United Utilities Water Limited – Written evidence (RPI0007)
1.1 United Utilities (UU) is the regulated water and wastewater company serving the North West of England. We provide around seven million people and 200,000 businesses with clean, fresh water every day and treat their wastewater before returning it safely back into the environment.
1.2 UU is the UK’s largest listed water company and is regulated by the Water Services Regulatory Authority (Ofwat), the Drinking Water Inspectorate and the Environment Agency.
2.1 Inflation is a fundamental feature of the economic regulatory model employed in the water sector. Under the current model, revenues and customer bills are adjusted by the annual change in the retail prices index (RPI) and the regulatory capital value (RCV) of each company is also adjusted for movements in RPI. An RPI-stripped real weighted average cost of capital (WACC) is set by the economic regulator, Ofwat, to provide a return to investors on the RCV.
2.2 Recognising legitimate concerns regarding the use of RPI as a measure of general household inflation, with effect from 1 April 2020, Ofwat is changing the basis of indexation in the regulatory model such that:
2.3 It should be noted that investors do not get excessive returns because of the use of RPI (over CPIH), as Ofwat will firstly determine a nominal WACC from which it then derives its RPI-stripped WACC and CPIH-stripped WACC, using Ofwat’s assessment of long-term RPI and CPIH.
2.4 The transition in the use of CPIH for the indexation of the RCV recognises that it is likely to take some considerable time for companies in the sector to adjust existing RPI-linked hedging, particularly absent the development of a CPI or CPIH-linked government bond market.
3. United Utilities’ inflation exposures
3.1 UU has exposure to inflation in the following areas.
3.2 Indexation of the RCV (c£11.2bn) and revenues (c£1.7bn per annum) as described in section 2 above.
3.3 Just over half of UU’s total net debt is in index-linked form (c£3.7bn).The vast majority of this is RPI-linked debt (c£3.5bn) split between bonds issued in the wholesale market (c£2bn), loans from the European Investment Bank (c£1.25bn) and the remainder in other bank loans. In addition, UU has issued a small amount of CPI-linked bonds (£165m as at 1 July 2018).
3.4 UU has issued RPI-linked debt over many years to provide effective long-term management of our balance sheet in line with responsible risk management policies, with RPI-linked debt providing an effective inflation hedge in relation to the RCV. Some of our RPI-linked debt has a maturity date extending out until 2057.
3.5 UU also has inflation exposure arising from its defined benefits pension schemes (c£4bn). Whilst these schemes closed to new entrants in 2006, they remain active for employees who joined prior to then and pay out benefits to pensioners.
3.6 The schemes’ rules provide that pensioner benefits are indexed to RPI. As part of a long-term de-risking strategy, the schemes are progressively implementing a full market hedge of RPI-inflation using index-linked government bonds (gilts) and swaps.
4. Response to request for evidence
4.1 The Committee is seeking evidence to address the following questions. As a business with intrinsic linkages to inflation, and a financing and hedging strategy designed to mitigate inflation exposures, UU welcomes the opportunity to provide its response primarily from a financial markets user’s perspective.
Question 1: The current situation regarding the retail price index is untenable. Do you agree? If so, what would you recommend is done to improve this situation?
4.3 Response to Question 1:The current situation is widely recognised and understood, particularly by the users of financial markets.
4.4 Whilst the deficiencies in the construct of RPI have been well publicised, there has been little study as to whether CPI better reflects general household inflation. We recognise that CPIH has been developed to be a more ‘inclusive’ measure of household inflation but this measure is relatively new and untested. We also note the ONS programme to develop a Household Cost Index but such developments take time, particularly in establishing trust and confidence in any new measure.
4.5 Therefore, notwithstanding the known shortcomings with RPI we do not necessarily agree that the current situation is untenable, and we consider it would be difficult to address issues regarding RPI without also addressing what alternative measure of inflation best reflects household inflation.
Question 2: Should the retail price index be abolished? If so, how should that be achieved?
4.7 Response to Question 2: No on the grounds that the abolition of RPI could lead to widespread financial disruption, contract uncertainty and an arbitrary transfer of economic value, which could have far-reaching implications for the UK economy.
4.8 There are many pension obligations, commercial contracts and financial instruments that reference RPI and do not necessarily contain ‘fallback’ provisions.
4.9 If the Government wants to encourage the progressive withdrawal in the use of RPI, then there are a number of steps it can take such as: phasing out the use of RPI in duties and tariffs; passing legislation to enable pension schemes to choose the measure of inflation by which benefits are indexed; determine which measure of inflation should be used by the Bank of England for inflation targeting purposes (i.e. continue with CPI or move to CPIH); and start to build a CPI or CPIH-linked government bond market with a gradual transition away from RPI-linked issuance.
4.10 We consider that from a market users perspective there is unlikely to be a sufficiently developed CPI(H)-linked gilt market until around £40bn (c10% of current RPI-linked gilts outstanding) has been issued in CPI(H)-linked form, across a range of maturities from 5 to 30-years (e.g. £10bn in each of 5-year; 10-year; 20-year and 30-year maturities).
4.11 We know from the consultation that the UK Debt Management Office (DMO) undertook in 2011, that the Government had concerns about a splintering of the index-linked gilt market, particularly around market liquidity considerations, so one would assume that the DMO would be tasked with undertaking a further study.
Question 3: If not, how should the retail price index be changed? If so, how should that be achieved?
4.13 Response to Question 3: Whilst we do not believe that the current situation is untenable, we do think that a programme of more active incremental improvements to the RPI might be feasible.
4.14 Whilst we believe that the decision taken in January 2013 to maintain the continuity of the RPI’s long-time series without major change was well founded (given legitimate concerns regarding the impact of making a fundamental change on users of the RPI), a long-term path for continuous and incremental improvement could be feasible without necessarily triggering a ‘fundamental and detrimental’ review by the Bank of England.
4.15 The way goods and services are consumed has changed significantly in recent years and is likely to evolve further. Developments in data collection and processing are also advancing rapidly. Over time, this may mean that more effective ‘real time’ price collection methods could be deployed, which could eliminate the need to rely on aggregation methods to the extent currently used.
4.16 Consideration of new data collection techniques etc. would have to be carefully worked through, and the UKSA/ONS would need to embark on a programme of feasibility work.
4.17 We consider that it might be possible to incrementally improve the RPI over time by making changes that in themselves do not constitute a fundamental change but overall improve the index. To illustrate, the change in the collection of clothing prices in 2010 was not in itself considered to be a fundamental change.
4.18 This is probably a more acceptable position versus the current ‘low maintenance’ position adopted by the ONS (particularly given that RPI-linked gilts extend out until 2068), but we would emphasise that the reasons for avoiding any ‘fundamental and detrimental’ change remain well founded.
Question 4: What would the implications be of changing or abolishing the retail price index?
4.20 Response to Question 4: We believe that there could be far-reaching implications in making a fundamental change to or abolishing the RPI. In particular, we are concerned that the abolition of RPI could lead to widespread financial disruption, contract uncertainty and an arbitrary transfer of economic value.
4.21 Many of these considerations were taken into account at the time of the 2012/13 UKSA consultation into the impact of the ‘formula effect’ on RPI versus CPI, and we believe that the decision taken then not to make any fundamental changes to RPI was well founded.
4.22 Given that the construct of RPI is very well understood by market participants and the decision taken in 2013 not to make fundamental changes provided certainty for market users going forward, there is a material risk that if a fundamental change were made to RPI or it were abolished, then this could have a significant impact on existing contractual arrangements that reference RPI – creating commercial imbalances and winners and losers on a purely arbitrary basis.
4.23 We also consider that such a move would likely render existing market hedges and asset-liability management strategies defunct, with the potential for ensuing chaos in the financial markets as lawyers seek to pick through the various ‘fundamental change/cessation’ provisions that exist in what are disparate, non-uniform, corporate bonds, index-linked gilts and swap contracts.
4.24 We also note that some commentators have observed that abolishing RPI would mean considerable savings for the UK Government on index-linked gilts. We would caution that such a move could be a rather short-sighted ‘win’, as it would at a stroke inflict material damage on the economic interests of the existing index-linked gilt holders, which are the same investors that the DMO relies on to buy new government debt.
19 July 2018