weighted average”) NPV of the revenues from the pipe should equal the cost of the pipeline.[18] Any downside risk of stranding should be balanced by an equivalent upside risk that revenues will exceed costs. Figure 1 illustrates this principle with a simple numerical example. In the example, the revenues may lead to recovery, over-recovery or under-recovery of costs, depending on the extent of future offshore production (and therefore of pipeline use and revenues).

2. Any system of charges that is consistent with this principle, and with the standard requirements of non-discrimination, transparency and objectivity, is reasonable. One possible system would have the Corrib partners bear the volume risk associated with the pipeline. For example, the partners might simply be asked to build the pipe themselves, or equivalently, to provide BGE with financial guarantees of revenue recovery should BGE build the pipeline. We discuss below the details of such a system.

3. Alternatively, BGE could choose to bear the volume risk associated with the pipeline. For example, BGE could set a fixed tariff path that would not alter in the future even if volumes were lower or higher than expected. Higher-than-expected volumes would therefore lead BGE to over-recover, while lower-than-expected volumes would expose it to under-recovery. The expected NPV of future revenues should equal BGE’s cost (including a return on capital), as in the simple example of Figure 1. We discuss below the details of such a system.

Risk-bearing by the Corrib partners

Under the first system outlined above, entry charges would be designed so that the volume risk associated with the pipeline is borne by the Corrib partners. One way to achieve this would be to have the Corrib partners build the pipeline themselves (i.e., provide the initial investment cost upfront in return for future use of the pipeline).

Alternatively, the entry charge could be set to recover the full cost of the spur based on forecast volumes, and supplemented by additional payments to or from the Corrib partners if necessary. The entry charges would be designed to recover the costs of the spur over a forecast of volumes by all users, including Corrib and future West Coast gas developments. However, a mechanism would be needed to place the Corrib partners at risk for these volumes. If after several years it became clear that actual volumes would not recover the full cost of the spur, then the Corrib partners would remain financially responsible for the resulting shortfall in BGE revenues. If volumes exceeded forecasts and the entry charges stood to recover more than the full costs of the spur, then the Corrib partners would be entitled to retrospective rebates from BGE. The entry charge paid to bring Corrib gas onshore should also be reduced to reflect benefits from other uses of the spur, for example in extending access to local towns. The Corrib partners should not be at risk with respect to these benefits. The entry charge can be based on reasonable forecasts of these benefits.

In technical terms, this means that:

· The Corrib partners’ payments for the transport of gas from the Corrib field should have an NPV equal to the cost of the pipe, less the sum of (1) the expected NPV of revenues from entry charges associated with the transport of gas from other West Coast fields, and (2) the expected NPV of the value arising from other uses of the pipe.

· Any difference between the expected and actual NPV of revenues from entry charges should be borne by the Corrib partners.

Revenues should be calculated according to standard methods (i.e., using probability-weighted averages). NPV’s should be calculated using the cost of capital for BGE Transmission.

Risk-bearing by BGE

Under the second system outlined above, entry charges would be designed so that the volume risk associated with the pipeline is borne by BGE. BGE would set a tariff designed to recover its costs, in expected NPV terms, as illustrated in Figure 1 above. In order to do so, it would need to take a view as to the probability of different future usage scenarios of the pipeline, again as illustrated in Figure 1.

As with the first system, this exercise would also require BGE to estimate the value of future usage of the pipeline. Also as with the first system, expected revenues should be calculated according to standard methods, and NPV’s should be calculated using the cost of capital for BGE Transmission.

Comparison

The Corrib partners have two alternatives: building the spur themselves and owning it directly, or letting BGE build it in exchange for future entry charges. Our proposals are designed to promote an efficient choice between these two options. Corrib should be motivated to build the spur independently only if it can do so at lower cost than BGE. Thus, our proposals seek to place the Corrib partners in the same economic position whether they build the pipe efficiently themselves or BGE does it. Although the two approaches allocate risk differently, they do so symmetrically: the party that bears the risks of stranding and under-recovery is exactly compensated in NPV terms by a corresponding “risk” of over-recovery. Corrib is therefore prevented under either approach from either bearing a penalty or receiving a subsidy, as might occur under alternative proposals. We recommend that the choice of approach be made by the parties involved.

Details of the arrangement should of course be subject to approval by the regulator and/or DPE. In particular, customers who import gas from the United Kingdom might fear that either system described above might not fully allocate risks to either Corrib or BGE, but instead place them on the users of the on-shore network. For example, “Corrib risk-bearing” might be incomplete if the stranding of the spur is identified only after the Corrib field has been depleted and financial redress can no longer be obtained from the Corrib partners. Other users would understandably object if BGE instead sought compensation by raising the onshore postalised rates. Alternatively, BGE might not care to ensure symmetrical risks of over-recovery and under-recovery under the “BGE risk-bearing” approach if eventual indemnification could be obtained from the onshore system.

If BGE were allowed to recover losses from the spur by increasing on-shore rates, the result would be to discriminate against other interconnectors. Customers who imported gas from the United Kingdom would effectively have to pay for part of the Corrib spur even if they did not use the asset. The same arguments that we made in favour of the Irish Entry/Postalised Exit system would counsel against this outcome. We therefore recommend that the Corrib “rate base” be kept separate from BGE’s other assets and cross-subsidies not be allowed.

An alternative strategy for designing the Corrib entry charge might be to allocate to Corrib all the costs of the spur, plus some fraction of the capital costs of the Ring Main, under the argument that the Corrib field is a key beneficiary of the Ring Main. However, we understand that the Ring Main will be built whether or not the Corrib field is developed, and it appears to be justifiable as a network extension whether or not Corrib is developed. All Corrib’s customers will already pay for a share of the Ring Main’s costs when allocated to the onshore postalised rate, and we understand from BGE that Corrib use of the Ring Main will therefore not impose additional costs on other users. We would therefore view it as a penalty for Corrib to be allocated some unique additional share of the Ring Main’s costs.

Moreover, the beneficiaries of the Ring Main include customers in the south who will enjoy additional flexibility and security of supply, new customers in geographic areas that will be able to receive natural gas supplies for the first time, and potential other developments off the west coast. Construction of the Ring Main would also avoid the need to reinforce the existing line from Dublin to Cork. When a project such as the Ring Main has network benefits that accrue to diverse groups, the argument for all its costs to be averaged in a postalised rate is especially strong.[19]

Benefits of the Spur for Local Towns

One benefit of the spur is allowing several towns to receive natural gas supplies for the first time. Both systems described above involve measuring this benefit and applying it as an offset to the costs that form the basis of the Corrib entry charge.

Estimating the benefit of serving local towns could be a complex and time-consuming exercise. We appreciate that it is most important for the Corrib partners to have an immediate indication of the approximate level of the total entry charge. We therefore put forward a practical proposal for measuring the benefit. Although our proposal can only provide an imprecise measure of the benefits to serving local towns, it is practical and can be implemented quickly.

Our practical proposal is that the benefits of serving local towns off the Corrib spur be measured by reference to the highest distance-related tariff that BGE could justify for any customers on its existing system or the Ring Main. In a separate recommendation below concerning potential bypass of the BGE system, we ask BGE to indicate what the hypothetical distance-related tariffs would be to different points of the onshore system.[20] We believe that the highest such tariff indicated by BGE is likely to constitute a useful proxy for the benefits of providing access to local towns off the Corrib spur.

Once the highest distance-related tariff of the BGE system has been determined, we recommend that it be applied to the total anticipated consumption by local towns off the Corrib spur. This will yield a set of hypothetical “revenues” that the local towns would pay if they were charged distance-related tariffs.[21] The hypothetical revenues should be deducted from the total annual capital charges of the Corrib spur in calculating the entry charge for each year. We recommend that BGE develop estimates of consumption by local towns off the Corrib spur and perform this exercise to provide the Corrib partners with indicative total entry charges as soon as possible.

A question has arisen whether the location of a new power station on the spur should be deemed an additional benefit that would warrant further reductions in the Corrib entry charge. We do not believe so. The distinction between local towns and a new power station is that local towns did not previously have access to the natural gas network. The Corrib spur was necessary to provide that access. In contrast, new power stations can choose from several possible sites on the existing onshore network. The construction of the Corrib spur is not necessary to provide them access. We therefore would not extend the calculations that we propose for local populations to new power stations that might locate on the spur.

Shippers who Use the Corrib Spur

Under the BGE Code of Operations, entry capacity is reserved by Shippers. It has been suggested to us that our first system described above is therefore mistaken in discussing retrospective payments between BGE and the Corrib partners, and should instead require retrospective payments between BGE and shippers who purchase Corrib gas. This suggestion misunderstands the economics of the Corrib field.

Under any plausible scenario, the delivered price of gas in Ireland will be determined by the UK price plus the cost of transporting gas from the UK to Ireland. The price received by the Corrib partners for gas from the Corrib field will be the delivered price of gas in Ireland, less the price of transporting gas from the Corrib field to the onshore system. Consequently, any increase in the Corrib entry charge reduces the price that the Corrib partners can command for their gas. Conversely any decrease in the entry charge will increase the price that the Corrib partners can command for their gas.

It follows that the level of the Corrib entry charge has no economic impact on Shippers, but is absorbed entirely by the Corrib partners. The Corrib partners benefit if in retrospect the entry charge has been calculated too low, and they suffer if the entry charge in retrospect has proven too high. If future utilisation of the spur is higher than anticipated and the charge has proven higher than necessary to recover the total cost of the spur, it is therefore appropriate to allocate any retrospective rebates of the Corrib entry charge directly to the Corrib partners. Conversely, if future anticipated volumes do not materialise and the costs of the spur have not been fully recovered, it is appropriate to seek compensation for any shortfall directly from the Corrib partners.

Back to Contents

Charges for the First Interconnector

In the course of our consultations, various parties expressed concerns to us that the proposed Moffat entry charge is not properly cost-reflective. We heard of two specific factors that allegedly depress the entry charge: the inclusion of an EU grant in the ratebase; and questionable allocation of infrastructure costs between the interconnector and the onshore network. These concerns were exacerbated by an alleged lack of transparency in the calculations.

An additional potential distortion arises in relation to Interconnector 2: it appears that the Moffat entry charge is calculated on the basis of a single ratebase that covers both the old and new interconnectors. The first interconnector arguably has a low ratebase reflecting the effect of depreciation, and possibly the two factors noted above (the EU grant and questionable cost allocations). Pooling this low ratebase with that of the new interconnector therefore effectively “subsidises” the latter.

An artificially low Moffat entry charge would be both discriminatory and inefficient. It would discriminate against entrants seeking to build competing infrastructure, by effectively subsidising the Moffat-Dublin interconnector(s). The potential for inefficiency would then exist, because these subsidies would reduce the incentive to develop alternative infrastructure, and/or indigenous sources of gas. Although some parties may argue that efficient pricing of Interconnector 1 may temporarily raise gas prices in Ireland, it would have the effect of reducing long-term prices. Long-term cost reduction is supported by the same arguments that we developed above against a fully postalised system.

We therefore recommend a revision of the charges for Interconnector 1 as follows:

1. The EU grant should be removed from the ratebase of Interconnector 1, and applied instead to the onshore network, ensuring that it benefits all gas users in Ireland, without discriminating against potential new pipelines.[22]

2. The “transparency” obligation discussed above should be applied to require BGE to publish full details of its tariff calculations, including cost allocations.

3. BGE should keep the rate-base and tariffs on Interconnector 1 entirely separate from the prospective rate-base and tariffs for its proposed Interconnector 2.

4. The tariff for Interconnector 1 should be reset to the tariff that would now apply, had tariffs for Interconnector 1 been based from the start on “RPI-2%”.[23]

This scheme could in principle lead to a significant price difference between capacity on the two interconnectors. However, the indicative tariffs prepared by BGE suggest that this problem is unlikely to arise in practice. BGE’s indicative tariffs for Interconnectors 1 and 2 are higher than the Moffat entry charge published in the Bord Gais/Arthur Andersen study, but the indicative tariff for Interconnector 1 does not differ significantly from the indicative tariff for Interconnector 2.

Preventing Inefficient Bypass

A number of the currently proposed projects have the potential to bypass the BGE system and serve customers directly. Many of the gas users we spoke to expressed concern that such bypass would involve “cherry-picking”, i.e., would lead to higher prices for customers who remained with the BGE system. We have focused on a separate concern with cherry-picking: the postalised onshore system could give an artificial competitive advantage to inefficient bypass pipelines, and prevent competition among alternative projects from proceeding on the merits. We therefore propose a competitively neutral tariff mechanism that prevents the onshore postalised system from motivating inefficient bypass.

Example of Inefficient Bypass

The potential for inefficient bypass arises primarily because of the cross-subsidy inherent in any postalised scheme, as the following example illustrates.

Example:[24] suppose that the Irish Entry/Postalised Exit scheme were implemented, and that the onshore tariff was 2p/therm, but the true cost of serving a user in the Dublin area from the “entry point” of Interconnector 1 was only 0.8p/therm. The difference of 1.2p/therm would effectively fund postalisation for more distant users whose cost of service is higher than the 2p/therm postalised onshore tariff. Suppose also that a new pipeline could connect Dublin users to Interconnector 1 at a cost of only 1p/therm. Then the new pipeline is less efficient than BGE’s onshore service (the new pipeline’s 1p cost of serving the Dublin area is higher than BGE’s 0.8p cost), but the new pipeline can profitably take customers away from BGE (by charging 1.9p/therm, which is less than the onshore postalised rate). The new pipe in this example can compete only because disconnecting from the BGE onshore system allows customers in the Dublin area to avoid subsidising more distant users.

The Geographic Equalisation Public Service Levy

Bypass by inefficient pipelines is economically inefficient, wastes resources, raises prices for the majority of customers, and distorts downstream markets (e.g., the market for electric power) by creating artificial competitive advantages for those firms able to bypass. As we argued above, geographic equalisation of access charges is a legitimate goal of public policy, and the requirement to use a postalised exit fee is therefore a legitimate PSO. It would nonetheless be hampered or even frustrated if inefficient bypass were encouraged.

We therefore propose a competitively-neutral Public Service Levy (“PSL”) to be applied to all pipelines, for the purpose of funding geographic equalisation. In the example above, the levy on Dublin users would consist of 1.2p/therm. Thus the BGE tariff paid by the user would have two components: a 0.8p/therm cost-reflective charge, and a 1.2p/therm PSL. The revenues from the PSL would go into a special “Geographic Equalisation Fund”, used to fund postalisation across the onshore network. For example, suppose that the true cost of serving a user in Limerick is 3.5p/therm. The user would still pay a 2p/therm tariff, but again this would now have two components: a 3.5p/therm cost-reflective charge, and a negative component consisting of a –1.5p/therm PSL. This –1.5p/therm would be funded as a withdrawal from the Geographic Equalisation Fund. Tariffs would be designed so as to keep the balance in the Geographic Equalisation Fund close to zero over time.

Crucially, if the Dublin user switched from BGE to an entrant, the entrant would be required to pay the 1.2p/therm PSL into the Geographic Equalisation Fund. The inefficient entrant in the example given above would therefore not be able to engage in cherry-picking. To take a customer from BGE, it would have to charge no more than 2p/therm, from which it would have to pay over the 1.2p/therm PSL, leaving just 0.8p/therm revenue, which would be insufficient to cover its costs of 1p/therm.

Non-Discrimination

The PSL in no way discriminates against private pipelines. Rather it removes potential competitive distortions arising from postalisation, and thus ensures a genuinely level playing-field. The combined effect of the postalised BGE exit charge and the PSL is to create a competitive situation identical to that which would arise if BGE were to use fully cost-reflective (“distance-related”) exit charges. For example, the use of distance-related charges would mean that a BGE customer in Dublin would pay an exit charge that reflected the true cost of service. In the illustrative example above, this would correspond to a charge of 0.8p/therm, rather than 2p/therm. Any entrant would therefore have to compete with the 0.8p/therm. From the point of view of a potential entrant, this is identical to competing with a 2p/therm charge under the obligation to pay a 1.2p/therm PSL.

Finally, we note that this mechanism would not prevent bypass by a more efficient pipeline. If in the example of bypass given above, the new pipeline could serve Dublin area users at a cost of 0.5p/therm (which is less than the hypothetical 1p/therm cost incurred by BGE), then it could afford to charge the PSL levy, out-compete BGE, and still make a profit (for example, it could charge 1.9p/therm, leaving a 0.2p/therm profit.

Some parties have suggested modifications to the principle underlying the PSL. For example, it has been suggested that payments into the fund by parties other than BGE need only be large enough to fund expansion of the BGE “backbone” alone, and that BGE should then be responsible for any additional costs of postalisation that do not relate to the “backbone”. However, as the example above illustrates, efficiency requires that the PSL should be equal to the difference between postalised and cost-reflective charges. Any departure from this principle threatens to allow inefficient bypass.

PSL Subsidies to Private Pipelines

To ensure that this mechanism is non-discriminatory, the PSL would also apply to an entrant seeking to serve more distant parts of Ireland. In the example above, an entrant serving a customer in Limerick would be able to claim the same 1.5p/therm subsidy from the Geographic Equalisation Fund as BGE would receive to supply that customer.

This mechanism is entirely compatible with the Gas Directive. It involves no discrimination, since all potential pipelines attract the same PSL. The principles behind the PSL are objective, and support the valid goal of facilitating a legitimate PSO. Neither does the mechanism involve any form of state aid. Without the mechanism, BGE would be no better off, as the payments received from Dublin area users would be specifically designed to exactly balance the costs incurred in serving more distant users.

Some parties have expressed concern that the PSL mechanism should not represent an open-ended commitment to subsidise new infrastructure in remote areas. In our view this problem would only arise if postalisation itself were viewed as an open-ended commitment in Ireland, i.e, a commitment to extend infrastructure and provide service at the postalised rate irrespective of cost. However, that is not our understanding. We therefore recommend that the granting of a payment from the Geographic Equalisation Fund to any party, including BGE, be limited by a requirement for approval by the DPE/Regulator. It will therefore be necessary in the future to establish a mechanism and objective, non-discriminatory criteria for the approval of payments from the Fund. In particular, if the Department approves payment for serving a particular location, then the same payment (on a per therm basis) must automatically be available to all companies serving or wishing to serve that area.

Preventing Abuse by BGE

As a practical matter BGE would necessarily be responsible for calculating the PSLs, at least for the present. This leads to a natural concern that although the mechanism is in principle non-discriminatory, BGE might be able to discriminate in practice by distorting the PSL calculations. Specifically, BGE might misstate the extent of the PSL in specific areas simply to deter competitors. If a competitor served only the Dublin area, then an exaggerated PSL would lead to a reduction in its revenues and profitability. However, for BGE an exaggerated PSL would probably be approximately revenue neutral, reducing its revenues (net of PSL) in the Dublin area but increasing them elsewhere.

Consequently, it is essential that the scheme be implemented with rigorous safeguards against any potential abuse by BGE. Again, we recommend that BGE be given a “transparency obligation” in relation to the calculation of PSLs. We refer to our discussion above for details of this obligation.

Indicative PSL Charges

At our request, BGE has calculated indicative PSL charges based on our recommendations, for a number of alternative demand and supply scenarios. These are attached in Appendix 3.