UBS Investment Research Cheung Kong Infrastructure/Power Assets

Transcription

1 ab UBS Investment Research Cheung Kong Infrastructure/Power Assets Global Equity Research Hong Kong Utilities Sector Economics Regulatory review risk is only modest We expect manageable risks on the upcoming overseas regulatory reviews We expect the risks on the upcoming regulatory reviews in the UK and Australia for Cheung Kong Infrastructure (CKI) and Power Assets (PAH) regulated assets to be manageable. We also do not expect significant regulatory changes in 213 or 218 in Hong Kong. As the companies have diversified portfolios, the effect of any one regulatory reset should be relatively modest. Thus, we do not expect any major share price performance overhang. Lower costs and growth are to be expected Although we think the allowed WACC will fall because of a lower allowed cost of debt in the UK and Australia, this could be mitigated by a lower actual cost of debt in theory. We also expect slower asset growth in Australia, which should be positive because of reduced upward pressure on tariffs and shareholders need to retain too much equity in the business. Maintain Buy rating on CKI and Neutral rating on PAH We maintain our positive stance on CKI as the company has dependable cash flow and dividend growth. We also think it is still actively looking for acquisition opportunities to capture the deleveraging trend of corporates, investment funds and governments. We have already assumed a lower return for the overseas businesses during the next regulatory periods in our estimates. 31 July Stephen Oldfield Analyst stephen.oldfield@ubs.com Yuxiao Peng, CFA Analyst yuxiao.peng@ubs.com Ken Liu Associate Analyst ken.liu@ubs.com Valuation: Price targets of HK$61. for CKI and HK$73. for PAH We base our price targets on DCF, with explicit cash flow to 218E, 3% terminal growth for dividends received from overseas associates, and an 8.% terminal ROIC on the scheme of control business. We assume a 5.6% WACC for PAH and 5.7% for CKI. Table 1: Valuation summary Reuters Share Price Price PE P/BV ROE Dividend yield code Rating price target Upside 213E 214E 213E 214E 213E 214E 213E 214E CKI 138.HK Buy HK$53.2 HK$61. 15% 12.8x 12.8x 1.7x 1.6x 15.4% 14.2% 3.4% 3.5% Power Assets 6.HK Neutral HK$69.45 HK$73. 5% 14.4x 14.2x 2.2x 2.x 15.7% 14.8% 3.7% 3.7% Note: Above data as of 3 July 213. Source: Bloomberg, Company data, UBS estimates This report has been prepared by UBS Securities Asia Limited ANALYST CERTIFICATION AND REQUIRED DISCLOSURES BEGIN ON PAGE 24. UBS does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

2 Investment summary The most significant non-hong Kong regulatory assets for CKI and PAH are in the UK and Australia. This report reviews the regulatory issues for the companies operations in both these markets as well as Hong Kong. There are two main issues to consider for the UK and Australian markets. (1) What levels of regulated returns will the regulator set for the next regulatory period; and (2) what changes will be made regarding expenditure allowances that drive growth in regulatory asset values. For the Hong Kong market, the key areas are: 1) what will be the long-term regulatory return on average net fixed assets post-218 (we think it will drop to about 8%); and 2) will the regulatory and industry structure change significantly after 218 (we think the answer is no). We remain positive on CKI given the following. We do not think regulatory uncertainty will be a major overhang for the share price. We believe the deleveraging trend by corporates, investment funds and governments, particularly in Europe, will likely produce further acquisition opportunities. The company offers sustainable and growing dividends. For companies owning a single or concentrated portfolio of regulatory assets, their shares tend to underperform during the lead up to a regulatory review because of uncertainty on the outcome of the review. However, as CKI and PAH have diversified portfolios, we believe the effect of any one regulatory reset will be relatively modest. Table 2 shows the key dates of the regulatory reviews for CKI s UK and Australian businesses. Overall, we think there is some risk that allowed returns will fall in upcoming resets because of lower costs of capital. In theory, the effect of a lower allowed cost of debt should be mitigated by a lower actual cost of debt, so the risk would be mostly on the cost of equity. We think the regulatory trend is for more stringent reviews/oversight of expenditure. We also expect a lower rate of growth in regulated asset bases particularly in Australia which we believe is a positive thing; lower asset growth should mean reduced upward tariff pressure (and hence less political pressure) and less need for shareholders to retain quite so much equity in the business (perhaps eventually increasing the dividend distributions). UBS 2

3 Table 2: Key dates for different regulatory reviews in CKI and PAH s main markets Event Date Northumbrian Water Company business plan submissions Draft determination Final determination Start of the new regulatory period Jan-14 May-14 Dec-14 Apr-15 UK Power Networks Company business plan submissions Draft determination Final determination Start of the new regulatory period Mar-14 Jul-14 Nov-14 Apr-15 Victoria Power Networks Final framework and approach paper released by the AER Company business plan submissions Draft determination Start of the new regulatory period Final determination Sep-14 Apr-15 Oct-15 Jan-16 Apr-16 SA Power Networks Final framework and approach paper released by the AER Company business plan submissions Draft determination Start of the new regulatory period Final determination Mar-14 Oct-14 Apr-15 Jul-15 Oct-15 Source: AER, Ofgem, Ofwat Valuation CKI s shares are still trading at 12.8x 12-month forward PE, close to its historical average. We think the current valuation has not factored in any acquisition potential as the shares were trading at 13.6x 12-month forward PE in June 21, a few weeks before CKI announced its acquisition of UK Power Networks. Our estimates have already incorporated a lower return for its overseas businesses (particularly in the UK and Australia) during the upcoming regulatory reviews. Over the last few years, the defensive earnings and cash flow profile of the constituent businesses in CKI and PAH s portfolios, and their reliable and growing dividend streams had led to share price outperformance. However, with signs of economic stabilisation (if not recovery) in key global markets, we think demand for defensive high-yield stocks could wane. In particular, the dividend yields at only about three and a half percent are no longer very compelling. We still have a Buy rating on CKI as acquisition opportunities in both regulatory assets and other infrastructure businesses should continue as we think deleveraging remains a priority for a number of existing owners of these assets in the developed markets, particularly in Europe. UBS 3

4 In our view, PAH is likely to participate in acquisitions that CKI makes if those acquisitions are in the power or gas utility industries. Our preference is for CKI with more upside to our DCF-based price target. Chart 1: PAH 12-month forward PE 25x 2x 15x 1x 5x x Average: 11.5x Jul-79 Oct-81 Jan-84 Apr-86 Jul-88 Oct-9 Feb-93 May-95 Aug-97 Nov-99 Mar-2 Jun-4 Sep-6 Dec-8 Mar-11 Jul-13 Chart 2: PAH 12-month forward dividend yield 1.% 9.% 8.% 7.% 6.% 5.% 4.% 3.% 2.% 1.%.% Avg: 5.8% +1 SD: 7.% -1 SD: 4.6% Jan-8 Jun-82 Nov-84 Mar-87 Aug-89 Jan-92 May-94 Oct-96 Mar-99 Jul-1 Dec-3 May-6 Sep-8 Feb-11 Jul-13 Source: Bloomberg, Company data, UBS estimates Source: Bloomberg, Company data, UBS estimates Chart 3: CKI 12-month forward PE 25x 2x 15x 1x 5x x Average: 12.2x Aug-96 Oct-97 Nov-98 Jan- Feb-1 Apr-2 May-3 Jul-4 Aug-5 Oct-6 Nov-7 Jan-9 Feb-1 Apr-11 May-12 Jul-13 Chart 4: CKI 12-month forward dividend yield 7.% 6.% 5.% 4.% 3.% 2.% 1.%.% average: 4.% +1 SD: 5.% -1 SD: 2.9% Jan-96 Dec-96 Dec-97 Nov-98 Oct-99 Sep- Aug-1 Jul-2 Jun-3 May-4 Apr-5 Mar-6 Feb-7 Jan-8 Dec-8 Nov-9 Oct-1 Sep-11 Aug-12 Jul-13 Source: Bloomberg, Company data, UBS estimates UBS versus consensus Source: Bloomberg, Company data, UBS estimates Our earnings and dividend forecasts for PAH and CKI are roughly in line with consensus. Our earnings forecasts for CKI are 3-4% lower, which we think is because consensus may have included the around HK$3m non-cash exchange gain from the Japanese yen borrowings. UBS 4

5 Table 3: Earnings and dividend forecasts UBS versus consensus Net profit (HK$m) DPS (HK$) Price UBS 213E 214E 215E 213E 214E 215E Target Rating Consensus Rating CKI UBS 1,122 1,148 1, Buy Consensus 1,39 1,6 1, Variance -3% -4% -3% -2% -2% -3% 8% Buy Hold Sell % 2% 4% 6% 8% 1% Power Assets UBS 1,267 1,445 1, Neutral Consensus 1,228 1,522 1, Variance % -1% -1% -2% -5% -2% 3% Buy Hold Sell % 2% 4% 6% 8% 1% Source: Bloomberg, Company data, UBS estimates Upside/downside scenario analysis CKI Upside scenario: valuation of HK$72 per share. In our upside scenario, we assume PAH s scheme of control business will earn an ROIC of 9% post 218, and the earnings decline for the UK and Australian businesses during the upcoming regulatory reviews would be 1ppt less than we currently assume. Downside scenario: valuation of HK$52 per share. In our downside scenario, we assume PAH s scheme of control business will earn an ROIC of 7% post 218, and the earnings decline for the UK and Australian businesses during the upcoming regulatory reviews would be 1ppt more than we currently assume. PAH Upside scenario: valuation of HK$84 per share. In our upside scenario, we assume PAH s scheme of control business will earn an ROIC of 9% post 218, and the earnings decline for the UK and Australian businesses during the upcoming regulatory reviews would be 1ppt less than we currently assume. Downside scenario: valuation of HK$63 per share. In our downside scenario, we assume PAH s scheme of control business will earn an ROIC of 7% post 218, and the earnings decline for the UK and Australian businesses during the upcoming regulatory reviews would be 1ppt more than we currently assume. UBS 5

6 United Kingdom CKI has four major regulated utility businesses in the UK. UK Power Networks (UKPN), Northern Gas Networks (NGN) and Wales and West Utilities (WWU), which are regulated by the Office of Gas and Electricity Markets (Ofgem). Northumbrian Water (NWC), which is regulated by the Office of Water (Ofwat). The UK portfolio contributes around 75% to CKI s group earnings (including the 39% stake in PAH) and 45% to PAH s group earnings. Major regulatory changes concerning the next regulatory period include the decision on allowed returns, the move to a total expenditure (totex) approach instead of segregating capital expenditure (capex) and operating expenditure (opex) for the UK, and other regulatory changes for the UK water businesses. We are neutral on the overall regulatory changes: Although the allowed return on regulated assets value (RAV) would likely drop given a lower cost of debt, we think value to the equity holders is likely to be unaffected if the companies refinance at a lower cost of debt to match with the regulators assumptions. We think the totex approach for the UK regulated water, electricity and gas distribution businesses should allow companies to exert greater control over cash flows than at present, and although not quantifiable, this should help reduce risk. We also think regulatory changes in the UK water businesses are evolutionary rather than revolutionary, and harbour relatively little additional earnings risk. On an industry-wide basis, we believe it is not regulatory changes that would create the greatest earnings risk during the next regulatory period, but rather the allowed return on RAV (i.e. the area in which regulatory changes are relatively minor). WACC is likely to fall with lower cost of debt We think the next regulatory reviews for the UK electricity distribution and water companies are likely to result in a lower allowed real return (WACC) as the cost of debt has come down significantly over the last few years. The decision on gas distribution companies was made for the April 213-March 221 period and announced on 17 December 212, with the post-tax real WACC decreased from 4.4% to 3.9% primarily because of a lower cost of debt assumption. We also estimate the post-tax real WACC for electricity distribution companies will drop from 4.1% to 3.6%, while UBS s UK utilities analysts expect the WACC for water companies to decrease from 4.5% to 4.%. We think the assumption on the cost of equity and notional gearing have relatively less downside risk and are likely to remain largely unchanged during the coming regulatory period. UBS 6

7 Table 4: Estimated WACC (real) for UK electricity distribution companies 21-15E E Cost of equity 6.7% 6.7% Cost of debt (gross of tax shield) 3.6% 2.4% Leverage 65.% 65.% Vanilla WACC 4.7% 3.9% Fully post-tax WACC 4.1% 3.6% Source: Ofgem, UBS estimates Table 5: Estimated WACC (real) for UK water companies 21-15E 215-2E Cost of equity 7.1% 6.9% Cost of debt (gross of tax shield) 3.6% 2.85% Leverage 57.5% 65% Vanilla WACC 5.1% 4.3% Fully post-tax WACC 4.5% 4.% Source: Ofwat, UBS estimates A lower cost of debt is likely The regulators benchmark their cost of debt assumptions to the iboxx GBP nonfinancials bonds yields with broad A and BBB credit ratings, and the UK water bond yields. These have fallen below the regulators assumptions in their final decisions announced in December 29 (Chart 5 and Chart 6). The Office of Gas and Electricity Markets (Ofgem) assumed a real cost of debt of 3.6% for electricity distribution companies between 21 and 215, but the bond yields in real terms (which Ofgem benchmark to) are only 1.1%. The Office of Water (Ofwat) assumed a real cost of debt of 3.6% for water companies between 21 and 215 as well, but the UK water bond yields in real terms are currently at around 1%. Chart 5: Ofgem assumptions of cost of debt (electricity distribution companies) versus market real cost of debt Chart 6: Ofwat assumptions of cost of debt versus market real cost of debt 1.% 1% 8.% 8% 6.% 6% 4.% 2.% 2.4% 4% 2%.% % Jan-98 Aug-99 Apr-1 Dec-2 Jul-4 Mar-6 Nov-7 Jul-9 Feb-11 Oct-12 Nominal cost of debt Real cost of debt Trailing av erage Ofgem real cost of debt assumption Jun-14-2% Apr-1 Apr-3 Apr-5 Apr-7 Apr-9 Apr-11 IBox x UK Water 15+y r Nominal Yield Implied Real Cost of Debt Ofw at AMP6 Allow ed Real Cost of Debt Source: Ofgem, Bank of England, Bloomberg, UBS estimates Source: Ofwat, Bloomberg, UBS estimates For gas distribution companies, the cost of debt has been lowered from 3.6% (for the April 28 to March 213 period) to 2.9% (for the April 213 to March 214 period) by indexing to the 1-year trailing average of the iboxx GBP non-financials indices of 1+ years maturity with broad A and broad BBB credit ratings. For electricity distribution companies, we think the real cost of debt is likely to be reduced from 3.6% (for the April 21 to March 215 period) to 2.4% (for the April 213 to March 214 period). This is because according to the strategy decisions announcement published by Ofgem, both electricity and gas distribution companies will apply the same cost of debt indexation UBS 7

8 methodology to the same benchmark. If we were to assume the real bond yield for both indices remain unchanged until March 215, the 1-year trailing average bond yields would be 2.4% (Chart 5). During the next regulatory period for both electricity and gas distribution companies, the cost of debt will be adjusted annually to reflect the 1-year trailing average of iboxx indices for bond yields with broad A and BBB credit ratings. Although this implies that the allowed return may not be consistent throughout the regulatory period, we think this will not bring significant volatility on WACC because the 1-year trailing average bond yields are unlikely to move significantly. However, there may be a trailing effect between the assumed cost of debt by the regulator and the actual cost of debt when interest rates start to increase, the companies borrowing costs will rise, but changes to the cost of debt assumption would be delayed to the next year or two because of the trailing effect. For water companies, UBS s UK utilities analysts believe their previous assumption of a 2.2% real cost of debt for the next regulatory period was too bearish after Ofwat published its final regulatory methodology document for on 25 July 213. Ofwat admitted that the current low cost of debt is unsustainable and therefore UBS s UK utilities research team revised its real cost of debt assumption to 2.85%. However, this is still lower than the 3.6% assumption for the current regulatory period. We think the companies could somewhat mitigate the impact of a lower allowed return by reducing their cost of debt. In this scenario, we do not expect the returns for equity holders to have a material impact. Risk on lower cost of equity is modest While we think the reduction in cost of debt is likely, we see less risk on the cost of equity for the next regulatory period. Over the last few years, reductions in the risk-free rate have been offset by an increase in the implied market equity risk premium, leaving the market cost of equity at least as high as the current regulators assumption for both the water and electricity distribution companies (Chart 7 and Chart 8). UBS 8

9 Chart 7: The market equity risk premium has more than offset the fall in the risk-free rate Chart 8: leaving the market cost of equity at least as high as the current regulators assumption (if not higher) 15.% 15.% 12.5% 1.% 7.5% 12.5% 1.% 5.% 2.5%.% 7.5% 5.% 7.4% 7.25% -2.5% Apr-5 Apr-6 Apr-7 Apr-8 1y r Real Gilt Yields Apr-9 Apr-1 Apr-11 Apr-12 Implied Market Equity Risk Premium Apr-5 Apr-6 Apr-7 Apr-8 Apr-9 Apr-1 Apr-11 Apr-12 Implied Market Cost of Equity Ofw at market Cost of Equity Ofgem market cost of equity (electricity distribution) Source: Bloomberg, UBS estimates Source: Bloomberg, Ofwat, UBS estimates Historically, UK regulators and the Competition Commission have looked through short-term volatility and distortions in the risk-free rate (Chart 9) when determining their market cost of equity (Chart 1). Chart 9: UK regulators and Competition Commission have ignored short-term distortions when determining both the risk-free rate 5% 4% 3% 2% 1% % -1% -2% Ofgem, 2.8% Ofgem, 2.5% CC, 2.% Ofw at, 2.9% CC, 2.5% Ofcom, 2.% Ofcom, 1.4% Ofgem, 2.8% Postcom, 2.5% Ofgem, 2.% Ofcom, 2.1% Ofgem, 2.5% ORR, 2.% Ofw at, 2.% Ofgem, 2.% Ofgem, 2.% CC, 2.% CAA, 1.8% Jan- Jan-1 Jan-2 Jan-3 Jan-4 Jan-5 Jan-6 Jan-7 Jan-8 Jan-9 Jan-1 Jan-11 Jan-12 Jan-13 1y r Index -Linked Gilt Yields Source: Oxera, Ofwat, Ofgem, Bloomberg, UBS estimates Chart 1:...and the market cost of equity 16% 14% 12% 1% 8% 6% 4% 2% % Ofcom, 6.6% Postcom, 6.9% Ofcom, 7.% Ofgem, 7.% Ofgem, 7.3% Ofw at, 7.4% Ofgem, 7.3% ORR, 6.8% CAA, 7.1% Ofgem, 7.3% CC, 6.8% CC, 6.6% Ofgem, 7.3% Ofcom, 6.4% Ofgem, 7.3% CC, 7.% Apr-5 Apr-6 Apr-7 Apr-8 Apr-9 Apr-1 Apr-11 Apr-12 Implied Market Cost of Equity Source: Oxera, Ofwat, Ofgem, UBS estimates UBS 9

10 For gas distribution companies, the decision announced in December 212 set the cost of equity at 6.7%, with an equity risk premium of 5.25%, a riskfree rate 2.% and equity beta of.9. Table 6 shows Ofgem and Ofwat s precedents on the cost of equity. For electricity distribution companies, we estimate the cost of equity for the next regulatory period will remain unchanged at 6.7%. We believe the decision is unlikely to differ significantly from that of the gas distribution companies. The discussion on cost of equity in the strategy decisions for the electricity distribution companies published in March 213 (the document that summarises the responses from companies and the decisions made by Ofgem concerning the coming regulatory change) is very similar to the strategy decisions for the gas distribution companies published in March 211. In brief, the strategy decisions conclude that Ofgem would keep the initial range of the cost of equity (post-tax real) between 6.% and 7.2%, which is consistent with the strategy decisions for gas distributions. The initial ranges for the components of the real cost of equity including the risk-free rate and the market risk premium are also the same as that for gas distribution. Risk-free rate: Ofgem s proposal for the initial range of the risk free rate is 1.7% 2.%. Compare to the risk free rate assumption of 2.% in the current regulatory period, we think the initial range indicates that Ofgem could see through the short term distortion in the risk free rate and is unlikely to make significant changes. Equity risk premium: Ofgem thinks this is likely to be at around the upper end of % according to the DMS (Dimon, Marsh and Staunton) methodology. (Ofgem assumed 5.25% for the current regulatory period). Equity beta: In the final decision (213-21) for gas distribution companies, Ofgem assessed that the cash flow risk faced by gas distribution companies is similar or slightly lower than that for the electricity distribution companies for the current regulatory period (21-15). Unless Ofgem has a strong case to argue that the risk of cash flow would be lower for electricity distribution companies during the next regulatory period (215-23) than the gas distribution companies, we do not expect any material changes in the equity beta. For water companies, UBS s UK utilities analysts expect the real cost of equity to decrease slightly from 7.1% to 6.9% during the next regulatory period, based on the most recent update from the Ofwat regulatory methodology published on 25 July 213. Based on its initial analysis, Ofwat believes that market evidence appears to be consistent with a lower cost of equity than it has assumed for the current regulatory period (21-15). UBS 1

11 Table 6: Regulatory precedents on cost of equity from Ofgem and Ofwat Date announced Risk free rate Equity risk premium Equity beta Cost of equity Ofgem Gas distribution Dec % 5.25%.9 6.7% Gas transmission Dec % 5.25% % Electricity transmission Dec % 5.25%.95 7.% Electricity distribution Dec 29 2.% 5.25%.9 6.7% Gas distribution Dec % 4.75% % Electricity and gas transmission Dec % 4.5% 1. 7.% Ofwat Water Nov 29 2.% 5.4% % Source: Ofgem, Ofwat Notional gearing less problematic for electricity than water For gas distribution companies, Ofgem has decided to raise the notional gearing assumption slightly from 62.5% to 65% for the next regulatory period (213-21). The 65% gearing assumption is the same as that for the electricity distribution companies during the current regulatory period (21-15). For electricity distribution companies, Ofgem has made it clear in the strategy decision document that based on the information it has to date, it does not see any compelling reasons for a reduction in notional gearing levels from those used during the previous regulatory period. For water companies, Ofwat has indicated a possible increase in the notional leverage assumption from the current 57.5% to 6-7%. We think concerns on capital structure remain as Ofwat had previously expressed concerns over private equity capital structures delivering excess equity returns. Ofwat s Chairman Mr Jonson Cox highlighted the increasing levels of leverage within the sector in a recent lecture. Mr Cox has (in passing) stated the case for moving away from a single industry cost of capital to one that reflects companies actual leverage. UBS s UK utilities analysts think that should Ofwat move to use the companies actual leverage in the WACC assumption, it would likely only be concerned with leverage at the regulated entity level, as this is the business that provides the service to customers and could potentially need to be rescued by taxpayers. Gearing for the regulated business of Northumbrian Water was 63% as of March 212, but the annual report mentioned this was mainly because of a special dividend paid after the acquisition by CKI in October 211. Average notional gearing for the regulated business of Northumbrian Water was 59.5% for the last two years, which is near the 57.5% notional leverage assumed by Ofwat. We estimate its gearing is likely to drop to 59% when approaching 215, should the company s debt remain relatively stable and its regulated assets grow organically according to the capex plan. We think this UBS 11

12 represents little threat to Northumbrian Water even if the regulator marks the notional gearing assumption to reflect each individual company s situation. Totex more control for companies; should reduce some cash flow risk As part of the totex (total expenditure, i.e. capital expenditure plus operating expenditure) approach, companies will be able to request what portion of their expenditure is opex (or fast money ) and is compensated for immediately through allowed revenue, and what portion is capex (or slow money ) to go into the regulated capital value (RCV) to earn a return going forward. While companies will have to justify their chosen opex/capex split, it will also give them far greater control over cash flow than at present, and although not quantifiable, this should help reduce risk. Other key regulatory changes in the UK water industry The UK water industry is going through a period of regulatory change beyond the level associated with typical regulatory price reviews. The most significant of these planned changes for the next regulatory period (from April 215 to March 22) include the following. Average cost-to-serve: Remuneration for household retail expenditure is to be made in reference to the UK water industry s average costs. Business retail competition: Non-household retail will be opened up to competition in an attempt to improve customer service levels. Despite these changes, the regulator decided the wholesale majority of the business will continue to attract the familiar real return on RCV (with revenues to be driven by the current RPI+K formula), with the most contentious of the changes that is, those affecting retail (average cost-to-serve, and nonhousehold competition) only affecting a relatively minor portion of the businesses (around 11% of industry revenue). Ofwat has committed to allocating 1% of the water companies RCV to the wholesale function (Chart 11). Chart 11: Ofwat breakdown of industry revenue and AMP6 RCV allocation 1% 9% 8% 7% 6% 5% 4% 3% 2% 1% % Rev enue Business Retail - Competition, 2% Household Retail - Ave Cost to Serve, 9% Wholesale - Regulated Return on RCV, 89% Wholesale - 1% RCV allocation Source: Ofwat, UBS estimates UBS 12

13 Average cost-to-serve: winners and losers The most contentious of the regulatory changes is with respect to the calculation of the industry average cost-to-serve, which will form the basis of household retail remuneration during the next regulatory period (Chart 12). Under this concept, the water companies retail costs will be reimbursed with respect to the industry average, i.e. if the per-household retail cost is higher than the industry average, this additional cost will not be compensated for, and if the cost is below the average, then firms may be reimbursed based on their actual costs, leaving them with no initial financial advantage, but allowing for outperformance should they further lower their costs. Chart 12: Indicative household retail cost-to-serve shows both Pennon and United Utilities currently above the industry average Cost to serve ( /unique customer) B&W Hants Portsmouth Bristol Cambridge Sutton & E Surrey Severn Trent South Staffs Wessex Affinity Dee Valley South East Yorkshire Thames Anglian Northumbrian Southern Dwr Cymru South West (Pennon) United Utilities WoC WaSC Listed Companies Average Note: WoC = water-only company (supplies only water to customers); WaSC = water and sewage company Source: Ofwat, Company data, Moody s, UBS estimates We think the original formula that Ofwat deploys to calculate the average costto-serve has unfairly penalised the combined water and sewage companies (including Northumbrian Water) and unfairly benefits the water-only companies, as the extra cost of sewage services increases the size of household bills, and thus, the /household cost of bad debt expense. In response to this, Ofwat has revised its calculation methodology (as published on 25 July 213) and made an allowance for additional costs faced by water and sewage companies. Chart 13 shows our fairest case scenario, by indexing the companies bad debt costs to the size of the customers bills, which accounts for both the provision of sewage costs and other variations. In the two scenarios (Charts 12 and 13), the average cost-to-serve for Northumbrian Water is still very close to the industry average, which poses little risk to its earnings. UBS 13

14 Chart 13: UBS fairest case scenario average cost-to-serve indexing bad debt costs to the size of the household bill Cost to serve - Bad debts indexed to bill size ( /unique customer) B&W Hants Portsmouth Bristol Cambridge Wessex Severn Trent Sutton & E Surrey Thames Anglian Yorkshire Northumbrian Affinity South East Dee Valley South Staffs Southern South West (Pennon) Dwr Cymru United Utilities WoC WaSC Listed Companies Av erage Source: Ofwat, Company data, Moody s, UBS estimates Business retail competition little expected impact; gains as likely as losses Although Ofwat has yet to give an indication of what margin these retail businesses will be allowed to earn, UBS s UK utilities analysts believe that based on other retail companies/functions and previous deregulations, and considering the asset-light nature of the operations, an allowed EBIT margin equivalent to around 5% of contestable revenue would be reasonable. We estimate that Northumbrian Water earns around 13m from its business retail customers. A 5% margin would give an allowed EBIT of around 6.5m, or 2% of the group s total EBIT in 212 under UK GAAP. Should 5% of these business customers change retailers every year, this would likely leave only a small proportion of earnings (around.1%) at risk in any particular year. Moreover, on average, it would seem just as likely for the water companies to gain retail business customers as lose them. UBS 14

15 Australia Power prices are driving regulatory pressure The regulatory regime in Australia has been coming under increased scrutiny in the last few years because of a significant increase in power prices, with rising network costs being the main contributor (Chart 14 and Chart 15). Chart 14: NSW real electricity prices a 55-year view Chart 15: NSW default tariffs $MWh Real 21 $ $M 3X F Retail Environment Energy Network total Source: Simshauser, UBS estimates Source: IPART, UBS estimates A combination of rapid regulated assets base (RAB) growth and limited (or even negative) demand growth have driven network prices higher. This is because the regulated utilities earn a regulated return on assets. Any asset growth or weak volume would require a higher network price to meet the regulated returns. UBS s Australian utilities research team believes the weak power demand growth can be attributed in part to the rapid increase on power prices. Chart 16 and Chart 17 show throughput for Victoria Power Networks (VPN) and ETSA (also called South Australian Power Network, or SAPN), which demonstrated that volume per customer was weak and has been decreasing in most cases over the past six years. Chart 16: VPN throughput (December year-end) Chart 17: SAPN throughput (December year-end) 1.% % 14.4.% -1.% %.% % -2.% -3.% -4.% mwh/cust (y/y) mwh/customer (RHS) % -3.% -4.% mwh/cust (y/y) mwh/customer (RHS) Source: Spark Infrastructure Company Fact Book Source: Spark Infrastructure Company Fact Book Another reason for higher electricity prices is the rapid growth in RAB (Chart 18 and Chart 19). The increases in RAB have been partly justified because of UBS 15

16 the need to replace aging equipment, rising peak demand, and higher finance costs (at least expectations of this at the time of the last regulatory review). However, the Australian Energy Regulator (AER) is of the view that weaknesses in the regulatory framework have also contributed to higher prices, and it has made changes for the upcoming regulatory period in order to address the concerns. Chart 18: Victoria s electricity RAB and volume growth index 25 2 Chart 19: South Australia s electricity RAB and volume growth index Volume index Rab index RAB Index Volume Index 215 Source: Company data, NEM, UBS estimates Source: Company data, NEM, UBS estimates The next five-year regulatory period will commence on 1 July 215 for SA Power Networks and 1 January 216 for both VPN companies. UBS s Australian utilities research team believes that regulators in an attempt to contain tariff growth may force down returns for regulated businesses. However it seems that this risk may be reduced by the reining in of RAB growth, which would reduce upward tariff pressure. Our discussions with VPN management suggest that lower RAB growth is likely in the next regulatory period. In November 212, the Australian Energy Market Commission (AEMC) published its opinion on some rule changes proposed by the AER. The new rules will likely apply firstly to the ETSA review for the period commencing 1 July 215. In particular, they concluded the following. Rate of return: A rate of return framework will be applied for electricity transmission, electricity distribution and gas distribution, but the new rule is somewhat vague. Cost of debt: The AEMC views the AER as being in the best position to determine the best methodology to determine a return on debt. Capex: The AEMC confirmed that the network service provider is the starting point for determining network capex and opex. The AER must accept any reasonable proposal (it cannot simply reject and replace it with its own). However, the AER can apply a number of tools such as expenditure sharing schemes and efficiency reviews. Benchmarking: The AER will be required to publish benchmarking reports. UBS 16

17 Enhanced discretion in determining WACC One key change planned by the AER is to move away from a strict, formulaic approach of considering the individual components of WACC to also being able to consider the overall total cost of capital. The AER said some stakeholders viewed the existing method of determining the WACC as being too mechanical. The AER mentioned that apart from the traditional CAPM model that derives the WACC from a bottom-up approach by estimating the cost of equity and cost of debt, it would also consider other top-down approaches such as adjusting the outcome from WACC based on broker reports, trading multiples, and intangibility tests or estimates from other regulators. Details on the discussions were set out in AER Rate of Return Guidelines, a consultation paper published by AER in December 212. A particular area of focus for the next regulatory period is likely to be the assessed cost of debt. Given that government bond yields in Australia are at a historical low, the cost of debt for the next regulatory period would likely drop. Another possible change could be the way the regulator estimates the return on debt. Historically, the AER would set the return of debt using an on the day approach where it estimates the cost of debt on the first day of the new regulatory period and then lock in the cost of debt throughout the whole regulatory period. The AER has stated that it is considering an alternative approach by indexing the cost of debt to a trailing average. For example, a trailing average return on debt could be calculated as an average of debt raised over the last five or 1 years, using one of several possible weighting approaches. Expenditure Management of VPN, which comprises CitiPower and PowerCor, expects growth in the regulatory asset base to slow during the next regulatory period as a lot of the higher capex for upgrading networks has been completed. Expenditure on advanced metering infrastructure (AMI) such as smart meters should be largely complete by the end of this year, and this expenditure, which is presently accumulated in a separate account, should be combined into the RAB for the next period. Like in the UK, the regulator is refining its expenditure assessment framework, making a more holistic assessment of the required expenditure. According to the AER, the processes for standardisation and analysis of expenditure are still under development and consultation on changes is already underway. According to VPN management, the guidelines should be completed by 29 November 213. The AER will use a range of tools including benchmarking for assessment of efficient expenditure needs. The AER plans to develop new rules to provide stronger incentives for efficient investment. The incentives will include a capex sharing scheme, where the benefits of any under-spend will be shared with consumers and the costs of overruns will also be shared. In addition, the AER is planning ex-post reviews, which could lead to the exclusion of inefficient capex from the RAB, commencing in 214. UBS 17

18 Hong Kong 213 SOC review no big changes expected PAH and CLP are holding discussions this year regarding the development plan, which will drive capex plans for the remainder of the scheme of control (SOC) duration, and on changes to the SOC terms, if any. We expect discussions to be concluded by the end of 213. We do not expect any significant changes to the SOC agreements for the period. Any changes would need the agreement of the power companies and cannot be unilaterally imposed on the companies by the Hong Kong government. Growth over the next few years should be more modest because of lower capex in the generation sector as the emission reduction upgrades have been completed. We expect new gas units of 312.5MW each to enter service in 217 and 218. The level of capex will likely depend on the construction of new gas-fired power plants. The Hong Kong government is still examining options regarding plans to reduce carbon omissions by 22, which would result in amendments to the development plan. We believe companies will be able to earn the permitted returns and recover their fuel costs despite the ongoing tariff pressure. The tariff pressure is a direct consequence of the government s legislation to tighten emissions and its requirement to buy gas from China. Even if we were to assume that the companies cannot earn the permitted returns, the downside to their share prices has been limited in the past. Table 7: Historical, forecast tariffs and scheme of control accounts balances for Hongkong Electric (Power Assets) Tariff (HK$ cents /kwh) E 214E 215E 216E 217E 218E Basic tariff Fuel Clause Rate reduction and special rebates Net tariff Growth YoY -.1% 2.9% 6.3% 2.9% 1.4% 2.5% 3.1% 3.9% 3.5% Fuel cost (HK$ cents /kwh) Unit fuel cost Growth YoY 11.2% 21.6% 1.1% 1.3% 1.8% 2.7% -.2% 1.6% 19.2% Scheme of control accounts (HK$ m) Tariff stabilisation fund account (543) (497) (425) (351) (34) (273) (227) (27) (25) Fuel clause account 569 1, (59) (288) (616) (52) 657 Rate reduction account (4) (1) (2) (14) (24) (33) (4) (46) (53) Note: For scheme of control accounts, positive account balance means it is a receivable; negative account balance means it is a payable. Note: The tariff numbers for 213 are actual, while fuel cost and scheme of control accounts numbers are our estimates. Source: Company data, UBS estimates Longer-term outlook: lower returns post 218E Despite political noise about the Hong Kong power market, we think the SOC which is a contractual undertaking between the government and the power companies will remain intact. In particular, we believe the following. UBS 18

19 While complaints by populist politicians and the local media may become more vocal, the regulatory regime will likely remain intact. The broader implications of any unilateral breach of the SOC contracts by the Hong Kong SAR Government by not allowing tariff increases or a fuel cost recovery would more than offset any benefits of marginal reductions in near-term power prices. A material change in the Hong Kong regulatory restructure particularly the opening up of the Hong Kong power market to new entrants is extremely unlikely in the foreseeable future. However, a reduction in the permitted return if not in 218, then at least by 223 is probable. This is already reflected in our discounted cash flow estimates. However, the extent of any reduction would likely depend on the inflation and interest rate environment at that time. We assume a lower rate of return after 218 of 8% (compared to 9.99% now). We do not expect the SOC to be replaced by competitive power markets with new entrants. Superficially, a move towards competition sounds alluring as one could assume that competition would result in efficiency gains and lower returns. However, there are several reasons why we think competitive markets are not a viable option in Hong Kong. These relate to the following: The long-term contractual commitments with respect to gas supplies and nuclear power purchases with particular stranded cost risks for gas contracts in a competitive environment would be recoverable under the SOC contracts in place. The establishment and operating costs of a competitive power market. The limited interconnections between CLP and Hongkong Electric (HKE) and the absence of interconnections between HKE and Guangdong. The potential for unfavourable tariff consequences to "grass roots" customers if larger customers are cherry picked by new entrants. The relative system reliability levels of Guangdong and Hong Kong's two power networks. The likely contradiction between power purchases from Guangdong (which would likely be subcritical coal-fired) and Hong Kong's environmental policies. UBS 19

20 Power Assets Income statement (HK$m) 12/8 12/9 12/1 12/11 12/12 12/13E % ch 12/14E % ch 12/15E % ch Revenues 12,463 1,221 1,312 1,246 1,486 1, , , Operating expenses (ex depn) (3,339) (2,892) (2,85) (3,567) (3,424) (3,362) -1.8 (3,557) 5.8 (2,996) EBITDA (UBS) 9,184 7,386 7,553 6,886 7,134 7, , , Depreciation (1,384) (1,724) (1,92) (1,947) (2,3) (2,1) 3.4 (2,172) 3.4 (2,245) 3.4 Operating income (EBIT, UBS) 7,8 5,662 5,651 4,939 5,14 5, , , Other income & associates 732 1,53 1,899 4,193 4,665 5, , , Net interest Abnormal items (pre-tax) Profit before tax 9,29 7,616 8,131 9,933 1,564 11, , , Tax (1,31) (919) (937) (858) (835) (849) 1.7 (852).4 (1,5) 17.8 Profit after tax 7,719 6,697 7,194 9,75 9,729 1, , ,57.6 Abnormal items (post-tax) Minorities / pref dividends Net income (local GAAP) 8,29 6,697 7,194 9,75 9,729 1, , ,57.6 Net Income (UBS) 7,718 6,695 7,189 9,74 9,729 1, , ,57.6 Tax rate (%) Pre-abnormal tax rate (%) Per share (HK$) 12/8 12/9 12/1 12/11 12/12 12/13E % ch 12/14E % ch 12/15E % ch EPS (local GAAP) EPS (UBS) Net DPS Cash EPS BVPS Balance sheet (HK$m) 12/8 12/9 12/1 12/11 12/12 12/13E % ch 12/14E % ch 12/15E % ch Net tangible fixed assets 46,488 47,464 47,924 48,799 49,298 49, , , Net intangible fixed assets Net working capital (incl. other assets) 3,281 1,193 1,475 (56) (19) (632) (1,48) 65.7 (1,454) 38.8 Other liabilities (9,85) (8,722) (8,928) (9,562) (9,919) (1,113) 2. (1,332) 2.2 (1,567) 2.3 Operating invested capital 4,684 39,935 4,471 39,181 39,27 38, , , Investments 1,175 19,363 35,6 36,197 42,224 45, , , Total capital employed 5,859 59,298 76,71 75,378 81,494 83, , , Shareholders' equity 47,327 52,144 56,137 57,873 63,35 68, , , Minority interests Total equity 47,327 52,144 56,137 57,873 63,35 68, , , Net debt / (cash) 3,532 7,154 19,934 17,55 18,459 15, , , Other debt-deemed items Total capital employed 5,859 59,298 76,71 75,378 81,494 83, , , Cash flow (HK$m) 12/8 12/9 12/1 12/11 12/12 12/13E % ch 12/14E % ch 12/15E % ch Operating income (EBIT, UBS) 7,8 5,662 5,651 4,939 5,14 5, , , Depreciation 1,384 1,724 1,92 1,947 2,3 2, , , Net change in working capital (735) (515) (82) Other (operating) ,72 2,27 2, , , Operating cash flow (pre tax/interest) 9,16 8,353 8,396 9,443 9,322 1, , ,14 9. Net interest received / (paid) Dividends paid (4,375) (4,53) (4,53) (4,53) (4,951) (5,292) 6.89 (5,5) 3.92 (5,615) 2.9 Tax paid (1,182) (728) (868) (542) (566) (676) 19.4 (696) 3. (72).8 Capital expenditure (1,864) (2,464) (2,284) (2,533) (2,675) (2,137) -2.1 (3,587) 67.9 (5,185) 44.5 Net (acquisitions) / disposals Other (1,897) (3,58) (13,619) 442 (1,583) Share issues Cash flow (inc)/dec in net debt 499 (2,59) (12,577) 3, , , FX / non cash items 1,386 (1,32) (23) (711) (1,327) Balance sheet (inc)/dec in net debt 1,885 (3,622) (12,78) 2,429 (954) 2,789-1, Core EBITDA 9,184 7,386 7,553 6,886 7,134 7, , , Maintenance capital expenditure Maintenance net working capital (1,384) (1,724) (1,92) (1,947) (2,3) (2,1) (2,172) (2,245) Operating free cash flow, pre-tax 7,8 5,662 5,651 4,939 5,14 5, , , Source: Company accounts, UBS estimates. (UBS) valuations are stated before goodwill, exceptionals and other special items. Note: For some companies, the data represents an extract of the full company accounts. UBS 2

21 Global Equity Research Hong Kong Electric Utilities 12-month rating Neutral Power Assets 12m price target Company profile HK$73. Power Assets is an integrated electric utility servicing Hong Kong Island and the nearby Lamma Island. The business is regulated by a scheme of control agreement with the Hong Kong government, which permits a pre-determined return on average net fixed assets. Faced with a mature, low-growth business, the company is diversifying overseas, with a focus on regulated utility assets, typically in cooperation with 39% shareholder Cheung Kong Infrastructure. It has investments in Australia, Canada, New Zealand, Thailand and the UK. Valuation (x) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E P/E (local GAAP) P/E (UBS) P/CEPS Net dividend yield (%) P/BV EV/revenue (core) NM NM NM EV/EBITDA (core) EV/EBIT (core) EV/OpFCF (core) EV/op. invested capital Enterprise value (HK$m) 12/11 12/12 12/13E 12/14E 12/15E Average market cap 119, , , , ,331 + minority interests + average net debt (cash) 18,72 17,982 17,64 17,64 14,137 + pension obligations and other - non-core asset value (36,197) (42,224) (45,2) (47,8) (49,817) Core enterprise value 12,74 14,691 12, , ,651 Value (EV/OpFCF & P/E) 25.x 2.x 15.x 1.x 5.x 15.x 1.x 5.x Growth (%) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E Revenue EBITDA (UBS) EBIT (UBS) EPS (UBS) Cash EPS Net DPS BVPS x.x 12/11 12/12 12/13E 12/14E 12/15E EV/OpFCF (LHS) P/E (RHS) Profitability 55.% 15.% 53.71% 14.5% 52.43% 14.% 51.14% 13.5% 49.86% 13.% 48.57% 12.5% 47.29% 12.% 46.% 12/11 12/12 12/13(E) 12/14(E) 12/15(E) 11.5% EBIT margin (LHS) ROIC (RHS) ROE v Price to book value 16.5% 2.2x 16.% 2.2x 15.5% 2.1x 15.% 2.1x 14.5% 2.x 14.% 2.x 13.5% 1.9x 13.% 12/11 12/12 12/13(E) 12/14(E) 12/15(E) 1.9x ROE (LHS) Price to book v alue (RHS) Growth (UBS EPS) % % 11 2.% % % 12 5.% 1 12/11 12/12 12/13(E) 12/14(E) 12/15(E).% Rev enue (LHS) UBS EPS Grow th (RHS) Margins (%) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E EBITDA / revenue EBIT / revenue Net profit (UBS) / revenue NM NM NM NM NM NM Return on capital (%) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E EBIT ROIC (UBS) ROIC post tax Net ROE Coverage ratios (x) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E EBIT / net interest Dividend cover (UBS EPS) Div. payout ratio (%, UBS EPS) Net debt / EBITDA Efficiency ratios (x) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E Revenue / op. invested capital Revenue / fixed assets Revenue / net working capital NM NM NM NM Investment ratios (x) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E OpFCF / EBIT Capex / revenue (%) NM NM Capex / depreciation Capital structure (%) 5Yr Avg 12/11 12/12 12/13E 12/14E 12/15E Net debt / total equity Net debt / (net debt + equity) Net debt (core) / EV Source: Company accounts, UBS estimates. (UBS) valuations are stated before goodwill, exceptionals and other special items. Valuations: based on an average share price that year, (E): based on a share price of HK$69.5 on 29 Jul :36 HKT Market cap(e) may include forecast share issues/buybacks. Stephen Oldfield Analyst stephen.oldfield@ubs.com Yuxiao Peng, CFA Analyst yuxiao.peng@ubs.com Ken Liu Associate Analyst ken.liu@ubs.com UBS 21