DONG Energy A/S (DENERG.CO)

Transcription

1 Europe/Denmark Equity Research Electric Utilities Rating OUTPERFORM Price (29 Sep 16, Dkr) Target price (Dkr) Market Cap (Dkr m) 114,932.2 Enterprise value (Dkr m) 155,521.6 *Stock ratings are relative to the coverage universe in each analyst's or each team's respective sector. ¹Target price is for 12 months. [V] = Stock Considered Volatile (see Disclosure Appendix) Research Analysts Share price performance Mark Freshney mark.freshney@credit-suisse.com Vincent Gilles vincent.gilles@credit-suisse.com Specialist Sales: Jason Turner jason.turner@credit-suisse.com Ju l A u g Se p O c t DENERG.CO OM XC 2 0 The price relative chart measures performance against the OMXC 20 which closed at on 29/09/16 On 29/09/16 the spot exchange rate was Dkr7.45/Eu 1.- Eu.89/US$1 Performance 1M 3M 12M Absolute (%) Relative (%) DONG Energy A/S (DENERG.CO) INITIATION Turning CfDs into value Initiate with an Outperform, TP DKK 310: Since IPO in June, DONG is up c17% vs. the sector, but we still see underappreciated value in offshore wind (c77% of EV) and, in particular, the UK Contracts for Difference (CfDs). We are c0-4% above EBITDA consensus across E due to our view on gains on wind farm sales. Investment story: (1) Turning high-returning CfDs into value: DONG is the most-experienced offshore wind builder and has three offshore wind CfDs with the UK government which generate c15% post-tax nominal project-level returns (vs. c6.3% WACC). These are the highest risk-adjusted returns in the sector; (2) Returns and cash flow generation to 2020: DONG's policy is to farm-down (i.e. divest) c50% stakes in offshore wind assets to partners. This generates up-front cash flows through profits on disposal and construction gains to help fund DONG's own retained stakes. Infrastructure assets such as UK offshore wind are scarce and we expect bidders to still be prepared to pay high prices; and (3) Energy price exposure more limited: A major sensitivity is in the c34mmboe p.a. of gas and oil production (c13% of EV). Past value-destruction is in the share price, and the decision to run-down the business is cash flow enhancing, in our view. Each +/- c5% fall in gas and oil prices impacts group EPS by +/-c2.5%. Catalysts and risks: DONG aims to farm-down 50% of Race Bank in H2 16. We expect a large price (c2x build cost) to underpin the value of UK wind. Returns in competitive tenders have gone to low levels in recent months (e.g. c6-8%), hence we do not value post-2020 projects. The key risk is that DONG does not remain disciplined and wins assets on sub-wacc returns. Valuation: We reach our TP on the basis of an asset-by-asset SOTP. We value the UK assets in the same way as SSE, and this gets us to c11.6x EV/EBITDA for the UK wind assets (c62% of EV). Ex. farm-downs, DONG group trades on 8.8x FY+1 EV/EBITDA. We think it should trade on c9.7x on the same metric. Peers trade on c10x. CS HOLT arrives at DKK 301/share. Financial and valuation metrics Year 12/15A 12/16E 12/17E 12/18E Revenue (Dkr m) 70, , , ,916.6 EBITDA (Dkr m) 18, , , ,044.7 Adjusted net income (Dkr m) -12, , , , CS EPS (adj.) (Dkr) Prev. EPS (Dkr) ROIC (%) P/E (adj.) (x) P/E rel. (%) EV/EBITDA (x) Dividend (12/16E, Dkr) 6.00 Net debt/equity (12/16E,%) Dividend yield (12/16E,%) 2.2 Net debt (12/16E, Dkr m) -19,241.4 BV/share (12/16E, Dkr) 93.2 IC (12/16E, Dkr m) 26,237.8 Free float (%) 19.9 EV/IC (12/16E, (x) 5.9 Source: Company data, Thomson Reuters, Credit Suisse estimates DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, LEGAL ENTITY DISCLOSURE AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse 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 DONG Energy A/S (DENERG.CO) Price (29 Sep 2016): Dkr273.4; Rating: OUTPERFORM; Target Price: Dkr310.00; Analyst: Mark Freshney Income statement (Dkr m) 12/15A 12/16E 12/17E 12/18E Revenue 70,850 65,007 65,259 73,917 EBITDA 18,484 23,099 22,066 24,045 Depr. & amort. (8,701) (7,210) (7,668) (7,773) EBIT (7,250) 16,639 14,398 16,271 Net interest exp. (1,450) (1,594) (1,550) (1,664) Associates (8) (10) (10) (10) PBT (9,367) 15,034 12,838 14,598 Income taxes (2,717) (3,841) (3,484) (3,770) Profit after tax (12,084) 11,193 9,354 10,828 Minorities (31) (75) (75) (75) Preferred dividends (714) (652) (652) (652) Associates & other Net profit (12,829) 10,466 8,627 10,101 Other NPAT adjustments Reported net income (12,829) 10,466 8,627 10,101 Cash flow (Dkr m) 12/15A 12/16E 12/17E 12/18E EBIT (7,250) 16,639 14,398 16,271 Net interest (659) (944) (900) (1,014) Cash taxes paid (5,091) (4,152) (4,929) (4,895) Change in working capital 1, Other cash and non-cash items 25,132 5,390 5,152 7,273 Cash flow from operations 13,450 16,932 13,721 17,636 CAPEX (18,739) (18,776) (20,564) (26,098) Free cashflow to the firm (5,289) (1,844) (6,843) (8,462) Acquisitions Divestments 2,605 5,726 2,888 1,725 Other investment/(outflows) 16,948 16,952 13,741 17,656 Cash flow from investments 814 3,902 (3,936) (6,718) Net share issue/(repurchase) Dividends paid 0 0 (2,506) (2,775) Issuance (retirement) of debt Cashflow from financing (1,495) (702) (3,208) (3,477) Changes in net cash/debt 5,227 (3,200) 7,144 10,194 Net debt at start (17,214) (22,441) (19,241) (26,386) Change in net debt (5,227) 3,200 (7,144) (10,194) Net debt at end (22,441) (19,241) (26,386) (36,580) Balance sheet (Dkr m) 12/15A 12/16E 12/17E 12/18E Assets Total current assets 14,323 9,371 7,371 6,760 Total assets 98, , , ,878 Liabilities Total current liabilities 10,457 9,953 11,234 9,920 Total liabilities 60,393 56,462 60,295 69,513 Total equity and liabilities 98, , , ,878 Per share 12/15A 12/16E 12/17E 12/18E No. of shares (wtd avg.) (mn) CS EPS (adj.) (Dkr) (30.71) Prev. EPS (Dkr) Dividend (Dkr) Free cash flow per share (Dkr) (12.66) (4.40) (16.29) (20.14) Key ratios and valuation 12/15A 12/16E 12/17E 12/18E Growth/Margin (%) Sales growth (%) 2.8 (8.2) EBIT growth (%) (516.0) (13.5) 13.0 Net income growth (%) (115.6) (17.6) 17.1 EPS growth (%) (106.7) (17.8) 17.1 EBITDA margin (%) EBIT margin (%) (10.2) Pretax profit margin (%) (13.2) Net income margin (%) (18.1) Valuation 12/15A 12/16E 12/17E 12/18E EV/Sales (x) EV/EBITDA (x) EV/EBIT (x) (21.7) Dividend yield (%) P/E (x) (8.9) Credit ratios (%) 12/15A 12/16E 12/17E 12/18E Net debt/equity (%) (58.3) (42.3) (53.2) (64.9) Net debt to EBITDA (x) (1.2) (0.8) (1.2) (1.5) Interest coverage ratio (x) (5.0) Source: FTI, Company data, Thomson Reuters, Credit Suisse Securities (EUROPE) LTD. Estimates Company Background DONG Energy is a developer, builder, owner of offshore wind farms, mainly in the UK but also globally. There is a Danish-UK-Norwegian oil and gas production segment and Danish regulated distribution and sales businesses, as well as thermal generation. Blue/Grey Sky Scenario Our Blue Sky Scenario (Dkr) Brent crude oil at US$80/bbl. TTF natural gas at 20/MWh. Recurring 700MW p.a. of projects across with c20% valueaccretion / premium on farm-down. No LNG losses Our Grey Sky Scenario (Dkr) Brent crude oil at US$30/bbl. TTF natural gas at 10/MWh. No further cost-cutting in the upstream gas and oil business. Only c20% value-accretion on remaining UK projects. No accretion on any further products post Nordpool power falls to 18/MWh Share price performance Jul- 16 Aug- 16 Sep- 16 Oct- 16 DENERG.CO OMXC 20 The price relative chart measures performance against the OMXC 20 which closed at on 29/09/16 On 29/09/16 the spot exchange rate was Dkr7.45/Eu 1.- Eu.89/US$1 DONG Energy A/S (DENERG.CO) 2

3 DONG in six key charts Figure 1: DONG shares are up c17% (vs. flat markets) since IPO at a price of DKK Jun-16 Jun-16 Jun-16 Jun-16 Jul-16 Jul-16 Jul-16 Jul-16 Aug-16 Aug-16 Aug-16 Aug-16 Sep-16 Sep-16 Sep-16 Sep-16 Figure 2: Difference between day-ahead reference price and the 150/MWh CfD strike price, /MWh Value of top-up Apr-12 Oct-12 Apr-13 Oct-13 Apr-14 Oct-14 Apr-15 Oct-15 Apr-16 Oct-16 Apr-17 Oct-17 Apr-18 Oct-18 Apr-19 Oct-19 Apr-20 Oct-20 MSCI Europe Utils (EUR) FTSE 100 (GBP) DONG Share Price (DKK) Reference price (Day-ahead hourly) Strike 150/MWh Source: Thomson Reuters Datastream, Credit Suisse Research Figure 3: Post-tax nominal project-level IRRs for CfDs at different load factors (46%, 50%, 54%), % Project-level post-tax nominal IRR (%) 20% 18% 16% 14% 12% 10% 8% 6% 4% Source: Credit Suisse estimates CfD strike price ( /MWh) Figure 5: 2018E EBITDA mark-to-market sensitivity to different commodity price movements, % 4% 3% 2% 1% 0% -1% -2% -3% -4% DKK:GBP +/-10% Gas price + / - 2/MWh Source: Credit Suisse estimates Wind Load factor + / - 2% Oil price + / - US$5/bbl GB power price + / - 5/MWh Biomass 1/GJ ROC price 54% 50% 46% Thermal coal + / - US$10/tn Source: DBEIS, Credit Suisse research Figure 4: Balance sheet under the scenario of farming-down new projects (DONG metric), DKK bn A 2014A 2015A 2016E 2017E 2018E 2019E 2020E Adjusted interest-bearing net debt (LHS) FFO / Adjusted Interest Bearing Debt (RHS) FFO / Adjusted debt target (RHS) Source: Company data, Credit Suisse estimates Figure 6: EV/EBITDA for DONG and peers 12x 10x 8x 6x 4x 2x 0x 2016E 2017E 2018E 2019E EDPR SSE DONG (with development) Iberdrola DONG (without development) 100% 80% 60% 40% 20% Source: Thomson Reuters (Consensus estimates for EDPR), Credit Suisse estimates, Credit Suisse research (EDPR) 0% DONG Energy A/S (DENERG.CO) 3

4 Table of contents DONG in six key charts 3 1. Key point summary 5 2. Turning high-returning CfDs into value 8 3. We do not assume growth post Returns and cash flow generation to Balance of upside risk in energy prices Why we are ahead of consensus Trading at a discount to renewable peers 32 Appendix 1: Our SOTP-based valuation for DONG Energy 39 Appendix 2: CS HOLT Valuation 40 Appendix 3: Our comparative valuation tables 41 Appendix 4: Our forecast financials 42 Appendix 5: Our commodity forecasts vs market prices 46 Appendix 6: Credit Suisse PEERs 47 Appendix 7: Key management 48 DONG Energy A/S (DENERG.CO) 4

5 1. Key point summary Contracts for Difference (CfDs) generate the best risk-adjusted returns in the sector Key is offshore wind: DONG is the world's largest offshore wind developer, constructor, operator and owner. Around 77% of the EV is in offshore wind. Once operating, these assets generate large cash flows and attractive returns (c7.5-16%). There is c13% in North Sea oil and gas production and 9% in a regulated grid and energy sales. Of the EV, c32% is in four attractive UK projects: Three are under Contracts for Difference (CfDs), which are 15-year fixed-price contracts with strike prices of /MWh earning an average c15% post-tax nominal project-level return. Returns were set in 2013, ahead of a fall in costs. There is a fourth project (Race Bank) under the older 'Renewable Obligation' which earns a similar return. Lower returns more a near-term opportunity than long-term threat Competitive tendering for new projects : Most Governments have moved to reverse auctioning new projects, rather than setting prices administratively. Recent anecdotal evidence suggests returns for projects (in the Netherlands and Denmark) have fallen to WACC (c6-8%). There appears to be a "returns crunch", similar to in Brazilian tenders in We expect the next round of UK CfD tenders due in 2017 to result in strike prices of c 85/MWh. Hence we do not value anything beyond the current pipeline. but low returns an opportunity to monetise value: DONG can monetise partners' willingness to take low-returns. The company is seeking to farm-down (i.e. sell to third parties) c50% stakes in projects under construction. The most recent one (Burbo Bank) got c2x new-build cost. Race Bank is next due H and we expect a similarly high price. The funds keep DONG's financial leverage low (below 1.3x net debt:ebitda) and demonstrates the value for the c50% stakes retained. Regulatory risk less-likely, and Brexit may have no impact on multiples Change-in-law provisions protect CfDs from retroactivity: The CfDs are the bulk of the value within offshore wind and contain change-in-law provisions to protect against future adverse changes. While we cannot rule out changes to other aspects of regulation, no utility is immune. However, DONG partners to reduce the risk. We note the UK Conservative Government appears keen on offshore wind and the renewable budget is now under control, making changes less likely, in our view. Post-Brexit multiples: We think one reason why DONG is not trading close to fair value is uncertainty over the farm-down prices. It remains to be seen whether UK infrastructure returns remain low post the UK's referendum decision to leave the EU. Our call is that given strong macro data and scarcity of high-quality assets, high prices and lower returns are likely, despite our conservative approach. SSE and NG's upcoming sales of UK gas distribution stakes will have read-across on returns acceptable post-brexit. Oil and gas business just c13% of EV and diminishing Wind not sensitive to commodities: The CfDs and major European wind projects are at fixed power prices. Timing and prices achieved on the farm-downs are the major exogenous variable, not power prices. Lower financial gearing (c24% net financial debt to EV versus c40% for integrated utilities) further reduces the equity exposure to any one variable. Legacy oil and gas has to be considered against restructuring: Despite being only c13% of EV, it is still a swing factor for EPS in 2018E. Each +/-c5% on oil and gas DONG Energy A/S (DENERG.CO) 5

6 prices is +/-c2.5% on EPS. Our observation is that North Sea oil and gas production businesses were value-destructive at high energy prices and the strategy to run-down the business is value- and cash-flow enhancing. There is also self-help through costcutting. We value gas and oil production at just c3.6x EV/EBITDA. Conservative assumptions: We do not value the DKK 30.5bn of unrecognised deferred tax assets, and note related gas contracts (LNG, the aborted Hejre project and storage contracts) are in our SOTP at a negative 5% value which assumes a 'worst case'. Peers are EDPR / SSE at the group level and SSE's UK renewables business At the group-level, the peers are businesses such as SSE and EDPR: These trade on c10x consensus FY+1 EV/EVITDA. DONG trades on c8.8x excluding farm-down profits (7.3x including). We believe DONG should trade closer to c9.7x ex. farm-downs. Businesses such as EON (46% German and Eastern European networks) or NG (c96% regulated networks) are not peers, in our view. SSE's wind business for the UK assets: Within DONG, we value the UK wind assets which comprise c53% of EV at c x EV/EBITDA, once they are operating. This is a similar multiple to the c12.5x we put SSE's wind portfolio on. They tend to become annuity-type assets with high cash flow yields and low variability once operating. Dividend payout low (but growing) Dividend sustainable and growing: The dividend yield is just c2.2%, but with x cover in line with a company-set target of a "high single digit annual growth rate". We think the yield reaches c3.2% in 2020E (versus the sector on 5.4%). Higher cash returns are a strategic question for 2019, once the farm-downs are complete and the business has a c14% FCF to equity supported by new assets. That said, the book value per share doubles through to 2020E. Risks We highlight the following risks to our investment case: - Competitive auctions drive returns down and DONG wins and destroys value Adverse changes in energy trading and cross-border business post-brexit Construction risk on own projects (e.g. if projects are late or over-budget) Construction risk (where DONG undertakes EPC wraps for partners) Risk to ROC prices in the event that over-supply of ROCs continues Political risk in Denmark following the appreciation in DONG's value pre-ipo Political risk in the UK given the 107/household cost of DONG's CfDs Oil-linked contracts which result in lumpy profits if market prices are below oil For a full discussion of these risks to our investment case, please see the rest of this report as well as section 7.6. DONG Energy A/S (DENERG.CO) 6

7 Our SOTP-based valuation Figure 7: Summary of our SOTP-based valuation (as at 31 December 2017E) Asset EV EV FY+1 EV/EBITDA Rationale DKK bn % (once complete) Retained UK Wind subject to Renewable Obligation ('RO') 51 30% 11.6 x Mostly new. Some older assets. UK Wind subject to Contracts for Difference ('CfDs') 39 23% 11.7 x All new assets with 15-yr contracts Sale of assets and construction agreements in the UK (CfDs)* 16 9% 3.2 x Value of monetising CfDs and Race Bank (RO) Value of UK wind assets % European offshore wind 23 13% 7.1 x Mix of newer and older assets. Quick subsidy roll-off Sale of assets and construction agreements in Europe* 3 2% 3.2 x Value of monetising Borssele 1&2 and Borkum Riffgrund 2 Value of Continental European assets 26 15% Other 0 0% n/a Mix of businesses (A2SEA, Corporate) Gas and Oil production 22 13% 3.6 x Quick run-down and high tax rates in Norway Bioenergy and thermal power 2 1% 3.8 x Equivalent to value spent on upgrading to biomass Distribution and Customer Solutions 16 9% 8.4 x Sales, LNG losses and regulated business Enterprise valuation % Net debt (26) Hybrid and bonds, netted off with cash Provisions (15) Decommissioning, less some tax deductibility Equity value 130 Share count Valuation DKK 309 * This includes assets in which we expect DONG to farm-down a c50% stake (Burbo Bank, Race Bank, Walney Extension, Hornsea 1, Borkum Riffgrund 2, Borssele1-2). Source: Credit Suisse estimates Upside and downside potential Figure 8: Key assumptions in our valuation thesis Scenario Key basis Detailed assumptions ( ) Blue Sky 370p/sh (=) Base case- 310p/sh ( ) Grey sky- 225p/sh Commodity prices rise and DONG can generate value each year through to 2030 Core Credit Suisse assumptions Commodity prices fall and value-accretion becomes difficult Brent crude oil US$80/bbl, TTF natural gas 20/MWh. Recurring c700mw p.a. of projects across with c20% value-accretion on construction / farm-down. No LNG losses. In the long-term, Brent crude oil US$70/bbl, TTF natural gas 18/MWh c65% average value-accretion on remaining projects through to 2020 (most CfDs). None thereafter Ongoing losses in thermal power and LNG Brent crude oil US$30/bbl, TTF natural gas 10/MWh. Only c20% value-accretion on remaining UK projects and no value-accretion post Nordpool power falls to 18/MWh. Source: Credit Suisse estimates DONG Energy A/S (DENERG.CO) 7

8 2. Turning high-returning CfDs into value UK projects under construction generate cdkk130/share of value for DONG Value of DONG underpinned by CfDs: In late 2013, DONG was awarded 3 Contracts for Difference (CfDs) at c /MWh (real, 2011/12 prices). The returns profile has improved such that these now provide c15% project-level returns; the best in the industry on a risk-adjusted basis. As DONG builds these projects (and one further project under the Renewable Obligation), it will be converted into low-risk EBITDA. We do not see material risk of retroactivity: The CfDs contain change-in-law provisions to protect against future adverse changes and the UK Government budget for renewables is under control. While we cannot rule out changes to other aspects of regulation, no utility is immune and DONG partners with wind farms to reduce the risk (see next section). We note that the current UK Government appears keen on offshore wind. DONG is arguably the most-experienced builder, owner and operator of offshore wind, and remained committed as others withdrew in Pioneer profits in the UK DONG took an early start in offshore wind and persevered with a developer-owner model, especially in the UK. During , subsidy levels were uncertain and subject to Government review. The capital-intensity of offshore wind coupled with costs, which were at the time rising and deterred utilities from investing. The result was that some of the pioneering players pulled out of UK offshore wind (Centrica) or reduced exposure and focused on investments elsewhere (SSE). But DONG remained committed, and has been rewarded with an attractive pipeline of projects to In this section, we show how. 2.2 Re-cap of offshore wind support in the UK Figure 9: Difference between day-ahead reference price and the 150/MWh CfD strike price, /MWh Figure 10: Renewable obligation versus CfDs, /MWh Apr-12 Oct-12 Apr-13 Oct-13 Apr-14 Oct-14 Apr-15 Oct-15 Apr-16 Oct-16 Apr-17 Oct-17 Value of top-up Apr-18 Oct-18 Apr-19 Oct-19 Apr-20 Oct A 2010A 2012A 2014A 2016A 2018E 2020E 2022E 2024E 2026E 2028E 2030E Reference price (Day-ahead hourly) Strike 150/MWh Total ROC Total CfD Source: DBEIS, Credit Suisse research Source: DBEIS, Credit Suisse research, Credit Suisse estimates DONG Energy A/S (DENERG.CO) 8

9 We start by going through the two forms of support. To recall, there are two major subsidy schemes for offshore wind: 1. Renewable Obligation until March 2018 (old scheme): This rewards generators with a certificate that energy suppliers are obligated to buy. The face price of the certificates is 44.77, but the Government intention is to keep the supply in deficit to demand such that they trade at a c10% premium to the face value. The support is available for 20 years from commissioning. The generators receive the support in addition to the power price. This is 'open-ended' and is not subject to limits, hence led to over-build for other technologies. All existing UK offshore wind is under this, and receives support of c x renewable certificates per MWh. DONG has all operating UK projects under this scheme, and one development project (Race Bank on 1.8x). 2. Contracts for Difference from 2016 (new scheme): These are contracts awarded at the UK Government's discretion in competitive tenders. They are structured in a way that there is an inflation-linked "strike price" and to the extent that the market price for power is below- or above- the strike price, the difference (Figure 9) is taken from or passed-into consumer bills through a counterparty owned by the Government. The first contracts were awarded in December 2013 at /MWh 1. The contracts last for 15 years from commissioning, after which generators go to the market price. Much of DONG's value in offshore wind is in the UK. Support was under the RO (power price + subsidies), but new projects are at fixed prices (CfDs) Both types of contract are linked to inflation 2. We show in Figure 10 that across , the all-in CfD price is around c 160/MWh for 15 years, versus ROCs for c /MWh for 20 years. Once the subsidies end, the assets will be eligible for the baseload power price and capacity payments for the remainder of the c25 year asset lives. 2.3 DONG's CfDs bring cdkk75bn of top-up in excess of our power price estimates To ascertain how DONG got its contracts, we first go back to , when there was an investment hiatus in renewables. The Electricity Market Reform was introducing new means of regulation and support for the industry, including Contracts for Difference. But the complexity of what the Government was doing and requirement to get state aid approval from the European Commission meant that it was a long process to give the required clarity to industry. There also appeared to be disagreements in the coalition Government on costs. Investment across the offshore wind industry stopped. High strike prices for offshore wind CfDs at /MWh To re-start investment, the Government in March 2013 decided to run an 'early' tender to enable Final Investment Decisions. The strike prices were set on the basis of consultants' reports and benchmarked to the levels given under the Renewable Obligation. The return was calibrated to around c10-12% on a post-tax nominal project-level basis. The tender was not on the basis of price, but bids would be evaluated on the basis of two nonfinancial criteria:- Project deliverability (75% weighting): Project management, grid connection, procurement of long lead time items, and financing plans. DONG had advanced a number of its projects throughout and was in a position to proceed; and Impact upon industry development (25% weighting): This included expansion of workforce skills and capabilities. DONG committed to using Siemens turbines, where the blades would be manufactured in a new facility to be built in Hull. 1 Albeit no generator commissioned early enough for the 155/MWh 2 UK RPI in the case of the Renewable Obligation Certificates. UK CPI in the case of the Contracts for Difference DONG Energy A/S (DENERG.CO) 9

10 There were a total of c57 projects which applied. But we feel many of the offshore wind projects were not ready, given management teams had not focused upon them. Given DONG kept developing the supply chain and projects across this period, it was successfully selected for three Contracts for Difference. Figure 11: Key Investment contract CfDs awarded out by the UK Government (ex. EDF's Hinkley Point) Asset Technology Owner Start DONG offshore wind CfDs Length (yrs) Capacity (MW) Strike price ( /MWh) NPV of implicit subsidy ( m)* NPV of implicit subsidy (DKK m) Burbo Wind DONG (50%) LEGO (25%) PKA (25%) ,128 9,777 Walney Wind DONG (100%) ,833 24,548 Hornsea Wind DONG (100%) ,744 41,104 Other offshore wind CfDs 8,705 75,429 Dudgeon Wind Statoil (35%) Masdar (35%) Statkraft (30%) ,789 15,503 Beatrice Wind SSE (40%) CIP (35%) SDIC (25%) ,657 23,019 Biomass CfDs 4,446 38,522 Teesside Dedicated biomass CHP Macquarie ,704 14,764 Lynemouth Biomass conversion EPH Power 2018 to ,588 13,761 Drax Biomass conversion Drax Group Plc** 2016 to ,394 20,744 5,686 49,268 Total NPV of implicit subsidy for the eight investment contracts (ex. Hinkley Point) 18, ,219 * Using a c6% project-level discount rate ** Subject to state aid approval from the European Commission Source: Department of Business, Industrial Strategy and Energy, Credit Suisse research, Credit Suisse estimates DONG was awarded three CfD Investment Contracts with an implicit subsidy value of 8.7bn (i.e. an NPV DKK75bn in excess of the power price) We show in Figure 11 the CfDs awarded by competitive tender. The three that DONG was awarded were worth c 8.7bn (cdkk75bn) in excess of our estimate of the GB baseload power price. This is c46% of the total subsidy. Since , they have become even more attractive than the c10-12% headline IRRs for four reasons:- 1. Cheaper debt finance: Borrowing costs are c200bps lower than in 2013, and credit finance is available for construction (e.g. from the Green Investment Bank in the UK); 2. Costs have fallen with the oil price and other energy commodities: Deepwater ports and the vessels rates are cheaper, given reduced spending on offshore hydrocarbon production. We believe rates have since more than halved; 3. Turbine size has risen: The latest 6MW turbines from Siemens and 8MW turbines from Vestas are c2x the size of the previous generation 3-3.6MW machines. And yet the balance of plant (i.e. towers, cables) for a 6MW or 8MW machine is not much more than that for a MW machine; and 4. Turbine efficiency has risen: Longer blades and better turbines mean that load factors of wind farms built across have both improved and also been higherthan-expected. Load factors have reached levels of as high as c50%, when expectations were c42-45% for projects sanctioned in The result of the measures to improve returns is that the level of support that has been locked-in and is higher than the c10-12% envisaged. We think returns are closer to midteens. At the same time, hurdle rates have fallen from c8-10% down to c6-8% for these kinds of projects. From a cash flow perspective, as DONG builds these out across E, it will turn cdkk90bn of capex into DKK15bn of EBITDA. We talk through returns in the next section. DONG Energy A/S (DENERG.CO) 10

11 2.4 DONG's UK projects earn c15% project-level IRRs. This converts to cdkk15bn p.a. EBITDA. Figure 12: Nominal pricing for Contracts for Difference (real 2011/12 prices to the right), /MWh Figure 13: Post-tax nominal project-level IRRs for CfDs at different load factors, % E 2019E 2021E 2023E 2025E 2027E 2029E 150/MWh CfD (FIDeR) 140/MWh CfD (FIDeR) 114/MWh CfD (T1) 120/MWh CfD (T1) 105/MWh (Future tenders) 85/MWh (Future tenders) Project-level post-tax nominal IRR (%) 20% 18% 16% 14% 12% 10% 8% 6% 4% CfD strike price ( /MWh) 54% 52% 50% 48% 46% 44% Source: DBEIS, Credit Suisse estimates The investment contracts have since appeared to be highly generous. We think they earn post-tax nominal project-level returns of c15% Source: Credit Suisse estimates We show in Figure 12 the levels of strike prices for offshore wind, and what different load factors might mean (Figure 13). We see three different groupings of CfD: 1. The investment contracts signed at c 140 or 150/MWh real: We estimate that on the basis of the c48-50% load factors stated by the company, DONG's Hornsea and Walney extension projects earn c16% project-level returns. Burbo bank is lower, at c14%, given we estimate a c46% load factor. We estimate an average c15%. 2. Enduring CfDs have been won on the basis of /MWh real: The UK Government has run one subsequent auction to the investment contracts. The runners-up for the investment contracts were successful in the first auction. Iberdrola won 120/MWh for 0.7GW and Mainstream renewables won c 114/MWh for 0.4GW. We think these will have been at c10-12% project-level nominal IRRs. 3. Caps for future enduring auctions at /MWh real: In their March 2016 budget, the UK Government has already said that there will be a cap on future CfD strike prices, starting at 105/MWh but falling to 85/MWh for projects commissioning in We cover this in more detail later in this section. But we expect such levels to make for c6-10% project-level IRRs. The lower end of the range is more likely, in our view. The outcome is that we think the projects will be worth closer to c11-12x EV/EBITDA for a 15-year contract. They are cash-flow generative once running with generous tax writing down allowances there is also a residual cash flow from years This equates to c100% value-accretion, on our numbers. DONG Energy A/S (DENERG.CO) 11

12 2.5 Why we do not think that the CfDs will be subject to retroactive changes Given the CfDs have now led to attractive c15% IRRs for those developers holding them, we think it is worth considering retroactive cuts from the Government as a risk. Our view is that risk of these CfDs being contested is very low for the following reasons:- Change-in-law provisions: The CfDs contain provisions to protect the holder in the event of unforeseeable changes of law that are discriminatory or apply to CfD holders or the CfDs themselves (e.g. to protect against a generator tax on either the CfD or other projects where the Government determines it is over-earning). This includes protection from future Acts of Parliament. The CfD holder would be entitled to compensation for opex, capex and lost revenues in the event of a change. Renewable budgets under control: There is a budget envelope for subsidies for renewables, which reaches c 9.8bn by the year ended March This is called the "levy control framework" and is a cap the Government work to. In the past, this was exceeded. The Government ended support for many new renewable projects in (but not offshore wind) and we believe the budget is now under control. Government support: The current Conservative administration have a strong preference for shale gas and offshore wind. They cut subsidies for most mature technologies (e.g. onshore wind and solar). We expect most of the remaining funds to go towards new-build nuclear and offshore wind (see later this section). In the event of retroactive changes, we think the costs of future projects would increase. We think risk of changes to the CfDs and the c15% project-level IRRs that they bring is low. However, Governments are involved in more than just the support mechanisms, with the LEC exemption removal (see below) being one example as well as the CO2 tax (click here for more detail). We do not think any utility is immune, but we do think that the bulk of the return is safe and the CfD appears more secure than other sources of contract. DONG Energy A/S (DENERG.CO) 12

13 The CfDs have change in law provisions to protect their returns. Albeit we cannot rule out changes by Government around the edges (such as the LEC exemption) Case study: removal of the LEC exemption In May 2015, the Conservative Government won a surprise majority and launched a budget for Summer In this budget, the Government removed the exemption from the Climate Change Levy for renewable electricity from 1 August The Climate Change Levy was a c 5.4/MWh tax on energy supplied to businesses / public sector that comes from fossil-fuel fired generation. Renewable generators were able to internalise this and capture most of the c 5.4/MWh through a mechanism called the Levy Exemption Certificate (LEC). There was no legal commitment to the LEC (unlike CfDs). But the LEC was included when setting renewable subsidy levels and the c10-12% returns (e.g. the 140/MWh and c 150/MWh CfDs). The reasons given for the end of the exemption were threefold: (i) one-third the value would go overseas; (ii) the value falls post-2020; and (iii) more effective policies have been put in place. It was forecast to raise an extra c 450m revenue in , rising to 910m in Two-thirds of the impact will be incurred by the UK renewables industry. The impact on offshore wind was low, with just c0.6% reduction in project-level IRRs. The impact on DONG was also very low (e.g. c1.5% of our current EV). However, for biomass, where the margins are much lower, the impact was more pronounced. We estimated at the time it was c20% of the value for a biomass generator, such as Drax. Drax and Infinis the two major biomass generators pursued a judicial review against the UK Government. The judge ruled against Drax and Infinis on each of three grounds: (1) Officials are entitled to change their minds when debating policy outside the public domain; (2) Domestic courts cannot as primary decision makers; and (3) Property rights exclude the hope of future contractual income. Given our reading of the High Court documents, we are not sure that the Government realised what the impact on biomass generators would be. However, there has been no observable impact on future investments in the sector or for Drax's investment appetite. We therefore cannot rule out small changes in future to some of the regulation away from the CfDs, even if DONG's CfDs are protected by change in law provisions. DONG Energy A/S (DENERG.CO) 13

14 3. We do not assume growth post-2020 Subsidies have fallen to WACC. We do not expect growth for DONG post-2020 We think the best returns are over: Governments have moved to competitive tenders for wind i.e. reverse auctions rather than administratively set prices. This has pushed down returns. Recent examples in Europe suggest returns for new offshore wind projects have fallen close to WACC (i.e. c6-8%). An example is DONG's Borssele, which we think is on a return of c7.5%. Major tenders in the UK and Germany in 2017E: Key will be the potential 2017 CfD tender in the UK, where we expect a clearing price to be closer to the 2025E target of c 85/MWh for c1.5gw than the cap of 105/MWh. There will also be a tender in Germany for 1.7GW. We do not assume value-accretion or growth for DONG post Bigger question is where projects come from: Even before we get to returns, we think growth rates will not be high. We are not as optimistic as BNEF forecasts in the DONG IPO prospectus. That said, we think the CfDs we spoke about in the previous section are enough to be positive on DONG. 3.1 Lower levelised cost of electricity for offshore Figure 14: Levelised cost of electricity (2016 prices). Based on a c6% hurdle rate, but with illustrative dotted lines showing what it would be under a c9% hurdle rate, /MWh CCGT (US) Lignite (Europe) CCGT (Asia) CCGT (Europe) 59 Onshore wind Hard coal (Europe) Hydro 82 Offshore wind 118 Nuclear x ROCs Fuel cost CO2 Other variable Fixed O&M (cash) Capital repayment Return CfD ( 140/MWh strike) CfD ( 150/MWh strike) Note that we assume US$3.5/mmBTU US natural gas, US$10/mmBTU European and Asian natural gas, 4.2/tn CO2 emission permits, US$16/tn lignite, 6% cost of capital (except where illustrated), US$55/tn API2 coal and US$15/tn transport. CCGT= Combined Cycle Gas Turbine (gas-fired power station). Dotted line is for a c9% IRR hurdle (post-tax nominal project-level) Source: Credit Suisse estimates Cost of capital has come down, and this has reduced the levelised cost of energy for capitalintensive technologies. We think that the cost of offshore wind is closer to c 82/MWh. Subsidy levels and structures have changed of late. Most markets have now moved to competitive tender. This means that bidders must submit prices usually with a cap on DONG Energy A/S (DENERG.CO) 14

15 what they can bid rather than rely upon consultants' reports and administratively-set prices, as was the case with the initial CfD investment contracts. In the current environment, this should lead to lower prices. We revisit the levelised cost of electricity in a low-returns world. That is to say the all-in cost of generation from a given technology, including repayment of capital and a return on capital. We show in Figure 14 the costs, assuming a c6% IRR, which we think is the current WACC (with a c9% IRR super-imposed; what we think the WACC was five years ago). We find the following points of interest:- The fall in cost of capital outweighs lower fossil fuel prices: The fall in capital has made renewables more competitive, despite the fall in gas and coal prices. In a way, we see lower WACC (driven by monetary policy and lower bond yields) has been more relevant than the fall in Brent crude and thermal coal prices across the past 2-5 years. Offshore wind is now more competitive: We think the levelised cost is c 82/MWh. This is consistent with recent tenders in the Netherlands won by DONG (at 73/MWh excluding transmission costs, implying a c7.5% IRR) and in Denmark provisionally won by Vattenfall ( 64/MWh, but near-shore). 3.2 Likely to see returns closer to c6-8% from here Figure 15: Competitive wind tenders: clearing prices /MWh /MWh Dudgeon 179 Walnea Extension 179 Burbo Bank 179 Beatrice 167 Hornesa 167 Administratively-set prices East Anglia Phase 164 Neart na Gaoithe 157 Horns Rev Competitive tenders Borssele III and IV 120 Zuid-Holland I and II 108 Zuid-Holland III and IV 103 Noord-Holland I and II 100 Borssele I and II 73 Danish near-shore tender 64 0 May-13 Oct-13 Apr-14 Oct-14 Apr-15 Sep-15 Mar-16 Sep-16 Source: Platts We think returns for offshore wind are likely to be c6-8%, as per recent competitive tenders. The risk is that returns continue to fall In a world of quantitative easing and loose monetary policy, it has been the case that money has found its way into wind. We show in Figure 15 the clearing prices of the initial projects and the recent raft of competitively-awarded tenders. We saw that in Brazil in 2011 where real equity returns fell to c0% for onshore wind projects. Usually, such "return crunches" end following: (a) changes in macro conditions (recession and monetary tightening); and (b) projects that over-run and have no buffer to keep return above WACC. This forces bidders to be more rational in future. Given recent data points, we think returns are now close to c6-8% in developed markets. We believe that DONG won the 700MW Borssele project on c7.5%, and that the value accretion (i.e. the value in excess of WACC) is closer to c5-10% than c100%. This is DONG Energy A/S (DENERG.CO) 15

16 enough for an efficient player with a cost and operational advantage to win projects, but not enough to change our valuation for DONG. 3.3 We do not share consultants' bullish views of growth in offshore wind Figure 16: Bloomberg New Energy Finance forecasts (presented by DONG): Installed offshore wind (GW)- annual installations and accumulated capacity Installed offshore wind (GW) 2015A accum. 2016E 2017E 2018E 2019E 2020E 2021E 2022E 2023E 2024E 2025E United Kingdom Germany Denmark Netherlands Belgium France Sweden Other Europe total China Japan Taiwan Korea (Republic) Other Asia total United States Other North America Global Source: Bloomberg New Energy Finance, DONG The bigger issue pertains to whether the growth projects will come through post-2020, not the returns. We are more skeptical than consultant reports. We only focus upon projects under construction in our DONG investment case In our view, tenders in the UK in 2017 (where prices will likely fall to 85/MWh; real 2011/12 money) and Germany will be the ones to watch for growth DONG Energy A/S (DENERG.CO) 16 accum. We show above the data used by DONG in their IPO prospectus. We generally think that such forecasts are very ambitious (we note similarly aggressive forecasts for onshore wind in 2007 and 2008). We make the following points:- UK likely opportunity in three tenders: Offshore wind installations will grow from c0.7gw in 2009 to c6gw as of 2016, on our estimates. This is going to reach c10.8gw by 2021 given projects under construction. The UK Government has said there will be c 730m made available for new CfDs in the current parliament. This will be spread across three auctions, with the first one being c 290m, and likely in The Government has said that the auctions could support up to 4GW of further offshore wind, which would take the UK to c15gw of offshore wind (c19% of electricity production). The first auction in 2017E will only be for less-established technologies, and we think most will go to offshore wind. There will be a cap on the price at 105/MWh, falling to 85/MWh for projects commissioning by We think the first auction will clear closer to 85/MWh, and that it could support up to 1.5GW. We think DONG will seek to tender in remaining capacity at Hornsea. We see competition from some of the 4.8GW at Dogger bank (SSE, Statoil, Statkraft), 1.1GW

17 at Inchcape (EDPR, SDIC) and 0.9GW at Triton Knoll (Statkraft). The auctions will likely be >2x over-supplied, in our view. Germany returns also likely to fall: Bloomberg new Energy Finance forecasts show c6.8gw of offshore wind in Germany by 2020, all of which is under construction. The EEG renewable energy act has been passed in the German Parliament and will mean that there will be an auction for new offshore wind projects after it takes effect in January There is an additional c3.1gw, which the Federal Energy Ministry has proposed will be awarded in a competitive tender for 1.7GW in 2017 and 1.4GW in From 2026, the Government will present sites itself (rather than the developer) under a centralised model to encourage competition. There will be an additional c0.84gw annually from in order to reach c15gw by Alternative geographies are still less-mature DONG has flagged c2-3gw of potential in China and has a strategic ambition for some in Asia (China, Japan and Taiwan) and the US. We think better returns for pioneers may be possible, but we do not count on it. We yet to see any large projects reach FID stage in these regions. We run through them in detail:- United States: The first offshore wind farm is small at just c30mw, and is set to be complete by year-end. There have been press reports of offshore projects on the Eastern Seaboard for at least the eight years that we have been covering the industry. However, legal threats and lack of regulatory support have continually delayed projects. A new energy law in Massachusetts signed on 8 August supports offshore wind and requires 1.6GW by DONG holds some leases. China: Offshore wind is growing slowly. Bloomberg data presented by DONG suggest c27gw of offshore wind by Of the domestic players, only China Longyuan is a meaningful operator. Our Asian Capital Goods analyst, Edmond Huang, cites low returns as a reason the industry is held back. Our view is that China very quickly builds local champions that take the market in the OEM space, and development is done by local players. We expect DONG to need to partner. France: France does not have any offshore wind, but Bloomberg forecasts c5gw by There have been two major tender rounds in 2012 and DONG was in one of the tenders with EDF but has since sold it 40% stake to Enbridge. We note that local content (e.g. turbines by Alstom and Adwen) is used. Taiwan: The current Government is pro-renewables and aims to phase out nuclear by It is set to revise bidding mechanisms to encourage renewables (including offshore wind). The target is 3GW of offshore wind. Earthquakes and typhoons are a risk for offshore wind projects. Conclusion: We focus upon the CfD pipeline Overall, we keep Europe in focus for our projections and valuation on the projects up to 2020, and not beyond. The "returns crunch" needs to re-set first. We expect not only lower returns, but also see more uncertainty in emerging markets for wind. Hence we would be surprised to expect any large projects in Asian or North America. Our investment thesis solely focuses on projects under construction and out until We think the CfDs and growth out to 2020 are core of the value. DONG Energy A/S (DENERG.CO) 17

18 4. Returns and cash flow generation to 2020 Farm-downs necessary but will bring forwards value-accretion and set a precedent Farm-downs of c50% stakes essential: DONG has a strategy of selling stakes in projects months into construction. We think this is necessary as DONG does not yet have the balance sheet to fund them all and retain its BBB+/Baa1 corporate issuer rating. We expect 1-2 farm-downs per year to cycle the cash flows back to DONG. Farm-downs to earn high returns: Offshore wind is attractive, given little revenue variability and c0.5% 10yr bond yields. DONG's c50% stake in Burbo Bank was sold in February 2016 for a c100% premium to cost. While the investment climate for UK infrastructure remains to be seen post-brexit, we expect it to be robust. We look to upcoming assets sales by NG and SSE as comps and lead-indicators. We conservatively assume bidders take c7% post-tax nominal IRRs on acquisitions. Cash flow and returns: DONG will have one of the best cash-flow and returns profiles in the sector, on our numbers. Net debt:ebitda peaks at 1.3x in December 2018, before declining, with c16-29% group ROEs. From 2020E once the assets are complete the company would have a c14% FCF to equity with negligible debt. 4.1 Capital-intensity requires farm-downs The capital expenditure on DONG's offshore wind projects is cdkk90bn. This is c1.4x current invested capital and c4x EBITDA. Such a level would be unaffordable for DONG without issuing new equity or risking higher financial gearing and a lower credit rating. The company therefore needs to partner in order to capture value, and sells c50% stakes. Illustration of the benefit of farm-downs Figure 17: Balance sheet under the scenario of not farming-down new projects, DKK m Figure 18: Balance sheet under the scenario of farming-down 50% of new projects, DKK m 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10, A 2014A 2015A 2016E 2017E 2018E 2019E 2020E 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10, A 2014A 2015A 2016E 2017E 2018E 2019E 2020E 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Adjusted interest-bearing net debt (LHS) FFO / Adjusted Interest Bearing Debt (RHS) FFO / Debt target (RHS) Adjusted interest-bearing net debt (LHS) FFO / Adjusted Interest Bearing Debt (RHS) FFO / Adjusted debt target (RHS) Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates DONG Energy A/S (DENERG.CO) 18