FINANCING KEY MESSAGE

Transcription

1 3 FINANCING KEY MESSAGE Whilst a strong case can be made for more favourable financial parameters, we are proposing a business plan that accepts the debt index preferred by our regulator, Ofgem, (which will not cover the cost of our actual debt) and retains the present cost of equity of 6.7% and regulatory gearing assumption of 65%, although we expect to operate with less debt than that in our actual financing structure. We have arrived at this conclusion because, when we judge the package in the round, potential rewards associated with proportionate treatment have the scope to offset the downside that we will suffer in relation to the cost of capital assumptions. This is why we have proposed a financing package that takes this particular shape. It is similar to the one in our June 2013 plan, which our regulator judged to be welljustified in its November 2013 assessment, so we think Ofgem should have no difficulty in accepting it. March 2014

2 Financing DOCUMENT MAP WHAT YOU WILL FIND HERE This section sets out our proposition for the key financial parameters of the allowed cost of capital that will be required in the period. Our regulator, Ofgem, has challenged us to set out clear, unbiased and rigorous analysis to underpin our view and we present this analysis here. We also present the views of a variety of interested stakeholders, including debt investors, energy suppliers and customers, which have helped to inform our thinking and enable us to come to a rounded view of the appropriate cost of capital, as Ofgem requires. To meet our regulator s requirements we must assess the cost of capital taking account of the scenarios we have assumed, the cost forecasts we are making and the financial and operational uncertainties we face as a result. We also recognise that the regulatory uncertainty mechanisms and broader incentives contribute to an overall package of risk and rewards. We have factored this into our proposals, so that our stakeholders know we have considered the period in the round. In general, the rule that we have adopted is that we should bear the consequences of any risks and uncertainties that we face if we are better placed than customers to manage these risks. Our financing proposals take the distribution of risks between the company and customers into account, so that our investors can expect to achieve returns commensurate with the risks they bear on behalf of customers. In preparing our resubmitted plan, we have followed our regulator s guidance to assume it would make no changes to its methodology for setting the cost of capital in RIIO price controls, whilst also setting out what would need to change in our plan if it changes its methodology for setting the cost of equity. Since we now know that the methodology will change, we have ensured that we have provided a thorough justification for the alternative approach that we have set out. Northern Powergrid: Our business plan for March 2014

3 Financing This section at a glance KEY STATISTICS PERIOD OUR PROPOSED FINANCIAL PARAMETERS Cost of equity 6.7% Regulatory gearing 65% Cost of debt Indexed Overall cost of capital 4.1% Price reduction for domestic customers in 2015 Northeast Yorkshire 12% 6% KEY POINTS Our proposals reflect extensive stakeholder input and the net result of our plan is a cut in our average charges for domestic customers of 12% in Northeast and 6% in Yorkshire. The biggest single factor driving this reduction is the reduction in like-for-like costs in the plan period. Our shareholder has a long-term outlook, with a preference for capital growth over dividends and a preference for a conservative financing structure. We are already operating with over 200m more equity in our holding company than our regulator assumes in its notional financing structure for the regulated operating company. Our plan uses this strong balance sheet to bring the benefits of this financing structure to customers at no extra cost compared with Ofgem s assumption. The risks faced by equity investors in electricity distribution are increasing, and are higher than in gas distribution. The range of returns to regulatory equity in the next regulatory period is widening and acceptable credit metrics would not be sustainable without transitional arrangements for asset lives. We have adopted fixed capitalisation rates that average 71% and the move to a 45-year asset life is phased in over eight years. The overall package of returns available to us through the price control offered by Ofgem means that we can increase our allowed returns if we perform well. Our pension costs are efficient; we closed our final salary pension scheme more than a decade before some distribution network companies. Under Ofgem s previous methodology for setting the cost of equity, we were able to accept the cost of debt index by recognising the opportunities to view the deal in the round. Since our regulator has recently decided to change its methodology, this change needs to be evaluated as part of the overall proposition, taking into account the cost of debt, so that the overall answer remains fair when judged in the round. Northern Powergrid: Our business plan for March 2014

4 Financing CONTENTS 3.1 Overview The cost of debt The cost of equity The level of equity finance An alternative financing structure Capitalisation policies and asset lives Justifying our finance parameters Tax Pensions funding Stakeholder engagement The impact on customers bills We have included at annex GL.1 a glossary that explains the key technical terms and abbreviations used in our business plan. For more detail on how this plan differs from our June 2013 plan, please refer to annex G.12. Northern Powergrid: Our business plan for March 2014

5 Financing June Overview Our shareholder has a long-term outlook, with a preference for capital growth over dividends. Northern Powergrid is part of MidAmerican Energy Holdings Company, and ultimately owned by Warren Buffett s Berkshire Hathaway Inc. Our approach to financing the business is therefore guided by Berkshire Hathaway s longterm outlook and preference for capital growth over dividends. The last time Northern Powergrid Holdings Company paid a dividend to MidAmerican was 11 years ago, when we paid a dividend of 29m. This approach is a large part of the way that Northern Powergrid is able to confidently offer the financial flexibility necessary to deal with the uncertainties we face over the period. It also explains why the company has the highest credit rating of any of the standalone network groups. In part this financial strength is thanks to the fact that we have lower borrowings than our regulator, Ofgem, assumed when it reviewed the sector in 2009, which makes us more resilient to cost shocks. We believe this is particularly important in the light of the move to low-carbon technologies, where the rate of uptake is uncertain. Our financial flexibility will allow us to invest in the network assets required whatever the rate of low-carbon development. Our plan uses our strong balance sheet to bring benefits to customers in the form of lower prices. Our approach also means that we can use the strength of our company balance sheet to the advantage of our customers. We plan to use this strength to bring the entire price-drop forward into the first year of the new period, proposing a profile of revenues that are flat in real terms thereafter. This means lower prices in 2015 than we could otherwise offer, in recognition of the financial pressures being faced by some of our customers. We are also exploring other ways of using that strength, and have proposed changes to the regulatory rules for setting electricity distribution charges. These changes would give energy suppliers longer notice of changes to our charges than is currently the case, which they tell us would reduce their risks and allow them to offer lower bills to customers. We are continuing to work this proposal through with Ofgem and the other parties involved. This plan has been resubmitted taking into account guidance from our regulator. We last paid a dividend to our shareholder in 2003 In November 2013, our regulator expressed an intention to review its methodology for setting the cost of equity, rather than necessarily continuing the approach it used in previous price control reviews conducted under its new RIIO framework. Since the results of this assessment were expected to come late in the resubmission process, it asked us to prepare our resubmitted plan assuming that it would not change its methodology for the cost of equity. We have followed this guidance, and in doing so proposed a plan involving a very similar financing package to the one we submitted in June All of the statements about prices in this plan, and the majority of the financial sensitivity tests, are based on these financial assumptions. However, our regulator also asked us to say what would need to change in our plan if it were to change its cost of equity methodology in light of the Competition Commission s provisional findings in the Northern Ireland Electricity (NIE) price control reference. Our regulator has since decided that Northern Powergrid: Our business plan for Page 1 of 49

6 it should do so, and the alternative financing proposals consistent with this revised policy are set out in section 3.5 below. Our headline assumptions for the cost of capital are challenging, and are sustainable only when taken in the round. The key components of the cost of capital are the cost of debt and the cost of equity, and we have chosen to work with assumptions for each of these that we believe our regulator should have no difficulty in accepting. As far as the first of these is concerned, the cost of raising new debt is at an all-time low, and is expected to stay low for some time, but our existing borrowings were incurred some years ago when rates were higher, so the average cost of all our borrowing will be significantly higher than the up-to-date cost that our regulator s standard policy assumes. There are also greater risks for equity investors, and so the underlying cost of equity over is likely to reflect this and therefore be higher than that assumed by our regulator when the current price control was set. Given these risks, we feel that the regulatory assumption for the level of borrowing may be too high to provide sufficient financial headroom to cope with future shocks, so we will continue with our conservative financing structure and operate at a level of gearing that is lower than the regulatory assumption we propose to use in setting our charges. We propose to do this at no additional cost to customers. However, this package of assumptions is sustainable only when taken in the context of the potential upsides that are available to us if we are judged to have submitted a plan that warrants financial rewards through the proportionate treatment aspect of the price control review. Ofgem has made it clear that the information quality incentive applied to cost assessment results in the slow-track process will not provide these rewards, since a company meeting its cost efficiency benchmarks will receive no additional income, while the fast-tracked company (which was only just inside Ofgem s cost benchmarks) received a reward equal to 2.5 per cent of its costs, or 0.8 percentage points on its cost of equity. But it is important that there be no cliff-edge between the financial returns earned by fast-tracked companies and financial returns earned by equivalent companies that proceed via the slow-track. If there were a cliff-edge, it would distort incentives to submit well-justified plans, at future price control reviews as well as the current one. To avoid such distortions, cost-efficient companies should be able to earn rewards via the slow track that are similar to those received by Western Power Distribution. In theory, those rewards ought to have the potential to exceed the Western Power Distribution package, but in practice we accept that pragmatic reality is that any rewards at the slow-track stage will not match the fast track rewards. It is for Ofgem to determine the ultimate value of this reward, in the light of the commitment to treat companies proportionately in both process and financial terms. Our original plan was judged to be stronger than others in a number of areas, including the one submitted by Western Power Distribution (who were fast-tracked). The only reason given for not fast tracking our plan was a residual question in Ofgem s mind over the efficiency of some of our costs. Had Ofgem exercised its discretion to take a different, no less well-justified, approach to benchmarking we can readily see an outcome where we would have been fast-tracked. But that notwithstanding, the questions that were raised in relation to our costs by Ofgem s initial assessment of the business plan have been dealt with in full. In some cases, changes have been made. In many others our forecasts have been fully justified by detailed evidence or by the limitations of the benchmarking approach that prompted the question. Hence, we have assumed a proportionate treatment reward of 2 per cent of costs in this plan. This reflects a view that Ofgem should find our costs to be at, or close to, the level of efficiency it Northern Powergrid: Our business plan for Page 2 of 49

7 assigned to Western Power Distribution in its fast-track cost-assessment, and/or alternatively that it will judge that other aspects of our plan which were stronger than Western Power Distribution s warrant such a reward. Taking the scope for rewards associated with proportionate treatment in the round, we are willing to offer customers all of the benefits of our actual financing structure, at no extra cost compared with Ofgem s policies. In other words, we would not seek any additional customer contribution to our financing costs; rather we would set our charges based on our regulator s view, as set out by its policies and its recent decision for the gas distribution sector. Our base plan assumes our weighted average cost of capital will fall from 4.7% to 4.1%, and is likely to fall even further to somewhere in the range of 3.6% to 3.9%. The table below shows how the financial assumptions for which we would charge customers over the period compare with the assumptions currently in place for the period period period Cost of equity 6.7% 6.7% Gearing 65% 65% Cost of debt 3.6% Overall cost of capital 4.7% Indexed, currently 2.72%, likely to average 2.4% or even as little as 2.0% Currently 4.1%, likely to average 3.9% or even as little as 3.6% Asset lives 20 years Rising from 20 to 45 years Capitalisation rate 71% in Northeast* 73% in Yorkshire* 70% in Northeast 72% in Yorkshire Revenue profiling Rising in real terms Flat in real terms * Based on period allowances Table 1: Financial parameters in compared with the same parameters for Taken in the round, these financial parameters involve an exceptionally low overall cost of capital. No other UK utility infrastructure regulator has suggested setting a cost of capital lower than 3.8%, so the approach we are proposing under our base plan is only possible when the financial parameters are taken in the round, in light of the scope for rewards associated with proportionate treatment. Compared to some other recent utility price control proposals, the cost of equity assumed in our base plan is relatively high. Ofgem is however the only UK utility regulator to rule out setting a cost of debt that makes allowance for the cost of efficiently incurred debt issued more than ten years ago (or before 2013, by the time we reach the end of the price control period). Given the low rates available on debt since the financial crisis, this is likely to lead to extremely low debt allowances over the course of , well below the levels being contemplated by other regulators for price controls that end in 2020 or earlier, and well below the actual cost of our debt. We have consistently opposed our regulator s policy on the cost of debt because it has some significant weaknesses, and falls well short of being in the interests of current and future consumers. These issues are set out in full in section 3.2 below, on the cost of debt. However, looking past these Northern Powergrid: Our business plan for Page 3 of 49

8 issues to the overall cost of capital and the opportunity to take it in the round with the potential for rewards associated with proportionate treatment, our base plan uses our regulator s debt indexation approach to set allowances for the cost of debt. Under an alternative approach to setting financial parameters, our cost of capital would also fall significantly relative to the previous price control period. Although we were asked to assume in our base plan that our regulator would make no changes to its longstanding methodology for setting the cost of equity, our regulator recently decided to change this methodology, to take into account the provisional findings of the Competition Commission in the NIE price control appeal. We do not agree with our regulator s translation of the Competition Commission s findings on the cost of capital to the electricity distribution sector in Great Britain. There are strong reasons to believe that the longer time horizon applying to our price control, and the additional equity risks created by the strong incentives in place to deliver cost savings and service improvements that will benefit future consumers, would lead the Competition Commission to set a significantly higher cost of equity than the figure our regulator calculated that it would do. It is also quite plausible that the Competition Commission would calculate a value higher than the base value our regulator has set for those slow-track companies which are found to be somewhat less efficient than the fast-tracked company. However, we recognise that the 6.4% cost of equity awarded to the fast tracked company is likely to place an upper limit of the cost of equity that even a cost-efficient slow-tracked company will receive (short of significant new evidence which causes our regulator to re-evaluate its views on the cost of equity). So, since we have answered all of the questions our regulator raised with regard to our cost efficiency, under the alternative methodology our regulator has decided to implement for setting the cost of equity we believe our plan would warrant a 6.4% cost of equity (in addition to a reward associated with proportionate treatment, which we have assumed to be 2% of our costs). Even with this recognition of our cost efficiency in the non-financing expenditure areas of our plan, under our regulator s proposed alternative approach to setting the cost of equity, we would also require recognition of our efficiently incurred debt costs, with no arbitrary disallowance of debt simply because it was issued over 10 years ago (at a time when the company was operating and investing in the assets). Our regulator has recognised the sector wide impact of the ongoing equity risk created by its policy on the cost of debt. But it has not reflected the significant differential in the starting positions of the companies, which vary significantly in the extent to which their debt costs are likely to be underfunded by the proposed index. A change to its approach, and recognition of the cost of debt issued over 10 years ago, would move our regulator s approach to the cost of capital into line with the policy of other UK sector regulators, as well as the Competition Commission. Under this alternative financing structure, and taking into account other aspects of the overall settlement such as how interruptions incentive targets compare with current performance, and the assumed level of net real price effects, our overall package involves an effective allowed cost of capital which is exactly where we believe is appropriate for a highly cost-efficient slow-tracked company: close to, but below, the effective cost of capital allowed for Western Power Distribution. The financial settlement we are proposing therefore strikes a fair balance between customers and investors, at similar overall levels of cost to that Ofgem judged appropriate for the fast-tracked company, in light of the strength of our business plan and the fact we have answered all of Ofgem s questions relating to cost efficiency. Northern Powergrid: Our business plan for Page 4 of 49

9 Our proposals have been tested on a range of stakeholders, not just investors. Our willingness to use our balance sheet in the interest of customers has come from our dialogue with energy supply businesses. In our discussions, the more engaged energy supply companies were quick to highlight opportunities to help keep overall prices low and in particular they stressed the potential benefits that greater price certainty for them could ultimately bring for their customers, the end users of the network. Their messages chimed with the clear messages we have also heard from our domestic customers that they want us to do everything we can to keep prices low in the near future. The key ways this engagement affected how we developed this plan are as follows: Although our business sees pressures pushing the cost of equity upwards, and allowed gearing downwards, energy supply businesses have pushed back on these movements. They prefer to maintain the stability of these parameters, provided they remain near to the best estimate. Based on stakeholder feedback, we rejected the idea of offering lower prices over by reducing our financial gearing and moving immediately to 45-year lifetimes on new assets. Put simply, this would have cut prices now but increased the overall long-term costs, and on balance our stakeholders did not think this was worthwhile. Our plan includes a price profile that is flat in real terms, aside from legacy adjustments from the price control period or year to year adjustments to future revenues under our regulator s rules. In response to feedback received from some of our discussions with suppliers, we also considered alternative price profiles. In weighing up competing views, we decided that a price profile that was flat in real terms offered the best balance between charge reductions in and the price profile thereafter. Our plan includes a commitment to continue working towards developing industry arrangements that will give suppliers longer notice of changes to electricity distribution charges, something that they tell us will reduce the risk they face, and allow them to charge their electricity customers less. Following the commitment in our June 2013 plan, we brought forward proposals to change regulatory rules over the minimum notice period, with an increase from 40 days to 15 months. We will continue to support this initiative. Our plan sees our average charge for domestic customers fall by 12% in the Northeast and 6% in Yorkshire. The financial parameters we have arrived at, combined with the costs we have set out in our plan, would allow our average charge for domestic customers to fall by 12% in our Northeast area, and 6% in our Yorkshire area, relative to the last year of the current regulatory period, Further details are set out in Our plan on a page. These charges are based on the assumptions our regulator has specified to ensure comparability across different electricity distribution companies and annex 3.3 sets out a more detailed charging breakdown. The rest of this section of the business plan sets out in more detail our financial plans, covering the following aspects: The cost of debt. The cost of equity. The level of equity finance needed. Our policies for capitalisation of costs and asset lives. Northern Powergrid: Our business plan for Page 5 of 49

10 Tax liabilities. Pensions funding. How we have engaged with stakeholders over these plans. The amount we will charge customers. Northern Powergrid: Our business plan for Page 6 of 49

11 3.2 The cost of debt In order to keep costs to our customers low, a large share of assets are financed using debt, which is cheaper than other types of finance. Just like any other borrowing, our debt incurs interest costs, so we must have a clear idea of what interest payments we are likely to have to make over the plan period under a range of different scenarios. We also need a clear idea of the revenues we will receive from customers, from which we will make those interest payments. It is essential that we be confident that we can pay our debt interest over the long term, because otherwise the business would not be sustainable, earning a worse credit rating, which would make any new debt more expensive as lenders adjusted their pricing for a riskier business. Whilst we understand that it is the value of the regulatory settlement as a whole that determines the financial sustainability of the business, it is still relevant to consider the extent to which the regulatory assumption on the cost of debt reflects the actual cost of debt incurred. Our analysis suggests that we will not be recovering our full costs of debt, although the borrowings these costs relate to were incurred at good market rates at the time they were fixed. Given recent financial market conditions, we expect our actual pre-tax cost of debt over including issuance costs to be in the region of 6.1% nominal, or 3.3% real (subtracting 2.8% inflation), on average. The average cost of our existing borrowings, excluding issuance costs, is 6.3% nominal (or 3.5% real). The strategy decision our regulator took under its new RIIO framework for our price control, in March 2013, includes provision in its revenue allowances for debt costs, including issuance costs, based on a 10-year trailing average of a third-party index of debt costs. Based on equivalent assumptions for market conditions, we expect this to average at 2.4% (real, pre-tax) over , also assuming inflation at 2.8%. 1 This section has been prepared based on our regulator s guidance that we should assume it makes no changes to its methodology for setting the cost of equity, compared to the one used in previous RIIO reviews. In it we set out our supporting analysis, and also present the results under a range of alternative scenarios. Specifically, we provide further details on the following: The level of debt interest rates we currently anticipate we may be able to recover from our customers. The debt interest rates that we actually expect to have to pay on our borrowings and are planning to cover out of our overall revenues. How the interest rates we will have to pay, and will be able to charge customers for, could vary depending on market conditions. Of course, we now know that our regulator intends to make changes to its approach to setting the cost of equity. This section therefore serves to identify the issues we see with our regulator s preferred approach to setting the allowed cost of debt, while in section 3.5 below we set out what would need to change in our plan in light of our regulator s decision on the cost of equity. 1 The prices set out in this plan assume a 2.72% cost of debt, which is the level for , and which Ofgem requires all companies to use for comparative purposes. Northern Powergrid: Our business plan for Page 7 of 49

12 Our base plan adopts Ofgem s assumption of 2.72% for the index that sets the cost of debt... The prices set out in this plan factor in an allowance for debt costs at a rate of 2.72% (real, pre-tax). This is the assumption Ofgem has asked the whole industry to use to ensure plans can easily be compared. It is based on the level of a 10-year trailing average of two third-party indices of debt costs, commonly called the iboxx index of debt costs. 2.72% is the value being used to set prices in for energy transmission and gas distribution. Under our regulator s policy, our charges to customers each year will factor in the latest value of this debt-cost index, available at the point at which Ofgem confirms our allowed revenues for each year. This means that the debt assumption we make for the plan will vary over the course of the plan. but financial institutions agree that the index is likely to outturn lower than this. Both our own modelling and projections calculated by our banks suggest that this measure of debt interest rates is likely to be below 2.72% over We also need to have our own view so that we can be comfortable we have understood what is driving the results and can test sensitivities to the input assumptions. Although we have looked at a wide range of scenarios, the central case scenario below has been calculated using only two assumptions, listed below, and the debt index data we already have and know for certain: The first assumption is that interest rates on the nominal-debt index change in line with market expectations for nominal government 15-year gilts. We sourced these market expectations from the Bank of England. The second assumption is that retail prices index (RPI) inflation settles at 2.8%. Ofgem s cost-ofdebt allowance is well below our actual cost of debt These projections are set out below. The debt interest rates that are eventually factored into our charges will of course depend on how debt market conditions evolve over coming years, not on the 2.72% assumption we currently use, or on what people today think the iboxx index might do. As can be seen from the table, while our central scenario anticipates a recovery in debt rates, the banks who advise us all believe that our estimate is high. In this event, the cost of debt included in our charges will fall, acting to reduce prices further Plan assumption 2.72% 2.72% 2.72% 2.72% 2.72% 2.72% 2.72% 2.72% Northern Powergrid central scenario Average of three financial institutions 2.6% 2.5% 2.5% 2.5% 2.4% 2.2% 2.2% 2.3% 2.4% 2.3% 2.1% 2.0% 1.8% 1.6% 1.6% 1.5% Table 2: Forecast debt rates based on the iboxx index Northern Powergrid: Our business plan for Page 8 of 49

13 Real cost of debt (Ofgem method) The debt interest rates we will pay will be substantially higher than we will be allowed to recover. We have an existing book of long-term debt that we have issued over the last 19 years, some of which will mature and be repaid over , but some of which is locked in for many more years. We also expect to issue more long-term debt during , at the best market rates we are able to get at the time it is issued. To model our future debt interest costs, we have assumed that we will be able to issue this debt at the prevailing spot rate on the iboxx index, including any issuance costs we incur. Full details of our assumptions and calculations are set out in annex 3.1 on the cost of debt. Our key finding is that we expect our actual debt costs to exceed the amount that our regulator will allow for the cost of debt from , leaving us with additional costs that must be met by our shareholder. The figure below shows the results of the analysis, under our central scenario for future interest costs. 3.20% 3.00% 2.80% 2.60% 2.40% 2.20% 2.00% Book debt, inc. historical issuance costs Allowance using iboxx Figure 1: Forecast debt costs At 65% financial gearing, this gap between allowed and actual costs of debt converts to 80 to 130 basis points on the return on regulatory equity, with an average gap of around 110 basis points. This gap is driven by the fact that we took long-term debt in the past when we needed it to run the business. We issued this debt on the open market, so are confident we couldn t have secured better rates, and we have taken advantage of more innovative sources of funding, such as the European Investment Bank or debt insured by third parties in exchange for a fee, when these have been available and reduced overall debt costs. The weighted average maturity for these borrowings is 13 years, which seems more than reasonable given that the debt funds electricity distribution assets that can easily last 45 years. And the average coupon is low enough that any previous Ofgem- Northern Powergrid: Our business plan for Page 9 of 49

14 allowed cost of debt would have paid for it. All of this leads us to believe that our existing debt was efficiently incurred. The further contributor to the underfunding is that only 10 years are used to calculate a weighted average, and that interest rates have fallen to extremely low levels and stayed there for much longer than most experts expected. No one could have known this would happen when we issued our debt. And, in fact, rates could just as easily have risen over time. We therefore continue to believe that it makes sense to finance long-term assets with long-term debt. We have modelled several scenarios, but the shortfall is similar under each of these. As well as modelling our cost of debt under a base set of assumptions (our central view), we have undertaken significant scenario testing to assess whether different outturn financial market conditions could lead to significantly different results for actual and allowed costs of debt. These scenarios have included: allowing for different paths of underlying interest rates, to cater for either faster than expected recovery in the economy, or significant further delays in any recovery; and allowing for different levels of inflation (still maintaining consistency between expected inflation and actual inflation), at levels of 2.5% or 3.2%. Our conclusions for the actual cost of debt, and the cost of debt calculated according to a 10-year trailing average on the third-party index, are similar under these alternative scenarios: If interest rates rise faster than expected, the gap between actual and allowed debt costs will be slightly smaller, but it would still be material at around 110 basis points when converted to returns on regulatory equity. There would be a bigger reduction in the gap if rates rose to higher levels than our scenarios assume. If rates rise more slowly than expected, the gap between actual and allowed debt costs will be slightly larger than would otherwise be the case, as pre-financial-crisis data would rapidly drop out of the 10-year trailing average, and any nominal-rate debt issued before the financial crisis would be out of the money. There would be larger increase in the gap if rates failed to rise to the levels our scenarios assume. The inflation-rate assumption makes a small (5 basis points) difference to the underfunding on the real cost of debt, but a higher inflation rate reduces the underfunding on the nominal cost of debt during the period, since higher inflation would contribute to covering the cost of debt that we issued on nominal terms before the financial crisis. The scenarios above don t consider what would happen if the market breakeven expectations for inflation became structurally stuck at levels above true inflation expectations. If this did happen, we would face an even higher shortfall on our cost of debt, but Ofgem has considered the evidence and concluded that its proposed approach does not result in a downwardly biased estimate of the real cost of debt. Overall the view that there is likely to be a significant underfunding of our actual cost of debt under the index is robust to the scenario chosen. The robustness of this finding contributes to our conclusion that we could accept the indexed cost of debt only as part of an overall outcome that is assessed in the round, under the methodology our regulator used to set the cost of equity under previous RIIO price controls, and with other benefits offsetting this underfunding. Northern Powergrid: Our business plan for Page 10 of 49

15 We have consistently highlighted our concerns over this approach to setting allowances for debt costs Northern Powergrid has opposed our regulator s proposed approach to setting the allowed cost of debt consistently throughout the RIIO process, even before our own price control commenced. Concerns we highlighted included the following. It is inappropriate to set the cost of debt in a way that automatically rules out funding any debt that was issued more than 10 years ago. This is especially true since it makes natural sense to fund long term energy network assets with long term debt. The data our regulator believed demonstrated that its chosen index provided sufficient funding for the industry s debt costs had a serious flaw. Namely, it wrongly attributed the debt costs achieved by third parties with AAA credit ratings to the energy companies that paid them separate fees for providing insurance against default to bond holders, meaning the debt received the credit rating of the insurance company. While this approach was efficient for the companies to pursue, it means that the headline interest rates compared by our regulator to the level of its index didn t take into account the true costs of debt of the industry. This led to the mistaken belief that the chosen index provides headroom over the spot rates of debt that the industry can achieve when issuing new debt. The RPI inflation breakevens used by the index do not measure only expected RPI inflation, but will also be elevated on account of inflation risk premia (which is likely to be heightened in current financial market conditions) and other distortions driven by rules that mean pension funds must invest in inflation protected assets. By building this risk premium into allowed debt costs (by reducing them), it effectively assumes that the industry can access financing on these terms. But since utility financing is based predominantly on debt which pays a nominal rate of interest, there is a mismatch between the index assumptions and what can actually be achieved. Despite highlighting all of these issues, only one of our observations led our regulator to make changes to its index, which related to a technical inconsistency between how the Bank of England s inflation breakevens are calculated, and how they were being used by our regulator in converting the index to real terms. But since all of the transmission and gas distribution network operators accepted our regulator s overall financing package, which was based on the cost of equity methodology in place at the time, none of these issues has been tested. and we do not believe the approach is in the long term interests of consumers The drawbacks of the index are ultimately bad for customers as well as investors. Our regulator recently admitted, for the first time, that it had added to equity risk in the sector by imposing the index on an industry which had already issued fixed rate debt at a significantly longer average tenor than the 5-years that a 10-year trailing average implicitly assumes (since it gives equal weight to debt costs between 1 and 10 years ago). This equity risk would be added even if a company happens to have a debt cost that matches the current level on the trailing average. Previously our regulator had insisted that the index would reduce risk over the course of an 8-year price control period, relative to the alternative assumption of a fixed allowance. But this is only true if a company is able to match the financing structure assumed by the trailing average, which we, and other companies in our sector, are already unable to do due to previous financing decisions. As our regulator highlights, the index provides the strongest possible incentives for companies to cut their cost of debt. It is often a worthwhile trade-off for customers to accept higher equity finance costs in exchange for the lower long-term costs that strong incentives will deliver (for example on Northern Powergrid: Our business plan for Page 11 of 49

16 cost incentives, or interruptions performance). But in this case the strong incentives will never deliver any benefits for customers, since our regulator has committed itself to maintaining the indexed approach to the cost of debt, which means companies will retain any cost savings indefinitely. It is difficult to see how, now that our regulator has acknowledged that the debt index adds to equity finance costs, it can maintain that it is meeting its primary duty to take into account the interests of current and future consumers through this particular policy. Other UK sector regulators, and the Competition Commission, have taken an approach that provides incentives for companies to minimise debt costs on an ongoing basis, while still setting the cost of debt at subsequent reviews in a way that passes on any benefits from the incentives to customers. There are other, more pervasive, problems with implementing an approach to setting debt cost allowances like that our regulator is proposing. All of our decisions to issue debt were taken before our regulator decided to implement a 10-year trailing average in our sector, and all but one were taken before our regulator had started consulting on the form of debt indexation to apply; we were not able to take our regulator s approach into account in our financing decisions. By taking a retrospective decision not to cover these financing costs, our regulator may not be meeting its duty to ensure that the company is able to finance its functions. Moreover, the decision to rule out funding efficiently incurred debt simply because it was issued more than 10 years ago goes against the approaches taken by other UK sector regulators and the Competition Commission. This recognition on the part of other regulators is an important component of the regulatory stability that underpins decisions to invest in the industry. By retrospectively ruling that efficiently incurred debt will not be financed, our regulator is likely to add further to the equity risk faced by the industry. Equity investors are likely to perceive a heightened risk that our regulator will retrospectively determine that other efficiently incurred costs should not be allowed for arbitrary reasons. Moreover, under our regulator s view that it is acceptable to impose the index because it consulted on it and then announced it before the price control starts (expressed on page 12 of its equity returns decision), it would be perfectly acceptable for it to change its indexation approach (or remove it altogether) provided it consulted and announced its decision before the next price control. This is not a sustainable policy position from the point of view of regulatory credibility and commitment and we would be surprised if such a position were to be accepted and become established as good regulatory practice. although it still might be possible for it to form part of a settlement that properly allows us to finance our functions. Of course, no element of a price control settlement can be evaluated in isolation of the others, or indeed the costs that a company expects to incur, and any companies that accept the index may do so because it matches their actual debt costs more closely than ours, or because other elements of the settlement carry a value which outweighs any under-funding on the cost of debt. This is likely to be why transmission and gas distribution companies accepted the use of debt indexation in their price controls without an appeal, and also why it is likely the fast-tracked electricity distribution company will do the same. It is also a key reason behind the financing package we propose under both our base plan, and the alternative approach to financing our business our regulator asked us to develop in the event it decided to change its methodology for setting the cost of equity. Northern Powergrid: Our business plan for Page 12 of 49

17 3.3 The cost of equity This section of our plan has been prepared based on the guidance of our regulator that we should assume it continues to use the same methodology for setting the cost of equity as it used for previous RIIO price control reviews. In doing so, we assessed recent developments on the cost of equity; as a result, many of the issues discussed in this section are relevant regardless of whether there is a change in the methodology. However, we now know that our regulator has decided it will change its methodology, and in section 3.5 below, on an alternative financing structure, we have set out what would need to change in our plan in light of this decision, again as per the guidance from our regulator. There is a strong case for an increase in the allowed cost of equity of 6.7% that was used at the last price-control review. Our long-term investments in network assets are part-funded by equity finance. This equity finance represents a long-term commitment from our shareholder, MidAmerican Energy Holdings Company, to long-term assets. It also carries the bulk of the risk facing the company. The allowed return on that equity must therefore reflect the long-term nature of the investment, and also the risks it carries. There is strong evidence that the cost of equity in the next eight years will be higher than the current allowed cost of equity of 6.7% (real, post-tax). We see a number of influences acting to move the cost upwards: Risks to equity investors in electricity distribution are increasing, because of changing demands being placed on the network, and because of the fact that more money will now ride on our performance. We believe this makes sense and will deliver better long-term value for customers by placing the onus on us to deliver good outcomes. We are therefore proposing no changes to our regulator s proposals for incentives and uncertainty mechanisms. But, while the overall balance is one we feel represents good value to customers in the round, the package does mean our equity investors face more risk in the period than they did in the period. We face higher capital expenditure requirements, relative to our regulatory asset value, than some other network companies, which is an important factor in how our regulator assesses the risk of different sectors. We lack a safeguard that is available to other network companies. Our regulator has decided not to give electricity distributors the same disapplication rights as other network companies that enable those companies to force a reconsideration of the price control settlement by the Competition Commission before it has reached the end of its expected eight-year life. Estimates of the cost of equity using standard empirical techniques as adopted by our regulator at previous RIIO price controls could support a value of up to 7.2%, and the factors above suggest that a figure in the region of 6.9% could be justified, in combination with gearing slightly below 65%. Despite these influences, we have adopted a conservative figure for the cost of equity, as well as set aside our concerns about underfunding on the cost of debt, in favour of a view that the headline figure for the cost of capital must be assessed in the round, together with other benefits available in the price control. In doing so, we have explicitly recognised stakeholder views that the cost of equity Northern Powergrid: Our business plan for Page 13 of 49

18 should not rise relative to the current levels, 6.7% post-tax real, a view we can accommodate since this figure remains close to the slightly higher one that the evidence could justify. The sections below set out further details of the three main points supporting our view of the cost of equity. Comparisons with other energy network sectors indicate that a higher cost of equity for electricity distribution would be justified Ofgem has recently set a cost of equity for a number of other energy network sectors. These settlements are set out in the table below. Gas distribution Gas transmission Electricity transmission Risk-free rate 2.0% 2.0% 2.0% Equity risk premium 5.25% 5.25% 5.25% Equity beta Cost of equity 6.7% 6.8% 7.0% Table 3: Ofgem cost-of-equity decisions for other energy network companies We joined with our peer group companies in the Energy Networks Association (ENA) to commission Oxera, an economic consultancy, to compare the risks faced by the electricity distribution sector over with those faced by other sectors. Oxera found that the equity risk in the electricity distribution sector is likely to be higher than in the gas distribution sector, based on an assumption that gas distribution faces a similar level of underlying cost risk. It also concluded that levels of equity risk may be closer to those seen in electricity transmission, again assuming a similar level of underlying cost risk. These conclusions are primarily due to the likely scale of expenditure relative to the regulatory asset value (RAV) for electricity distribution over compared with these other sectors, the different licence disapplication conditions across the sectors, which give electricity distributors less protection, and the higher exposure of electricity distributors to pension cost risk. This report on relative risk is published on the Ofgem website at its RIIO-ED1 web forum. and the ratio of our capex to regulatory asset value is higher than for most companies in the recent network price-control reviews. In its recent reviews of transmission and gas distribution, Ofgem also found that the capex-to-rav ratio was an important factor to consider when looking at relative risk. The chart below shows how these ratios compared across various sectors. The three electricity transmission companies (to the left of the chart) were allowed a 7.0% cost of equity, while the gas transmission company (NGG) was allowed a 6.8% cost of equity, and the gas distribution companies (GDNs) were allowed a 6.7% cost of equity. In the first year of the period, the capex-to-opening-rav ratio in this plan is on average 12% across our two licensees. By the last year, this ratio is 9%, giving an average over the period of 10%. Based on these figures alone, an allowed cost of equity somewhere between 6.7% and 7.0% would be appropriate for electricity distribution over the period if we apply the capex-to-rav ratio from other RIIO price controls. Northern Powergrid: Our business plan for Page 14 of 49

19 25% 20% 15% 10% 5% 0% Scottish Hydro Electricity Transmission Scottish Power Transmission National Grid Electricity Transmission National Grid Gas Transmission Gas distribution (average) Northern Powergrid Northeast Northern Powergrid Yorkshire Figure 2: Average capex-to-rav ratios across energy network sectors Source: Figures for other sectors are taken from Ofgem Electricity Transmission Final Proposals, The figures include base, volume driver and strategic wider works capex, and also gas distribution repex. The figures also add weight to Oxera s view that electricity distribution is likely to face a higher cost of equity than gas distribution. The substantial capex demands placed on electricity transmission differentiate that sector and suggest that, using capex-to-rav alone, the cost of equity for electricity transmission must be higher than that for electricity distribution. This analysis and Oxera s conclusions are based on the assumption that underlying cost risk is the same across all sectors. We support that assumption, although it is important to note that electricity distribution companies have a unique role in supporting the transition to a low-carbon economy. Electricity distributors do not have the same rights as other network sectors to force an early appeal of an onerous price control to the Competition Commission. One of the underlying reasons for Oxera s findings is that electricity distribution companies have a significantly weaker ability to seek changes to price control settlements before their end date than the companies operating in these other sectors. In the transmission and gas distribution sectors, companies have an unfettered ability to appeal to the Competition Commission if Ofgem refuses to relax a price control that turns out to be unduly onerous, at any time during the eight-year period it covers, even if they are not in financial distress. In electricity distribution, companies have no such automatic ability to appeal to the Competition Commission. Northern Powergrid: Our business plan for Page 15 of 49

20 The risks faced by equity investors in electricity distribution in the next regulatory period are increasing compared with the present period. Equity investors in electricity distribution are facing upward pressure on risk as a result of the transition to a low-carbon economy. Although new uncertainty mechanisms are being introduced by our regulator, these (rightly) do not remove all the additional risk because it is important to maintain an incentive on us to keep costs as low as possible. The net effect is therefore upward pressure on the risks being borne by equity investors. We agree that the allocation of this risk is appropriate and would seek to make sure that this is explicit and accommodated in the round in the regulatory settlement. There is significant uncertainty. We are proposing no additional uncertainty mechanisms to protect us from the risks over the levels of uptake of low-carbon technologies, as we are well placed to control them. We believe that it offers better value for customers for us to have the strongest possible motivation to control our costs. But this means equity investors will be exposed to up to 21m in asset investment costs 2 before additional cost allowances can be applied for. Assuming this level of extra investment is required over the years from 2020 to 2023, the annual post-tax equity return for those three years could reduce by around 0.7 percentage points. Our involvement in the roll-out of smart meters means undertaking new activities, and taking on the risk associated with them. Once again, we are well placed to control many of these risks, and we are proposing no additional uncertainty mechanisms beyond those proposed by our regulator. But this still leaves some significant risks with equity investors. For example, in line with our regulator s policy, we have also made no allowance for data charges beyond March 2020, even though the roll-out will not yet have been fully completed at that point. The Department of Energy and Climate Change (DECC) estimates of these costs for the remaining three years have ranged from up to 2.2m (the latest figure) to as high as 24m (the estimate The transition to a lowcarbon economy is a source of uncertainty. Energy networks must be ready to drive this change and manage the uncertainty. early in 2013). This means that, in the years from 2020 to 2023, the annual post-tax return on equity could be reduced by around 0.5 percentage points if the final charges are at the levels estimated earlier in 2013, rather than at the level of the latest estimates. The Oxera report commissioned by the ENA also assessed whether equity risk was increasing over the period, compared with risks for equity investors in electricity distribution over the period. Oxera concluded that asset risk was increasing by 5% to 20% due to the factors they could quantify. Details of Oxera's risk-assessment framework can be downloaded from the Ofgem website at its RIIO-ED1 web forum. At the current 65% level of gearing, this would indicate a cost of equity of 7.0% or higher. The specific examples given above and the Oxera study of this issue all suggest to us that the risks faced by investors in electricity distribution are rising relative to their present levels. They therefore support a cost of equity higher than the 6.7% (real, post-tax) allowed for in current network charges. 2 The figure is quoted after tax and after the application of the efficiency-sharing factor that apportions most of any overspend against price-control assumptions to the company rather than its customers Northern Powergrid: Our business plan for Page 16 of 49

21 Changes to how data is gathered for price statistics have no implications for the cost of equity Our allowed return is set in real terms (i.e. before inflation), with inflation being allowed for through increases in the value of our asset base over time (and allowed revenues). This means that the price statistics calculated by the Office for National Statistics (ONS) are an important input to the price control calculations. In 2010, the ONS changed some of its guidance for the people who collect price data from shops. These changes addressed the fact that price collectors were frequently finding some items to be out of stock. However, this also fed through into the level of inflation that is measured for any given set of price changes in the shops, and increased the gap between RPI and CPI inflation that is caused by the RPI using a different formula to calculate inflation. This issue has not been well understood by many market commentators. The potentially permanent, structural, nature of the change has led to suggestions that the allowed cost of equity for regulated utilities should be reduced, since the formula effect gap between RPI and CPI inflation has now risen. However, a full analysis reveals that the issue does not warrant any adjustment for the following reasons. Many commentators have over-stated the scale of the impact on the RPI formula effect. The ONS publishes a decomposition of the gap between RPI and CPI inflation, with one component entitled the formula effect. Many commentators have used this data to quantify the scale of the change in the formula effect. But the ONS data actually measures the size the formula effect would be if RPI used the same weights as CPI. CPI actually places much higher weight on clothing, which causes the lion s share of the formula effect, so this ONS data significantly overstates the true formula effect, and its increase. The true effect can be closely approximated by the difference between RPI inflation and RPIJ inflation, which shows that it increased by around basis points after the changes in This is less than many people thought. The ONS is actively considering changes to reverse some of the impact of the changes in The ONS has been actively considering changes to its data gathering routines that will reduce the formula effect, and also reduce CPI inflation, based on the patterns of clothing prices that tend to be seen in the shops. It ran a major trial in 2012 that demonstrated these changes could reduce the formula effect. Its work plan for 2015 includes potentially making these changes. There isn t a long enough series of data on the formula effect to support an adjustment. Data on the size of the formula effect is only available back to It therefore isn t clear what level it took on before this date, so it requires significant assumptions to adjust estimates of equity returns based on historical data from before Ofgem s own consultants highlighted this point in their report and proposed an adjustment of only 0.25 percentage points on grounds of caution.3 Moreover, the weight on clothing in RPI has fallen significantly over the last 25 years. In 1989 it was in around 70% higher than it is now, so this reduction is likely to have contributed to a declining formula effect over time, even after the recent changes took place. There has been an offsetting and permanent structural break in council tax increases. Over , council tax increases averaged 6% per annum, and this contributed to the 3 Wright and Smithers, 2014, The cost of equity capital for regulated companies: a review for Ofgem, page 11 Northern Powergrid: Our business plan for Page 17 of 49

22 difference between RPI inflation and CPI inflation. The Localism Act (2011) means increases over 2% will need to be approved by a local referendum. Since approval will be difficult to secure, the average increase in future is likely to be around 2%. This reduction offsets around 15 basis points of the increase in the formula effect. Financial market instability over recent years makes it impossible to quantify how and to what extent these changes have influenced investor expectations. The changes to how price data is collected have not necessarily been well understood and it could take some time before their overall effect becomes apparent in financial market data. Some aspects of the changes, such as how they have impacted the measurement of CPI inflation, have received very little discussion. These issues will not affect the allowed revenues we receive, since the Bank of England inflation target remains at 2% (and so only the long run gap between CPI and RPI inflation matters for revenues). If investors fully appreciate the lower underlying price pressure in the economy, this may push down on real rates of return. But if they do not, then the tightening of monetary policy required by the Bank of England to hit its inflation target could lead to an increase in real rates. Moreover, both of these effects appear likely to be mitigated by the further changes to data collections routines that the ONS is proposing to make in These complex effects are likely to take some time to unwind in financial markets. They are unlike the ONS decision not to change the formula used to calculate the RPI at the start of 2013, where there was a clearly defined event which received significant attention, and which led to an immediate response from financial markets. It is not possible to say with any certainty the extent to which financial markets have taken these wider issues into account, since money markets have been subject to much more material issues (such as the flight to quality from Eurozone periphery countries) over the same time period. Taking all these factors together, including the fact that the ONS plans to make changes which will act in the opposite direction as early as 2015, and bearing in mind the offsetting effects associated with the structural break in council tax increases, there should not be any reduction in the cost of equity on account of this issue. while recent transaction premia tell us little about the cost of equity. Some regulated utilities have recently transacted at valuations above the regulated asset value. However, there are many factors that make it impossible to infer anything about the cost of equity from these transactions. Transactions often cover non-regulated businesses, or involve significant pension adjustments (for example where the vendor has made exceptional pension contributions to reduce deficits). This can make it difficult to identify the proportion of the transaction premium that relates to the RAV. Transactions often target underperformers, where improvements in performance and reductions in cost have the greatest scope to deliver additional returns for future shareholders. In an open bidding process these future returns will translate into a higher price paid now. Transactions in recent years have taken place at times of exceptionally low spot rates on the cost of debt. The regulatory frameworks of UK regulators such as Ofgem, the Competition Commission, Ofwat and the Civil Aviation Authority take into account, to varying degrees, the historical cost of issuing debt (since companies issue long term fixed rate debt). They Northern Powergrid: Our business plan for Page 18 of 49

23 also took place at a time when relatively high debt allowances had been given for new issuance, reflecting market expectations of high rates at the time controls were set. When spot rates are below allowed debt costs, the asset will be worth more than its RAV. Although transaction premia were not within the scope of the report written by Ofgem s consultants to support the decision on the cost of equity methodology, they did consider it in brief. They concluded that the most obvious explanation for transaction premia was the equity beta element of Ofgem s calculation. 4 This conclusion is however incorrect: even if a transaction premium resulted from the allowed cost of capital being above the spot cost of capital, the most obvious explanation would be the cost of debt, which accounts for a significantly higher proportion of the allowed cost of capital than the cost of equity. The allowed cost of debt (and the actual cost of debt of companies) depends on debt issued historically. Debt rates have tended to fall progressively further over the past 20 years. Since transactions are typically valued using the spot rate of debt, rather than the cost of debt actually faced by a company which has been operating in the industry for some time, a transaction premium will be observed whenever the spot cost of debt happens to be below the regulatory allowance. This will have been the case in the majority of transactions over the last decade, except for a short period during the financial crisis. The effect will have been a particularly strong effect over much of 2010 to 2013, since spot market debt rates have been at exceptionally low levels relative to the regulatory allowances used to set future cashflows, since these necessarily cover the cost of debt issued historically. Overall, these issues mean that we have continued to calculate the cost of equity based on long term estimates of its underlying parameters, rather than attempting to back out highly inaccurate estimates from transactions over the last few years. But evidence of equity returns confirms that the current 6.7% remains within the reasonable range. In 2013 Ofgem estimated a range for the cost of equity using the standard empirical framework for estimating it based on market evidence, the capital asset pricing model (CAPM). Extensive work was undertaken on evaluating the appropriate range. Our regulator requires that our proposals for the cost of equity must take into account empirical evidence from markets and established economic models, as well as factors we have already considered above, such as relevant comparators and precedent. We agree with the analysis Ofgem undertook in developing this range and have therefore used Ofgem s range to inform how we have developed our business plan proposals on the cost of equity under our base plan, as well as an assessment provided by Oxera on this range and where electricity distribution might lie within it. The range is set out below, along with the levels assumed in the current electricity distribution network charges. 4 Wright and Smithers, 2014, The cost of equity capital for regulated companies: a review for Ofgem, pages Northern Powergrid: Our business plan for Page 19 of 49

24 period period range low period range high Risk-free rate 2.0% 1.7% 2.0% Equity risk premium 5.25% 4.75% 5.5% Equity beta Cost of equity 6.7% 6.0% 7.2% Table 4: Cost-of-equity range, based on a detailed evaluation of empirical evidence by Ofgem Along with other electricity distribution companies, we commissioned Oxera to undertake an evaluation of the cost of equity range proposed by Ofgem in its strategy consultation, and provide a response to our regulator s consultation on our behalf. Its report is published on Ofgem s web-page showing responses to its strategy consultation. Oxera confirmed that the range was appropriate and also found that, although the exact cost of equity would depend on specifics of company business plans, a point estimate towards the upper end of the range was appropriate. Finally, we have also cross-checked our view of the cost of equity we propose against long-term returns equity investors tend to be able to earn in the average year. These long-term estimates of returns are typically in the region of 7.0% (post-tax, real). For example, the Credit Suisse Global Returns Yearbook cites data showing that the arithmetic average of returns on global equities from 1900 to 2011 has been 6.9%. The evidence presented above, on risks we face relative to other energy network sectors, and the increases in risk we face going forward, also point to a cost of equity towards the top end of the range. The cost of equity figures recently proposed by the Competition Commission and Ofwat are at lower levels Some regulators have recently suggested that some other regulated utilities may face a lower cost of equity than we have proposed. This includes the Competition Commission in its provisional findings for NIE s appeal of its price control, and Ofwat in its guidance on risk and reward for the water sector. Neither set of proposals has yet resulted in a price control being established with a lower cost of equity than the one we propose in our base plan. But we recognise that the proposals of these other regulators raise valid questions amongst our stakeholders as to why the cost of equity proposed in our business plan is higher. There are, however, very good reasons why these approaches are not transferable to this price control review, given the longer control period, the greater risk to cashflows that are built into this review and the much tougher treatment that Ofgem is proposing in relation to allowances for debt cost. The table below summarises the figures proposed by the Competition Commission and Ofwat, and compares these with our base plan. 5 Northern Powergrid: Our business plan for Page 20 of 49

25 Northern Powergrid s business plan Competition Commission for NIE (provisional) 6 Ofwat guidance for water sector Cost of equity 6.7% 4.8% 5.65% Gearing 65% 50% 62.5% Cost of debt Cost of capital Indexed, average likely to be 2.0% to 2.4% Average likely to be 3.6% to 3.9% 3.4% based on actual expected debt costs 2.75%, at or around expected debt costs of large water companies 4.1% 3.85% Table 5: Cost-of-capital assumptions of Northern Powergrid compared to those used by other regulators This table highlights four issues. The cost of equity we have proposed is higher than the cost of equity proposed for some other regulated utilities by the relevant regulators. The other regulators have proposed lower levels of allowed gearing. This is part of the reason for their low assumed cost of equity. If they had assumed 65% gearing, the Competition Commission would have proposed a cost of equity of 6.1% for NIE, while Ofwat would have proposed a cost of equity of 6.0% for the water sector. Each of these other regulators has proposed a higher cost of debt than the average that is likely to be in allowed under our plan over , and has set the cost of debt at a level that covers company costs (unlike our regulator s preferred debt index approach). The overall cost of capital proposed by these other regulators is similar to, or higher than, the level Northern Powergrid has proposed in its business plan. In other words, the financial proposals of these other regulators are closely aligned to our own, when taken in the round. There are also some material differences between the sectors that these financial proposals relate to. The table below summarises some of the most significant of these issues. 6 The Competition Commission proposed to use the 60 th percentile of its range for NIE s cost of equity, so we have presented the 60 th percentile of its ranges here. Northern Powergrid: Our business plan for Page 21 of 49

26 Northern Powergrid s business plan Competition Commission for NIE (provisional) Ofwat guidance for water sector Years covered Cost risk exposure Delivery risk exposure Financial risk exposure Higher: up to 65% post-tax Higher: bottom end of RORE range is around zero Higher: trailing average does not match pre-existing funding profile Lower: 50% pre-tax i.e. 40% post-tax Lower: limited incentive downside Lower: fixed allowance with 5 yearly re-set matches funding profile reasonably well Typically lower: proposed by companies Lower: bottom end of RORE range at cost of debt Lower: fixed allowance with 5 yearly re-set matches funding profile reasonably well Table 6: Comparison of equity risks under Northern Powergrid s business plan compared to equity risks in other sectors The table shows that our proposals relate to a longer time horizon than the proposals of these two regulators. The longer time horizon covered by our plan means it is more likely that interest rates will recover, and so allowed returns should be higher. It also shows that the price control framework for electricity distribution places more onus on the companies to take on and control delivery and cost risk where they are best placed to do so. This should deliver overall benefits for customers, through lower costs and better services. But it exposes equity to slightly more risk, and that carries a small amount of offsetting cost. Both the Competition Commission and Ofwat have proposed to increase significantly the level of onus placed on the companies for cost and delivery performance compared to historical levels in the other sectors they are setting proposals for. But they have not placed as much onus on the companies as is the case in the electricity distribution sector in Great Britain. Lastly, the electricity distribution sector is exposed to significantly more financing risk than the other sectors. Although a 10-year trailing average to the cost of debt would reduce financing risk for a sector that had historically adopted a 10-year financing approach (and had an average maturity of its debts of about 5 years) this is not true of the electricity distribution sector. Northern Powergrid s existing borrowings have an average maturity of 13 years, while the sector s borrowings have an average maturity of 17 years. This exposes the sector to a material risk that the allowed cost of debt will diverge even further from its actual cost of debt. Overall, there are strong reasons to believe that the appropriate cost of equity for our business plan is significantly higher than that being proposed by these other regulators, due to the higher assumed financial gearing in our plan, the longer horizon over which the returns will be applied, and the greater degree of cost, delivery and financing risk to which we are exposed. Moreover, these other regulators have proposed different approaches to the cost of debt that set a higher figure than the index adopted by our plan is likely to allow on average over Nevertheless, we recognise that the proposals of these other regulators for the cost of equity in isolation may limit the acceptability to our stakeholders of an increase in our cost of equity for the period, relative to that in place at present. Northern Powergrid: Our business plan for Page 22 of 49

27 and data on equity beta estimated purely on market evidence can justify a lower figure. Some commentators have highlighted that estimates of the equity beta of UK utilities, based purely on market data, are currently below the assumption in our plan (0.9). Using a lower equity beta assumption would lead to a lower overall cost of equity assumption. The 0.9 figure in our plan is based on a longer term view of the betas seen for UK regulated utilities. The direct market evidence suggests that utility share prices do move closely with the overall stock market at times when there are significant, market wide pieces of news (such as during the stock market declines of 2003, or more recently during 2008 and 2010). It isn t uncommon to see movements which are 1:1. This suggests that investors must perceive themselves to be just as exposed to economy wide risks when investing in utility stocks as they would be in buying an index tracker. But in between, often when markets are less volatile, there are sometimes periods where the link isn t as close, and estimated equity betas are lower. Estimates of beta based on market data can however have various drawbacks which are well documented. For example, a low estimate of beta might in itself indicate that the beta has been under-estimated, and a number of methods have been developed in order to adjust for this. Furthermore, an especially important issue for regulated utilities is the role of regulatory and political risk. There is no reason to expect that this risk would tend to be correlated with movements in the stock market. This means that it will not show up in estimates of equity beta, which simply measures the correlation of equity values with the value of the stock market as a whole. It is just as likely that the realisation of such risk, and a downwards movement in the value of a regulated firm, could happen on a day when stock values in general are increasing. This type of risk needs to be accounted for somewhere in the allowed return, and equity beta is a natural place to do so. There is therefore a strong case that equity beta should be set at levels above those observed from the market data. The assumption on equity beta used in our plan is actually lower than the mid-point of the range the Competition Commission recently estimated for utilities in Great Britain in its provisional findings for NIE. Although it set a slightly higher range for NIE, it found that GB utilities have an asset beta of 0.35 to 0.45, which at 65% gearing converts to an equity beta in the range of 0.8 to 1.1, or a midpoint of The equity beta Ofwat recently proposed for the water sector is however 0.86 (at 65% gearing), which is slightly lower than the level proposed in our business plan. Although an increase in the cost of equity could be justified, we accept that the current level of 6.7% lies within the reasonable range. Overall, we believe that an allowed cost of equity of up to 6.9% could be justified on the basis of the risks faced by the business. But the difference between this and the current regulatory assumption is not large, and the current assumption of 6.7% remains within the reasonable range. We also recognise that, on some measures, comparisons with other network sectors suggest that 6.7% remains a reasonable outcome. In particular, we recognise that the magnitude of our forward expenditure requirements as a proportion of our RAV is smaller than for some of the other network companies whose price controls have recently been settled. We also recognise that some other regulators have recently proposed lower costs of equity than assumed under our plan. And finally, feedback from stakeholders on our financial proposals, and from energy suppliers in particular, has reinforced their view that the cost of equity ought not to rise for electricity distribution companies. We have therefore chosen to hold the cost of equity we factor into our plan at the same level as is currently in place (6.7%, post-tax, real). Northern Powergrid: Our business plan for Page 23 of 49

28 3.4 The level of equity finance Equity finance is an important part of ensuring long-term financial sustainability. The use of equity finance is a core part of Northern Powergrid s strategy for coping with unexpected events. Equity finance is much more flexible than debt finance and, given the uncertainties facing the sector over , maintaining a sustainable equity wedge, and the lower levels of debt (financial gearing) this entails, is an important aspect of being able to ensure we can maintain the good service we plan to offer customers. Our shareholder s preference is to operate with slightly more equity in the financing structure than the 35% that Ofgem currently allows for price-setting purposes The cash flow risks facing the company are increasing. The new regulatory framework is designed to increase the amount of money riding on our performance, meaning more money can be made or lost depending on the cost levels and output performance we achieve. The uncertain timing of uptake of the low-carbon technologies also means there could be significant additional requirements to invest in the network. We have reflected these higher cash flow risks in our consideration of the cost of equity set out above, and concluded that they do put some upward pressure on the cost of equity. Our A-minus credit rating is the best achieved in our sector We also need to consider if there is sufficient equity finance in the overall capital structure. Maintaining the current regulatory assumption that only 35% of the asset base is financed with equity, with the remaining 65% of assets financed through debt, entails a degree of risk. The business would have less headroom to deal with unexpected costs, such as additional costs due to higher than anticipated uptake of low-carbon technology. In order to be financed on a more sustainable basis, and to be on a footing that allows such cost shocks to be absorbed, Northern Powergrid is already operating well below the financial gearing levels assumed by Ofgem during Our current expectation is that we will continue to do the same over the period. If we do, this would allow us to target an A-minus credit rating, and provide the financial flexibility necessary to immediately adapt to a significant upswing in lowcarbon technology uptake (relative to the low levels assumed in our base plan). It would also mean that the possibility of the business experiencing financial distress that might require a bail-out from customers would be even more remote than if the company made use of less equity funding. but we accept that this is our choice and we are not factoring the costs of our preferred conservative financing structure into our proposition for this price control review. Although we think that there are advantages from maintaining a larger equity wedge than the current charges assume, we can also see that there is value in maintaining consistency of the financial parameters over time. This business plan therefore factors in an equity wedge of only 35% of the asset base, with financial gearing of 65%. This approach means that, if we continued to maintain a thicker equity wedge than the regulatory assumption, customers would continue to enjoy, at no extra cost, the risk insulation that lower financial gearing brings. Northern Powergrid: Our business plan for Page 24 of 49

29 The increasing stress on equity in energy and water companies was the subject of an analysis published by the credit rating agency Moody s in May In their paper, Moody s observe that Ofwat and Ofgem may be inclined to change the speed of money to overcome potential cash flow problems caused by the extension of regulatory asset lives. The agency proposes to set aside the beneficial effect of faster money in analysing companies financial strength, which may put more highly geared companies at a disadvantage. In this scenario, an approach such as ours gives added financial strength. To ensure we can sustainably manage our business at these gearing levels, we have extensively stress-tested our financial profile. We find that we can do so, although not without some signs of financial stress as the effect of the lower cost of debt and the lengthening of regulatory depreciation (essentially the period over which long-term assets are paid for by customers) from 20 years to 45 reduces the operating cash flows. This is in spite of the potential financial shocks we could face, and the fact that our actual cost of debt is well above the cost of debt we have factored into these proposals. The headline results for our equity and credit metrics are set out below: Under this plan, our expected return on regulatory equity if our costs and operational performance are in line with our targets would be 6.3%, factoring in the base 6.7% cost of equity, a 0.7 percentage point uplift recognising the rewards we assume may accompany proportionate treatment and a 1.1 percentage-point headwind from our actual cost of debt being well above the one factored into this plan. Ofgem s new framework increases the exposure of equity returns to business performance, which is a direct consequence of its new policies and should deliver good value for customers by encouraging even better performance. At 65% financial gearing, the range of potential returns on equity runs from around 0% to around 13%, which is wider than the range Ofgem expected in 2009 in its review for the period. Our credit metrics maintain an investment grade during under our base plan, provided there are no major shocks. Plausible scenarios for business performance, the uptake of low-carbon technologies and financial market conditions would place significantly more stress on our credit metrics at 65% financial gearing than under our base plan. Overall, these results suggest that the upper limit on sustainable financial gearing is 65%. There is even a case to say that, given the uncertainty the sector faces, the sustainable level of financial gearing is lower than this. Below we set out more detail on the reasons for our view, including why transitional arrangements for asset lives are needed, how the range of returns on regulatory equity for compares with the forecast range for , and why we conclude we can maintain the assumption for gearing we factor into our charges at 65%. A 65% regulatory gearing assumption can be sustained only if it is accompanied by transitional arrangements for the move to longer asset lives. These results all factor in transitional arrangements for the lengthening of asset lives, which reduce the stress on our credit metrics, and help to meet the test of being confident we will be able to 7 Moody s Investor Service: UK Water Sector: Speed of money cannot address potential financeability concerns,16 May Northern Powergrid: Our business plan for Page 25 of 49

30 finance our business. Full details of these arrangements, which are allowed under Ofgem s policy provided they are justified, are set out in section 3.5 below. The range of returns to regulatory equity in the next regulatory period is widening The chart below shows how the potential range of returns to regulatory equity is widening for relative to the range of returns expected in 2009 for the period from 2010 to This chart is centred on a 6.3% expected return on regulatory equity, assuming cost and operational performance in line with the targets in this plan. This reflects the 6.7% base cost of equity assumed by our base plan, a 0.7 percentage point uplift in recognition of rewards we assume may accompany proportionate treatment, and a 1.1 percentage point headwind from the fact that our debt costs will be well above those we have factored into our charges. 16% 14% Cost of debt 12% Tax trigger 10% Health index 8% Connections 6% Losses 4% 2% Customer satisfaction Reliability Costs 0% -2% average company Northeast Yorkshire Expected return Figure 3: Potential returns on regulatory equity for the period If financial gearing is maintained at 65%, the range of plausible returns over the period is wider than was expected when the price control for the period from 2010 to 2015 was set. In fact the lower end of the range would lead to returns on regulatory equity close to 0% or even slightly below. Overall these equity metrics therefore suggest that, to maintain the level of equity risk between the two price control periods, shareholders should be prepared to invest more equity and support a slightly lower level of financial gearing over the period than the 65% level assumed in current network charges. These results use Ofgem s standard methodology for calculating the range of plausible returns on regulatory equity. We have also applied Ofgem s standard assumptions on financing structure, except that we have factored in the effect of our actual cost of debt on our expected return. Northern Powergrid: Our business plan for Page 26 of 49

31 and our credit metrics will not be without signs of strain. As well as testing the range of regulatory equity returns we could plausibly see over the period, we have also stress-tested our credit metrics. Our credit metrics matter because they are an important part of the assessment that ratings agencies undertake to set our credit rating. Our credit rating needs to remain comfortably at investment grade, otherwise the costs of borrowing will rise. This stress-testing has involved evaluating our credit metrics under a range of scenarios. The analysis was supported by Centrus, an independent advisor with significant expertise in debt finance. We asked Centrus to review our June 2013 plans and perform an independent assessment of our financeability under a range of scenarios. These results remain valid as the financial profile of our March 2014 plan remains similar. The scenarios described below all build in a gradual transition to longer asset lives. We have separately tested our credit metrics without these transitional arrangements and concluded that this would not be sustainable over the long term. This is because we judged the impacts on the financial metrics of putting in place no transitional arrangements to be so severe that they would not be in the long-term interests of stakeholders generally. 8 This is in line with our regulator s policies, which allow transitional arrangements provided they are justified. Full details of these arrangements, and the reasons they are necessary, are set out in section 3.5 below. In figures 4 and 5 below we set out the credit metrics that result from our plan, under the simplifying assumptions that Ofgem typically makes regarding financing and pension costs, and using Ofgem s approach to calculating the inputs to the financial ratios. In this scenario, our credit metrics at a 65% gearing and at a 6.7% cost of equity would typically be either A-minus or BBB, although they would show more signs of strain on the Moody s and Fitch ratios. Overall, and given the other factors that ratings agencies take into account, this indicates we could retain an investment grade credit rating. Our calculations for this scenario of the key metrics used by the three major ratings agencies are shown below, with colour coding used to indicate where each ratio lies in relation to the relevant rating agency thresholds. Moody s ratios S&P ratios Fitch ratios Adjusted interest cover Net debt / RAV FFO / Net debt RCF / Capex FFO interest cover FFO / debt Net debt / RAV Adjusted interest cover Net debt / RAV Key: A BBB Strained The move to 45-year asset lives needs to be phased in Figure 4: Our credit metrics under this plan, matching Ofgem assumptions for debt and pension costs 8 See page 6 of the Centrus report, included as part of annex 3.2. Northern Powergrid: Our business plan for Page 27 of 49

32 While these results are presented for our regulated activities taking both of our licensed businesses together, we have also considered these metrics for each of our licensees taken separately, and confirmed that there are no significant differences. The metrics above are simplified because they make the assumption that we are able to finance our debt costs at a different interest rate from the one we actually face on our existing debt. When ratings agencies look at our business, they quite properly factor in our actual cost of debt and full pension costs. Using our actual costs, the financial ratios we would achieve would be less favourable. Moody s ratios S&P ratios Fitch ratios Adjusted interest cover Net debt / RAV FFO / Net debt RCF / Capex FFO interest cover FFO / debt Net debt / RAV Adjusted interest cover Net debt / RAV Key: A BBB Strained Figure 5: Our credit metrics under this plan, based on our actual debt and pension costs Our actual credit metrics will depend on our actual level of gearing and operational performance. But this analysis shows that, under our base plan, at 65% financial gearing, the credit metrics would be consistent with maintaining an investment grade credit rating on the S&P ratios. The position is more ambiguous using the Fitch and Moody s ratios. But these credit metrics could become significantly more stressed under alternative scenarios, particularly if there are high levels of uptake of low-carbon technologies, or if the allowance for cost of debt falls even further, creating an even larger gap between the Ofgem allowance and our actual costs. The potential for additional stress on our credit metrics from such factors is one of the reasons we currently maintain a lower level of financial gearing than our regulator s assumption, and fund a higher proportion of our asset base with equity. Although customers benefit from lower risk because of this, they carry none of the cost. This plan would entail our continuing to set prices for customers based on a 65% financial gearing level, in line with the current assumption. If we maintained a thicker equity wedge and a lower level of financial gearing over the period, customers would continue to enjoy the benefit of the lower risk it brings, at no additional cost. 3.5 An alternative financing structure Our regulator asked us to prepare this plan assuming that it makes no changes to its methodology for setting the allowed cost of equity. It also asked us to say what would need to change in our plan if it decides to change its methodology, and as we were finalising this plan it confirmed its intention to do so. In taking this decision, it set out a view that an efficient company could receive a settlement involving a base cost of equity of 6.4% as part of a package that boosts the return on regulatory equity to well above 6.7%, but that it expects companies that are found to be somewhat less efficient will receive a base cost of equity of 6.0%. Northern Powergrid: Our business plan for Page 28 of 49

33 This section of our plan sets out the necessary changes to our plan in light of our regulator s decision, although much of the evidence used in developing our base plan, set out in the three sections above, is also relevant. We do not believe any changes would be necessary to our plan for allowed gearing. As much as anything else, the approach proposed by the Competition Commission sees the cost of equity flex upwards (and downwards) with gearing, which means the cost of capital is relatively unresponsive to gearing in any case. The key issue is therefore whether the cost of capital (and the calibration of the overall package for highly efficient companies) is being set at the correct level, which we consider in the rest of this section. As a highly cost-efficient company, our plan warrants rewards close to those being allowed for the fast-tracked company. Our resubmitted plan has answered all the questions that Ofgem has raised in relation to the efficiency of our costs. We therefore expect to be found to be similarly cost-efficient to the fast tracked company, Western Power Distribution. Our plan was also found to be stronger than Western Power Distribution s in a number of important respects, so there is merit in ensuring that it receives overall rewards which are proportionate to its status. In light of this, an overall package would be justified that sees our plan rewarded to a degree that approaches the fast-track rewards for Western Power Distribution. We therefore believe that a 6.4% cost of equity would be warranted for our plan even if Ofgem concludes that the cost of equity for a less efficient company should be 6.0%. This would be in addition to the reward associated with proportionate treatment our base plan assumes it will qualify for, which we have assumed will be at a lower level than Western Power Distribution s fast-track reward (2.0% as opposed to the 2.5% awarded to Western Power Distribution). There is also significant evidence that the base cost of equity should in any case be above 6.0%. In developing our base plan, we already took into account the latest evidence from market data, the provisional findings of the Competition Commission in relation to NIE s price control appeal, and the decisions and guidance recently issued by other infrastructure regulators. Our conclusions are included in section 3.3 above. Most importantly, we do not agree with our regulator s assessment that the Competition Commission would set a cost of equity of 5.5% if it were setting the cost of equity for electricity distribution in Great Britain over , at 65% gearing. Firstly, a direct translation of the Competition Commission s statements on its estimates for asset betas for utilities in Great Britain, and the Competition Commission s use of the 60 th percentile in its various ranges, would give a cost of equity of 5.8%. Although it took into account our regulator s estimates for RIIO price controls when assessing the cost of equity to set for NIE, it was very clear in the Bristol Water case (in 2010) that it is much more for other regulators to take into account its views than the other way around. Secondly, there is significant evidence that the Competition Commission would, if it were today opining on the cost of equity for electricity distribution under the RIIO framework for , set a higher cost of equity. Firstly, the price control will apply further into the future than NIE s price control, which would warrant a higher risk free rate. Secondly, there are significantly more sources of risk for companies under the RIIO framework than NIE is being exposed to, which should ultimately bring benefits to consumers but which do come Northern Powergrid: Our business plan for Page 29 of 49

34 at a cost. In section 3.3 we compare the levels of equity risk we are exposed to over with that under the Competition Commission s proposed framework for NIE, and Ofwat s proposed framework for water companies, and conclude that the risks we face are higher. All of these factors point to a cost of equity well above 5.5%, and even well above 5.8%, if electricity distribution s cost of equity were to be assessed under the provisional approach developed by the Competition Commission for NIE. Recognising, however, that the cost of capital for slow-tracked companies is likely to be limited, to some degree, by Western Power Distribution s acceptance of a 6.4% cost of equity, we view this as the realistic upper limit on the figure which could be awarded by Ofgem even for a cost-efficient company. A different methodology on the cost of debt would also be necessary. Putting to one side the reasons that the 6.7% cost of equity assumed in this plan is justified in its own right, the much tougher treatment that Ofgem is proposing in relation to allowances for debt cost already means our business plan proposes a cost of capital at or below the levels being discussed by other regulators. Any reduction in the cost of equity below the 6.7% proposed in our base plan would reduce the overall cost of capital further below the levels that the Competition Commission and Ofwat have proposed, and below levels that are sustainable. In making a move towards the cost of equity methodology provisionally proposed by the Competition Commission in the NIE price control reference, our regulator would therefore also need to adapt its methodology on the cost of debt so it is consistent with the tried and tested practice of the Competition Commission (and several other UK regulators) in setting the allowed cost of debt over many years. Our regulator has stated that its baseline figure for the cost of equity for slow-tracked companies is set at 6.0%, rather than 5.5%, on account of the context created by the existing debt book, and the possibility of actual debt costs diverging from allowances. Since all companies have existing debt books with significantly longer average tenors than the 5 years implied by the 10-year trailing average, it is true to say that the trailing average approach adds to ongoing equity risk across the industry. But the starting point for risk that has already been realised (and the shortfall on existing debt costs) varies significantly across companies, and some have debt portfolios that there is a reasonable prospect the index will cover the cost of (albeit with both upside and downside risk). We agree with our regulator that even these companies carry higher ongoing equity risk than they would if a fixed cost of debt was being set, since the index will be much more responsive to latest market rates than their actual cost of debt, to the degree that a simple fixed allowance would be a more accurate approximation. An uplift on the cost of equity is appropriate in these circumstances and may mean they cost of capital is being set at a high enough level for those companies. But other companies also face this additional risk, and start from a position where the debt cost allowance underfunds the directly observable cost associated with financing their previous investments. Therefore, our regulator s policy on the cost of capital still fails to recognise the cost of embedded debt across the industry, and where this debt burden falls. The debt was incurred to enable us to finance the business with an appropriate level of gearing at rates that were the best that could be obtained in the market at the time. It is therefore important that the approach taken to remunerating these debt costs is compatible with the approach taken to setting the allowed cost of equity. Fundamentally, the overall cost of capital needs to be high enough to cover our actual cost of debt and equity finance, provided our capital structure is sensible and has been constructed efficiently. In adjusting its approach to setting the cost of equity in light of the approach taken by another regulatory body (in this case the Competition Commission), our Northern Powergrid: Our business plan for Page 30 of 49

35 regulator should ensure it is being consistent with all aspects of that regulatory body s methodology in setting the cost of capital. None of the other UK regulators of similar businesses, including the Competition Commission, takes an approach to setting the allowed cost of debt that rules out making allowance for efficiently incurred debt simply because it was issued more than 10 years ago. Consequently, we propose that allowance should be made for the cost of our efficiently incurred debt. There are a number of ways in which this could be implemented. Our preferred approach would see us receive a fixed allowance for our efficiently incurred historical costs, and an indexed allowance for our cost of issuing new debt within the price control period. This alternative would be consistent with the Competition Commission s statement in its provisional findings for NIE s price control appeal, in which it indicated that an indexed approach to debt was only appropriate where a company has been able to take the approach into account in its financing decisions. 9 However, this approach would be a departure from our regulator s RIIO framework which specifies that a trailing average will be used to set allowances for all debt costs. A strong case can also be made for a longer term trailing average, since this would be fully consistent with the decision set out in the RIIO handbook that a trailing average (of an unspecified length) would be used to set debt allowances. A trailing average of around 25 years would closely mirror the average debt structure of the industry, though data availability on Ofgem s preferred index may necessitate a trailing average starting at 17 years in and subsequently increasing in one year increments to reach 25 years in Any indexation approach could also be implemented in conjunction with a sharing mechanism. We acknowledge our regulator s view that an index creates the strongest possible incentives to keep debt costs as low as possible, but this point alone is not a virtue since there is no mechanism for passing realised cost savings on to customers (or sharing the risk with companies to help reduce overall financing costs). Applying a 50% sharing factor on an ongoing basis would be one way of generating some benefits for customers from the incentive properties of indexation. All of these suggestions would help remedy some of the serious deficiencies with the 10-year trailing average approach, which are explained in section 3.2 above. Moreover, since the decision to adopt a 10-year trailing average for the electricity distribution sector was taken only in March 2013, it would be relatively easy to revise this approach given the small amount of time that has passed since then, and bearing in mind that only a small number of financing decisions are likely to have been taken since the finalisation of the policy was announced. Overall, this package would entail an effective cost of capital that is below that being offered to the fast-tracked company. To test whether the overall alternative package we propose above amounts to something which is credible overall, we have assessed the effective cost of capital that would be allowed under our original (June 2013) plan, Western Power Distribution s plan as it was eventually fast tracked, and our resubmitted plan under our alternative financing structure, taking into account three sources of additional potential returns. These sources are: 9 Competition Commission, 2013, Provisional findings for the NIE price control appeal, paragraph Northern Powergrid: Our business plan for Page 31 of 49

36 Fast-track/proportionate rewards: Ofgem allowed Western Power Distribution an additional reward of 2.5% of its forecast costs, on account of the quality and cost efficiency of its overall plan. Converted to equity returns, this amounts to 0.8 percentage points, on average across the four licensees. As set out above, our plan was judged to be higher quality in many areas, and we believe that we have now answered all the questions raised about the efficiency of our costs. We have therefore assumed our regulator will award a proportionate reward of approximately 2.0% of costs, or 0.7 percentage points on our cost of equity. Interruptions incentive targets (IIS): Western Power Distribution reduced its targets for this incentive scheme relative to those described on the face of its published business plan. In doing so, it still set itself targets that involve worse performance than the level it is currently achieving. This means that, even if Western Power Distribution makes no improvements and its performance stands still, it will earn significant rewards under the incentive scheme. Converted to equity returns, this amounts to around 0.6 percentage points of additional return built into its plan. In contrast, we must improve our performance simply to hit target. If we fail to improve on current levels of performance, our equity returns will be reduced by almost 1.0 percentage points. Of course, no benchmarking methodology can be perfect, and our plan adopts Ofgem s targets on the assumption that, in the round, the package will involve a reasonable return. We also aim to improve our performance over time relative to current levels (something we expect Western Power Distribution will also be targeting). But, considering this element in isolation, and relative to our starting point, we face a further material headwind on our equity returns. Net real price effects: Our regulator has highlighted that it believes that Western Power Distribution s assumed costs associated with real price effects are too high, but that it was able to accept them as part of a package that was cost-efficient in the round. This means that our regulator expects that Western Power Distribution will be able to out-perform this element of the package, thus earning further additional returns on account of its cost efficiency elsewhere. We estimate that these returns amount to around 0.8 percentage points on regulatory equity, when evaluated against a benchmark that assumes real price effects are paid for in full by productivity. In contrast, our revised plan assumes that productivity will more than pay for real price effects, which implies a 0.2 percentage point headwind on equity returns if it does not. The table below summarises the estimates developed using this analysis, for the fast-tracked company, our June 2013 plan, and our March 2014 alternative financing proposal. Northern Powergrid: Our business plan for Page 32 of 49

37 Fast-tracked Western Power Distribution plan Northern Powergrid June 2013 plan Northern Powergrid alternative financing structure Cost of debt 2.4% 2.4% 3.3% Gearing 65% 65% 65% Equity returns Base cost of equity 6.4% 6.7% 6.4% Fast track/proportionate rewards 0.8% 0.9% 0.7% IIS targets 0.6% -1.0% -1.0% Real Price Effects 0.8% 0.5% -0.2% Effective allowed cost of capital 4.6% 4.1% 4.2% Table 7: Comparison of effective allowed cost of capital for the fast-tracked company with that proposed by Northern Powergrid Significantly, the figures in this table highlight that the alternative financing package we propose in this plan is close to, but not quite as high as, that which Ofgem has decided is appropriate in the case of the fast-tracked company. While we believe our plan is just as cost-justified as Western Power Distribution s, and has been recognised by Ofgem to be stronger in a number of other respects, we recognise that Ofgem was not convinced of this cost-justification at the first time of asking, which warrants a proportionate differential under the RIIO process. These financial proposals therefore represent exactly what they should: a fair and sustainable balance between customers and investors. We also take some confidence from the fact that the alternative financing structure we have proposed with our March 2014 plan would deliver an effective allowed return that is comparable with that associated with our June 2013 plan, taking into account the difference between the two plans in areas other than financial assumptions. We still believe our June 2013 plan was welljustified, and so it is appropriate that our resubmitted plan would involve a comparable effective allowed cost of capital. Lastly, we note that in the analysis above we have not placed a value on the many aspects of our June 2013 plan which were judged to be stronger than those of other companies by our regulator and other stakeholders. Taking into account these wider aspects of our plan would provide even further justification for our plan to receive proportionate rewards at the level we have assumed, close to (but below) those awarded to the fast-tracked company. 3.6 Capitalisation policies and asset lives We have adopted capitalisation rates and asset lives that accord with our regulator s policy. Electricity distribution assets last many years and so, to take into account the benefits future customers will gain from these assets, the cost is spread over time. Two key financial policies determine when we are able to recover our cost allowances from customers: Northern Powergrid: Our business plan for Page 33 of 49

38 The capitalisation rate: when we spend money, it can either be treated as capital expenditure and paid for over many years, or paid for immediately. The proportion of our expenditure that is treated as capital is the capitalisation rate. Asset lives: once expenditure has been capitalised, it is not paid for immediately. Instead, customers pay it back slowly over time through depreciation charges, and in the meantime pay a cost of capital on the investment we have not yet recovered. The length of time over which the investment depreciates is known as the asset life. We have adopted capitalisation rates and asset lives that we believe are in line with the policy our regulator has set. The proposed capitalisation rate averages 71% over our two licensees, with slightly differing rates for each licensee given their different profiles of expenditure. The asset life of new assets will also gradually extend from 20 years at the start of the period to 45 years by the end. The evidence supporting this capitalisation rate, and the gradual transition to longer asset lives, is set out below. The capitalisation rate in our plan averages 71%... Our business plan assumes a weighted average capitalisation rate across our two licensees of 71% for the period. For the Northeast licensee, the capitalisation rate is 70%, while for the Yorkshire licensee it is 72%. This is in line with our regulator s policy that the capitalisation rate should be broadly equivalent to the average proportion of expenditure that will relate to long-lived assets. The chart below shows how our expenditure breaks down across categories with different expected lifetimes. 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Northeast Yorkshire Northern Powergrid Long-lived assets Short-lived assets Operating costs Figure 6: How our costs are split across different capitalisation categories As can be seen from the chart, there is a proportion of our expenditure that clearly relates to longlived capital assets, and so should be paid for by customers over many years. This includes both the costs of buying and installing network assets, and the costs of designing those network assets. There is also a proportion that is clearly operating expenditure and so should be paid for by customers in Northern Powergrid: Our business plan for Page 34 of 49

39 the year in which it is incurred, as customers realise the full benefits immediately. This includes costs such as running a contact centre to handle communications with customers over issues such as faults. However, in between there is a grey area, where the expenditure will relate to assets that last significantly longer than one year but have an expected economic life of far less than the 45 years for our long-lived distribution assets. This grey area means that, based on a regulatory view of costs included in Ofgem s definition of total expenditure (totex), the capitalisation rate for the Northern Powergrid businesses lies somewhere between 65% and 75%. In order to set the capitalisation rate for the period it is therefore necessary to split these medium-term assets between those that should be expensed immediately and those that should be capitalised and paid for by customers over many years. Our view of the capitalisation rate to assume for the period has therefore been informed by the following ways of estimating an appropriate capitalisation split: Using regulatory cost categories and allocating to capital expenditure all long-lived asset costs, all indirect costs that relate to these assets and those categories of our short-lived capital expenditure that are relatively durable. Expenditure on very short-lived assets such as tools and assets that relate to innovation, primarily operational IT and telecoms, has not been capitalised. Applying statutory accounting rules to our costs to check the split of expenditure over the period between long-lived assets and short-lived assets and operating costs. Using the rules set at the last distribution price-control review (the DPCR5 rules) to calculate the overall proportion of expenditure that is capitalised. The table below shows the capitalisation rates that result from these various approaches, both based on the expenditure we plan in the period and using period expenditure or allowances. In all cases, we have excluded pension deficit repair payments, costs that are passed directly into prices (pass-through costs) and some other specific exceptions. This is to make the comparison of capitalisation rates between the and periods as transparent as possible. It is in line with Ofgem s guidance for calculating totex for these purposes period Regulatory cost category split Statutory split of long-lived assets DPCR5 rules applied to planned expenditure from period Regulatory cost category split Statutory split of long-lived assets DPCR5 rules applied to period allowances Northeast Yorkshire Northern Powergrid average 70% 71% 71% 67% 72% 70% 73% 74% 74% 72% 75% 73% 74% 76% 75% 74% 75% 74% Table 8: Capitalisation rates calculated using a variety of approaches Northern Powergrid: Our business plan for Page 35 of 49

40 As can be seen in the table, each of these approaches gives either a lower or a similar capitalisation rate when applied to our forecast expenditure in the period compared with the period data. We do not, however, believe that it is appropriate for there to be significant changes in capitalisation rates from one regulatory period to the next, since over the long term the capitalisation rate for the business should be broadly stable. For our June 2013 business plan we took a long-term view of the capitalisation rate, by taking an average of the regulatory cost category split across the whole period for each licensee, to calculate the proposed capitalisation rate. This followed Ofgem s guidance on methodology. This long-term view was based on 13 years of expenditure, and so represented our best view of the longterm capitalisation rate for the business. The adjustments made to our plan between June 2013 and March 2014 have not materially changed the findings of this analysis, and so our March 2014 plan retains the same capitalisation rate assumptions. and the move to a 45-year asset life is phased in over eight years. At present, electricity distribution assets are depreciated over 20 years, which means customers pay back the initial investment faster than the assets are expected to last, on average. Ofgem has set a policy that regulatory asset lives should more closely match the expected economic life of the asset in future, so that current and future customers each pay their fair share of the cost of the assets. This means extending the life of new assets to 45 years, while respecting previous commitments to repay previous asset investments over 20 years. These major changes to financial policies will have a significant effect on network revenues and charges in the coming decades. In the near term, charges will be reduced as new assets will be repaid more slowly, but over time the asset base will build up. Many years from now customers will need to pay for a much larger unrecovered investment than would have been the case under the old rules. Our plan phases in these significant changes by moving gradually to the new policy and extending asset lives in eight equal steps from the current 20 years to 45 years in future. These transitional arrangements are justified based on long-term financial modelling that shows the effect of slower cash flows on key financial metrics. As well as improving some of the key metrics over the period and helping to manage the transition, this phased lengthening of asset lives also significantly improves some of the key metrics in later years, beyond By putting these transitional arrangements in place now, financeability is aided in the near future. The transition also serves to offset what would be potentially significant requirements for an increase in the proportion of equity funding after This will help to contain the cost of capital in future periods. We have assessed the impact that moving straight to 45-year asset lives, for new assets, would have on our credit metrics. The results under our base plan are shown below, in terms of how much worse our underlying credit metrics would be if we did not make use of transitional arrangements. They take both of our licensees together, factor in our actual cost of debt, and make an assumption for how the trailing average of the cost of debt index that will set our debt cost allowance might evolve. Northern Powergrid: Our business plan for Page 36 of 49

41 Moody s ratios S&P ratios Fitch ratios Adjusted interest cover Net debt / RAV FFO / Net debt RCF / Capex FFO interest cover FFO / debt Net debt / RAV Adjusted interest cover Net debt / RAV Key: Improved > 5% Improved < 5% Worsened < 5% Worsened >5% Figure 7: The impact that taking no transition would have on our credit metrics Overall, the difference transitional arrangements make to our credit metrics towards the end of the period is material. It becomes even more significant in the period. Given that our credit metrics will be placed under increasing strain from the change to asset lives, it makes sense to put transitional arrangements in place now, so they are not strained even further. Further details of the analysis, and results under other scenarios for the cost of debt, are set out in annex 3.1. We have also confirmed that we would see a similar impact from transition if we considered each of our licensees separately, and if we assumed the notional financing structure our regulator typically assumes when it assesses our credit metrics. Once we consider the potential for cost and performance shocks over the period, the need for transitional arrangements in the near term is more strongly pronounced. They will provide some headroom to deal with these shocks that an immediate move to longer asset lives would remove. Finally, our judgement that transitional arrangements are justified was backed by Centrus, an advisor with significant expertise in debt finance that we asked to perform an assessment of our financeability under a range of scenarios. Centrus assumed that transition was needed in their assessment, 10 as the impact on the financial metrics in ED1 and ED2 of no transition arrangements was sufficiently severe to be regarded as unbeneficial to the long-term interests of stakeholders generally. 11 The results of the Centrus analysis remain valid as the financial profile of our March 2014 plan is similar to our June 2013 plan on which the modelling was undertaken. 3.7 Justifying our finance parameters Although a strong case can be made for more favourable financial parameters, under the base scenario set by our regulator we are proposing a business plan that accepts the Ofgem-preferred debt index, retains the present cost of equity of 6.7% and adopts a regulatory gearing assumption of 65%. In this section we have set out our view of the key financial parameters, namely the cost of debt, the cost of equity, notional gearing, the transition to a 45-year asset life and the capitalisation rate. We have put forward evidence that suggests that: 10 Source: Centrus report, page 6, included as part of annex ED1 is the period; ED2 is the period. Northern Powergrid: Our business plan for Page 37 of 49

42 our actual cost of debt will be higher than the cost of debt indicated by the iboxx index that our regulator has decided to adopt in its Strategy Decision; the cost of equity for electricity distribution companies could justifiably be set higher than for the other energy networks that have been the subject of recent price-control reviews. This would also be a little higher than was assumed at the last price-control review for the electricity distributors because we think there is more risk in the next regulatory period than we are carrying in the current period. But we think the right number still lies within the range of 6.0% 7.2% (post-tax, real) that Ofgem identified in its Strategy Decision; and a gearing level of 65% is at the very top end of the justifiable range for an electricity distribution company in the period. It is likely that equity injection will be needed in the period after 2023, and the assumed gearing will have to fall to levels closer to our own actual gearing. With respect to the transition to longer asset lives, we propose to adopt a one-period transition (i.e. to move to the new 45-year life smoothly over the period). Ofgem allowed the transmission and gas distribution companies a similar transition over the eight years of their price controls. We have provided evidence that justifies the same treatment for us so that we can ensure we maintain reasonable credit metrics. This will enable us to keep our actual borrowing costs down and close the gap a bit between the actual costs that we will incur in servicing our debt and the allowed cost of debt that will be determined by the iboxx index. With respect to the capitalisation rate, this parameter is not one that has its own intrinsic merits or demerits or where we can present market evidence in favour of any particular treatment. Rather it is simply a way to balance the amount of money that we receive over the long term with the amount that we receive over the short term. Ofgem aims to ensure that the capitalisation rate should be fixed over the period and be consistent with the proportion of the expenditures that we make on long-lived assets. We see no reason to depart from this assumption. We conclude that a strong case can be made for a higher cost of debt, a slightly higher cost of equity and lower gearing, leading to a higher weighted average cost of capital than results from the application of our regulator s assumptions for these parameters. Nevertheless, under our base plan we are proposing a business plan that adopts the iboxx debt index proposed by Ofgem, a cost of equity that remains at the 6.7% (post-tax, real) that was assumed at our last price control review and gearing of 65%. These represent a challenging set of parameters, but our view is that they would be financially sustainable if they were accompanied by the benefits we assume may be associated with Ofgem s commitment to treat companies proportionately at the next stage of the price control review. To reflect the potential for such rewards, which it is for Ofgem to set, we have included an explicit 2.0% of totex in calculating prices under this plan. But our regulator plans to change its methodology for setting the allowed cost of equity, and its new approach is not compatible with the cost of debt index. Our regulator also asked us to say what would need to change in our plan if it were to decide to change its methodology for evaluating the cost of equity, something which it recently confirmed it intends to do. Our alternative financing structure takes as a given the reality that Ofgem s recent decision, and Western Power Distribution s fast-tracking, means that the cost of equity for the slow-track companies will be limited to a maximum of 6.4%, even for highly cost-efficient companies (short of significant, new, evidence supporting a higher figure). Since we believe we are a highly efficient Northern Powergrid: Our business plan for Page 38 of 49

43 company, our proposals assume we will receive this figure (in addition to a proportionate reward that is close to, but not as high as, the fast-track reward given to Western Power Distribution). This lower figure for the cost of equity is however not compatible with Ofgem s approach to setting the cost of debt. This is readily confirmed by comparison to other sector regulators and the Competition Commission, none of which rules out the costs of efficiently incurred debt simply because it was issued more than 10 years ago. In moving to this alternative approach to evaluating the cost of equity, our actual debt costs on historically issued debt would need to be covered in full, with further debt issuance being funded based on the indexed level. 3.8 Tax Our business plan factors in the taxes that we will pay in the next regulatory period. Our plans take account of likely tax rates, because Ofgem needs to recognise these in its financial calculations. Northern Powergrid Holdings Company is a major payer of corporation tax, paying around 60m in Other taxes we pay, such as business rates and employers national insurance contributions, took this to over 110m. And including other taxes we collect on behalf of the UK exchequer, the total was around 220m. Our plans assume that we continue to pay tax at prevailing rates. Northern Powergrid and our ultimate parent company, Berkshire Hathaway, have a conservative approach to tax. We are classified as having a low risk profile with HM Revenue and Customs. While we make sure we don t pay extra tax unnecessarily, we do not operate any aggressive corporate tax-planning schemes designed to reduce UK tax payments by exploiting complex tax loopholes. The corporation taxes we currently expect our electricity distribution licensees to pay on their regulated activities over have been factored into this business plan, as set out in the table below Corporation tax 45m 43m 40m 39m 38m 38m 38m 38m Table 9: Corporation tax anticipated by Northern Powergrid under this plan The other taxes we pay, such as the employers national insurance contributions, are also factored into our plans at the prevailing rate. 3.9 Pensions funding Pensions are a part of the benefit we offer to employees and a significant cost to the company because our industry carries historical pension commitments defined at privatisation, which we continue to honour. 12 These figures are not directly comparable with the 2012 actual figures quoted above for Northern Powergrid Holdings Company because they include electricity distribution activity only Northern Powergrid: Our business plan for Page 39 of 49

44 When the company was privatised, it had a defined benefit pension scheme, which was at the time common in the public and private sectors. These schemes are very costly. We took early action to close our scheme to new joiners, and this has helped control these costs (new joiners have instead been offered membership of a scheme that gives our employees benefits that are in line with those offered by any good modern employer). Below we set out: the actions we have taken to limit our pension costs; the ongoing costs that we will incur from the mix of legacy and new schemes we have; and the costs of repaying the sizeable historical pension deficits that have emerged in recent years. We took action to control our pension costs by closing our final salary pension scheme more than a decade before some distribution network companies. We were the first distribution network operator to bring to an end altogether the practice of offering final salary pension schemes to new joiners. We took this step in 1995 in Yorkshire and 1997 in Northeast, and since then have offered only defined contribution pension schemes to new joiners. Although some other network operators closed their original schemes before us, they all subsequently offered their new joiners alternative final salary pension schemes. The table below shows the years in which electricity distribution companies closed their last final salary pension scheme to new joiners. * Some new joiners have since been allowed to join the scheme at the Chief Executive s discretion ** A career average scheme was offered to new joiners until 2008 Source: Ofgem 2009 Defined Benefit Pension Scheme Questionnaires, and publications by individual pension-sponsoring companies. Table 10: The years in which final salary pension schemes stopped being offered to new joiners As the table shows, our move came more than a decade ahead of some distribution network companies. Year last final-salary pension scheme was closed to new joiners Northern Powergrid 1997 SSE 1999* Former E.On Central Networks 2005** Electricity North West 2006 Scottish Power 2006 WPD (South Wales and South West) 2010 UKPN 2011 One network operator (Western Power Distribution, in its South Wales and South West licensees) closed its original, pre-privatisation, final salary scheme to new joiners only from 1 April Northern Powergrid: Our business plan for Page 40 of 49

45 Another network operator, UKPN, closed its final salary scheme, which had replaced its original pre-privatisation scheme, to new joiners in Some electricity distribution network groups, such as Electricity North West, closed their pre-privatisation schemes before us, but they subsequently offered new joiners alternative final salary schemes, and did not bring this practice to an end until many years after we stopped. Looking more widely at network company pension schemes, to compare ourselves with a broader group of peers, we see a similar pattern. The non-electricity distribution defined benefit schemes covered by the Government Actuary s Department s analysis closed to new joiners between 2001 and The earlier that an electricity distributor closed its final salary pension scheme to new joiners, the more thoroughly it has protected its customers from the financial risks associated with the scheme. This means that we took the most significant step we could to limit the costs of pensions to electricity customers, and make these more controllable, well ahead of many of our peer companies. In fact, we were amongst the earliest private sector employers to close their final salary pension scheme to new joiners. But, although some other private sector companies outside our industry have gone on to close their final salary pension schemes to new accruals from existing members, we cannot do this. The legal terms on which the industry was privatised contained long-term commitments to continue providing pensions on the same basis to existing employees, and we have no choice but to respect these commitments. Companies that kept their final salary schemes open for longer will now have a higher proportion of their employees as members of the scheme than would otherwise be the case. Given that the liabilities relating to such schemes have been significantly increasing due to financial market movements in the last few years, customers of these companies will now be facing a larger pension liability than they would otherwise be carrying. Our alternative defined contribution scheme, available to new joiners since 1995, has proved to be a much more cost-effective way of providing pension benefits to employees. It is in line with the benefits that any good modern employer would provide for its staff, and so is a feature of our employment package that we plan to continue offering to new joiners. Our ongoing pension costs are efficient Our defined contribution scheme was opened in 1995 and is available to all employees. It offers benefits comparable with most other large employers schemes, with a company contribution matched by a (usually smaller) employee contribution. Our average employer contribution is 8.5%. The scheme members will receive a pension based on the value of the fund accumulated during their membership. 33% of our employees Our expensive final salary pension scheme was closed in 1997 are members of this scheme. Our defined contribution scheme costs much the same as most modern workplace pensions and is part of our ongoing employee costs, and as such is factored into the costs set out in other parts of this plan. Our defined benefit scheme is significantly more expensive to run than most workplace pensions, but comparable with other schemes in the electricity industry. 66% of our employees are members and, as we explain above, this proportion is lower than it would otherwise have been thanks to our early action to close the scheme to new joiners. Our employer contribution per member is currently set at 26.6% of salary (plus a 2% administration charge). Northern Powergrid: Our business plan for Page 41 of 49

46 and appropriate pension deficit repair costs have been factored into our plan. Pension schemes are created as separate legal entities, overseen by trustees. This way, the assets of the scheme are outside the company that supports that scheme. The scheme receives payments in the form of contributions from employees and their employer and makes payments to pensioners. The trustees of the scheme periodically check that the assets of the scheme are sufficient to meet the likely liabilities that the scheme is committed to, via its membership and rules. This calculation is always carried out by an actuary on behalf of the trustees, and takes place every three years. If this valuation finds that the assets are worth more than the estimate of the liabilities, there is a surplus. If the assets are worth less than the liabilities, there is a deficit. Assets and liabilities will vary significantly over time and an actuarial valuation provides a snapshot position. If there is a deficit, the sponsoring employer of the scheme has a legal obligation to pay in enough money over time to ensure that the deficit is eliminated. The period over which the deficit is eliminated is set out in a deficit recovery plan, which is agreed between the sponsoring employer and the trustees. The Pensions Regulator has powers to intervene if the recovery plan exceeds 10 years, is significantly back-end loaded or uses inappropriate assumptions. At present, there is a significant deficit on pension commitments made before Since we took all reasonable steps to minimise these costs, including closing our defined benefit scheme to new joiners in 1997, our regulator makes allowance for these costs through a separate update mechanism. The level of costs that have been factored into our allowed revenues over the period is shown in the table below, and the explanation is provided in section 1 of the core narrative to our business plan. These costs will be updated every three years, and by may have been replaced by revised, agreed figures. In the table below we also show the costs we are incurring under the existing repair plan we have agreed with the trustees of our pension scheme, which is currently in the process of reappraisal as part of the standard three-year process. Costs included in our revenues m 26m 26m 26m 26m 26m 26m 26m Current costs 35m 35m 35m 35m 35m 35m 35m 35m Table 11: Pre-2010 pension deficit costs currently anticipated under this plan 13 More recent pension deficit costs, which related to pension liabilities incurred from April 2010 onwards, are not covered by this automatic truing-up mechanism. They have therefore been included with our ongoing employment costs, in line with our regulator s expectations. 13 The table shows the pension-deficit cost that will be recovered from customers via allowed revenue, calculated in accordance with our regulator s required formula. Northern Powergrid: Our business plan for Page 42 of 49

47 3.10 Stakeholder engagement Our proposals for financing the business reflect extensive stakeholder input We have undertaken significant stakeholder engagement on all aspects of our business plan, including the financial aspects. We invited stakeholder views on financial issues when we published our Emerging Thinking document in November We invited further comment on our financial plans in April 2013 after we published our updated plan. When we published our updated plan, we also specifically asked for stakeholder views on key decisions on the degree to which we should use our balance sheet to offer lower prices over the period. This touched on key financial topics such as the level of financial gearing, the need for transitional arrangements and the profile of prices over the period Three questions were particularly important: We asked whether stakeholders would favour adopting a higher wedge of equity funding, at 37.5% rather than 35% (and commensurately lower financial gearing at 62.5% rather than 65%), in order to allow us to move immediately to 45-year lives on new assets. This would have given lower bills overall during , although we needed to find a blend that would have been good news for customers and workable for investors. We asked for views on the option we favoured for the profiling of prices, which entailed a profile that was flat in real terms to bring forward reductions that customers would otherwise have had to wait longer for. Although customers had told us clearly that they favoured lower prices, we listened actively to views that could have favoured other options (such as prices that stayed flat in nominal terms). We asked whether stakeholders would prefer it if we finalised our charges for each regulatory year further in advance than is currently the case. If industry arrangements are modified to facilitate our doing this, it could mean energy suppliers would know our distribution charges a year in advance of current timescales. This would provide 15 months notice, rather than the current three months, ahead of the date when the revised prices will take effect. on the possible use of a higher equity wedge to give lower prices today but at a higher cost in the long run In assessing stakeholder views on the first question, we concluded that we could not find a balance that would have been good for customers and workable for investors. Although customers valued lower prices in 2015, this would have come at the cost of a lower financial gearing and higher cost of capital. This higher cost of capital would have meant that, over the long term, customers would have effectively had to pay more. Going without transition and without a lower assumed financial gearing would also not have been acceptable for investors, for the reasons set out in the section on capitalisation and asset lives above. On balance, we therefore put this option to one side, and maintained the current assumed financial gearing of 65%, while also adopting transitional arrangements that will see asset lives for new assets gradually lengthening to 45 years from the current 20. on the profiling of revenues within the price control period In assessing stakeholder views on the second question, we maintained our view that the best option for customers of the network is to adopt a flat real-terms profile for prices over the period. Northern Powergrid: Our business plan for Page 43 of 49

48 This gives lower prices in 2015 but overall leaves customers paying effectively the same amount over the eight years. Although some stakeholders may have preferred a different profile, in weighing up competing views we decided that prices that were flat in real terms offered the best balance between charge reductions in and the price profile thereafter. and on the possibility of giving much longer notice of price changes. The third question was particularly relevant to energy suppliers. They told us that, if they know charges further in advance than is currently the case, it will reduce the risks they face, and mean they can offer lower prices to the end-users they supply. We believe that we could offer a longer notice period without significantly increasing our financing costs, and so offer an overall good deal for customers, provided that the regulatory rules are amended to ensure there is a net reduction of risk in the end-to-end energy supply chain. This plan therefore includes a commitment to continue to work with our regulator and other parties to the regulatory arrangements to explore new arrangements that could facilitate this longer notice period, or reduce the overall costs caused by charging volatility. We have already made significant progress in fulfilling the commitment we originally made in our June 2013, by bringing forward proposals for a change to regulatory rules that would increase the notice period from the current 40 days to 15 months. We are continuing to work on this proposal with Ofgem and the other parties involved The impact on customers bills The net result of our plan is a significant cut in our average charges for domestic customers from April As an electricity distribution company, we provide our society with an essential service in exchange for a fair payment for those services. Everything in this plan therefore reflects the proposition we put to customers explaining what they can expect us to provide and what they will pay for that service. The section of our business plan on outputs sets out what we will deliver. The section on our planned expenditure sets out what we expect it will cost us. And finally, since some of our costs are spread over many years to reflect the long-lived nature of the underlying assets, this section of the plan has set out the cost of sustainably financing our business. Below we set out: the bottom line in terms of what our customers will pay; what is driving the change in prices compared with ; and how our allowed revenues could change over the period if our performance is either better or worse than our targets. As request by our regulator, all the information we present in this plan that relates to the prices we shall charge for our services are expressed in constant prices and assume that there are no rewards or penalties under any of the regulatory incentives and that there are no under- or overrecoveries against our allowed income. Northern Powergrid: Our business plan for Page 44 of 49

49 The chart below shows that electricity distribution accounts for 16% or 85 per year of a typical average household bill of 531, as at January Wholesale energy, supply costs and profit margin Distribution charges Environmental charges Other costs VAT Transmission charges 0% 10% 20% 30% 40% 50% 60% Source: Ofgem factsheet 98 issued 16 January 2013 Figure 8: The share of electricity distribution in the average electricity bill Between now and 2015 distribution charges are set to rise due to the upward profile of those charges that were set for the period. But in April 2015 the net result of the costs in this plan and the financial parameters set out above is a significant cut in our average charges for domestic customers. Figure 9: Distribution charges for average domestic customers under this plan with a reduction of 12%in Northeast and 6% in Yorkshire. If this plan is approved, our overall allowed revenues will fall by 10%. Once that reduction has been allocated to the various customer groups through our approved charging framework, it results in a 12% real-terms price cut for an average domestic customer in our Northeast area in , and a 6% real-terms price cut for an average domestic customer in the Yorkshire area. 14 The overall effect 14 Average charge reductions have been calculated using the standard Ofgem average consumption figures. Northern Powergrid: Our business plan for Page 45 of 49