Section 5 - Shannon-McManus Report

Section 5: Shannon-McManus Report

5.1 Introduction

The Shannon-McManus report was drawn up in an effort to reconcile the two sets of data from Sofregaz and BG/KeySpan. Both of these evaluated similar proposals for satisfying the island of Ireland’s gas needs over the period to 2025, but came up with somewhat different results.

Sofregaz concluded that the best way to provide gas to the island of Ireland was to supply the two markets separately, twinning the existing Scotland-Ireland pipeline for Ireland, and putting more compression on the infrastructure to Northern Ireland (see Section 3). BG/KeySpan on the other hand, estimated that reinforcing the SNIP and building a North-South interconnector was the best way to fulfil the island's needs (see Section 4).

Following discussions with British Gas plc and BGE/Sofregaz, and with their agreement, a databook was developed by the two authors – Dr Tom McManus of the Department of Public Enterprise and Professor RWE Shannon, formerly of British Gas. This covered standard costs for various elements of the infrastructure, and also specified the demand level, discount rate and methodology to be used.

Sofregaz and BG were required to evaluate three options, based on the databook. The options were:

All future gas needs for the island of Ireland, after the existing Scotland-Ireland interconnector is fully utilised, would come via Northern Ireland, i.e. with a N-S interconnector on the island of Ireland and a second SNIP as required. This is the equivalent of BG/KeySpan’s preferred option and of Sofregaz's UK2.

All future gas needs for the island of Ireland, after the existing SNIP interconnector is fully utilised, would come via a second Scotland-Dublin interconnector, with a S-N interconnector as required. This is the equivalent of Sofregaz’s UK1-S.N. option. The BG/KeySpan report did not evaluate this option.

The two parts of the island of Ireland would continue to be supplied by separate interconnectors from Scotland, with separate enhancements as required. This is the equivalent of Scotland-Dublin in BG/Keyspan and UK1 in Sofregaz.

Sofregaz and BG were required to estimate the Capex and NPV of each option, based on the agreed databook. Both concluded that Option 3 is the cheapest, Option 2 is the 2^nd cheapest and Option 1 is the most expensive. In each case Sofregaz came up with higher costs than BG. The following table sets out the results:

An initial analysis of the infrastructure investment requirements reveals two reasons why the options rank in this order –

Option 3 is the cheapest, because it includes only one substantial piece of pipeline – Scotland-Dublin. The other two options include two substantial pipelines – the interconnector from Scotland, and the interconnector between the North and the South.

Option 2 is cheaper than Option 1 because the main market is in the South. Therefore the S-N interconnector in this option can be smaller, and be built later, than in Option 1.

On the face of it, the results support the Sofregaz analysis in Gas 2025, since both organisations, using a standardised methodology, came up with the same ranking as in that report.

The results of an exercise such as this are dependent on the underlying assumptions. Among the more important assumptions in the databook were:

The analysis is based on a single, high demand scenario. In the case of Ireland, it is based on Gas World B, and in the case of Northern Ireland it is based on the high demand case used in the BG/KeySpan report.

A discount rate of 7 per cent is used. It is the same rate as was used to evaluate the first Scotland-Ireland interconnector, but is higher than the 5% test rate recommended by the Department of Finance for evaluating non-commercial public investment projects (commercial projects in the public sector attract a higher discount rate to reflect the risk profile of the project). It is also higher than rates currently being proposed in the UK by the regulators of the various utilities.

5.4.2 BG/KeySpan

The following table compares the BG figures in Shannon-McManus with those in the BG/Keyspan proposal.

Because of the lack of detail in the BG/KeySpan report, it is difficult to account for the differences. We asked BG to provide a reconciliation. They did so, based on the NPV data rather than the capex figures. The differences are set out as follows:

The main difference is due to flow analysis and methodology. This relates to the assessment of how much gas can be put through the pipeline. Shannon-McManus required that a particular model of flow be used, based on a transient flow model, and the AGA (American Gas Association) flow equation.

BG argue that a steady state model of flow is more appropriate. Steady state flow is estimated on the basis that the rate of gas flow stays the same throughout the day, whereas transient flow accounts for the fact that demand varies over the day. Using the former implies that more gas can flow through the same pipeline, and hence the required supply of gas can be delivered for lower investment cost.

The AGA flow equation reflects fully turbulent flow in pipes, which also reduces the implied throughput. BG argue that, especially in the earlier years when the system throughput is reduced, this equation will underestimate the amount of gas that can be carried.

It is outside the expertise of DKM to comment on the validity of BG’s arguments. 5.5 Summary

In the Shannon-McManus analysis, both BG and Sofregaz re-estimated the costs of the various methods of supplying gas to Ireland, using the same set of underlying assumptions.

The results support the conclusions of the Sofregaz report, that the cheapest way to supply gas to Ireland from overseas is via a second interconnector from Scotland to Dublin, and compression on the Scotland-Northern Ireland interconnector.

However, there are two important assumptions underlying the analysis:

There is only one demand scenario.., and

There is no new indigenous gas find.

A new gas discovery changes the picture even more, since the need for a new interconnector would be postponed, possibly by a considerable period of time.

To Section 6