30 June 2017 Asia Pacific/China Equity Research Utilities

China Solar Sector Research Analysts INITIATION Gary Zhou, CFA 852 2101 6648 [email protected] The two new trends

Dave Dai, CFA 852 2101 7358 [email protected] Figure 1: Expanding market share of distributed solar and rising Gloria Yan demand for mono-Si solar components 852 2101 7369 80% 71% 74% [email protected] 60% 50% 50% Distributed solar as % of 40% 45% 32% 40% 25% 30% China's annual solar 18% installations 20% 9% 12% 0% Market share of mono-Si 2015 2016 2017E 2018E 2019E 2020E Source: NDRC, Credit Suisse estimates

■ Mono-Si gaining share on the upstream. Despite global solar market likely growing by single digits in the next few years, we expect demand for Mono-Silicon (mono-Si) solar components to more than double during 2016-20, driven by market share expansion (from 25% to 50%) due to its superior cost efficiency over competing multi-silicon (multi-Si) technology. Such a trend should benefit mono-Si wafer manufacturers (led by LONGi) with strong volume and earnings growth, while multi-Si rivals (such as GCL Poly) may suffer from shrinking demand. ■ Rising distributed solar in China. For China's solar operator market, another major change we expect is the shift from ground-mounted solar farms to distributed solar (largely rooftops). Such a trend is warranted by a lower land requirement, limited power curtailment and lucrative returns for distributed solar (an equity IRR of >15% vs 10% for solar farms). We believe that the forecasted 55% capacity CAGR (60GW target by 2020) during 2016-20 for distributed solar should help to drive up demand for mono-Si products (extra 100-200 bp IRR gains). ■ Headwinds for ground-mounted projects. In comparison, the traditional ground-mounted solar market may slow to only a 12% capacity CAGR during 2016-20. Key headwinds include further on-grid tariff cuts (we expect another 15-20% cut in 2H17), expanding competitive tenders (>80% in 2017) and rising interest rates. With >90% of existing project exposure in ground-mounted solar, we expect most listed operators (except Linyang) to face challenges in defending profitability. ■ Stock calls. We initiate coverage on the two upstream solar wafer producers: LONGi (OUTPERFORM, TP: Rmb23.0 with 34% upside, and the best positioned in the expanding mono-Si market) with a 25% FY17- 19E EPS CAGR and GCL Poly (NEUTRAL, mainly in multi-Si market). For operators, we downgrade XYS to NEUTRAL with headwinds for ground- mounted farms and upgrade Linyang to NEUTRAL given exposure to distributed solar. Key risks: Industry competition and policy changes.

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.

30 June 2017

Focus charts and table

Figure 2: Mono-Si has both cost and efficiency Figure 3: Mono-Si is gaining market share at the advantages expense of traditional multi-Si

(Rmb/W) 100% 20.0% 19.7% 2.3 18% 25% 2.2 30% 19.5% 80% 40% 45% 2.2 50% 19.0% 60% 18.5% 18.2% 2.1 Mono-Si 2.0 40% 82% 18.0% 75% 70% Multi-Si 2.0 60% 55% 50% 17.5% 20% 17.0% 1.9 Mono-Si Multi-Si 0% 2015 2016 2017E 2018E 2019E 2020E Cell efficiency Module production cost (Rmb/W) Note: Pure production cost excluding margins at wafer, cell and module production Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates processes. Source: Company, Credit Suisse estimates

Figure 4: Higher solar farm IRR by using mono-Si Figure 5: Mono-Si price premium supported by products due to lower BOS cost (land, cables, etc) higher efficiency

(IRR) BOS cost 200bp (Rmb/W) 18% 17% 20.0 15% 16% 15.0 14% 100bp 10.0 12% 11% 10% 10% 5.0 3.0 2.7 3.5 3.2 8% 0.0 Ground-mounted Ground-mounted Distributed (multi- Distributed (multi-Si) (mono-Si) Si) (mono-Si)

Note: BOS stands for Balance of System (the system cost of a solar farm, including land, Source: WIND cables, etc. and accounting for around half of total initial investment) Source: Credit Suisse estimates

Figure 6: LONGi is a key beneficiary of rising Figure 7: Decent profitability for mono-Si wafer (on demand for mono-Si solar components current spot price) (GW) Wafer Module 60 60% (Rmb/W) Poly Multi Mono Multi Mono 40 40% Cost 0.38 0.85 0.78 2.52 2.58 ASP 0.48 1.02 1.18 2.70 3.00 20 20% Gross margin 21% 17% 34% 7% 14% 0 0% LONGi's ex posure* - - 90% - 10% 2015 2016 2017E 2018E 2019E 2020E GCL's ex posure* 60% 40% - - - LONGi' output Global demand for mono-Si LONGi's market share Source: Company data, Credit Suisse estimates Note: * Measured by FY18E gross profit exposure. Source: Bloomberg, WIND, Credit Suisse estimates

Figure 8: Rating, target price and EPS change summary Company Ticker Rating Target price (FC) EPS change (%) New EPS (LC) P/E (x) New Old New Old FY17E FY18E FY19E FY17E FY18E FY19E FY17E FY18E FY19E LONGi 601012.SS O n.a. 23.00 n.a. n.a. n.a. n.a. 1.01 1.32 1.59 16.9 13.0 10.7 GCL 3800.HK N n.a. 0.80 n.a. n.a. n.a. n.a. 0.10 0.10 0.10 7.0 7.2 7.1 XYS 0968.HK N O 2.40 3.87 -14% -21% -31% 0.32 0.33 0.31 6.9 6.8 7.1 GCLNE 0451.HK N N 0.40 0.50 0% -28% -22% 0.04 0.04 0.04 7.8 8.9 7.9 Linyang 601222.SS N U 7.50 6.80 -9% 8% 13% 0.42 0.48 0.48 17.5 15.4 15.1 CNE 0182.HK O O 0.47 0.56 -23% -26% -38% 0.06 0.08 0.08 4.6 3.8 3.4 Note: Priced as of 28 June 2017. Source: Company data, Credit Suisse estimates

China Solar Sector 2 30 June 2017

The two new trends Prefer mono-Si over We initiate coverage on the China solar wafer manufacturing sector. Given higher cell multi-Si efficiency and reduced costs, mono-silicon (mono-Si) outperforms multi-silicon (multi-Si) on project returns—a key reason why it has gained favour in recent years. We forecast mono-Si's global market share to double from 25% in 2016 to 50% by 2020, and demand- supply balance should stay intact in the next two years. Turning cautious on On solar farm operators, we are turning more cautious given potential declines in annual Chinese solar farm ground-mounted project quotas, continuing tariff cut risks (the next round could take place operators in 2H17 with a likely ~15-20% cut), and expanding competitive project bidding (>80% in 2017). Rising interest rates are an additional risk. Mono-Si gaining share on the upstream Mono-Si has superior With ~10% higher efficiency, we believe that mono-Si modules are justified in having a cost efficiency ~10% price premium (~Rmb0.3/W) to multi-Si, due to the lower Balance of System (BOS) cost (such as land, cables, etc.) required. After technology upgrades and capacity expansion, we estimate the pure production cost of mono-Si modules (for market leaders such as LONGi) is 10% lower than multi-Si, leaving mono-Si makers flexibility in balancing market share expansion and margin protection. Rising distributed solar in China Distributed solar is a Despite a likely growth slowdown for China’s solar power sector, we believe that growing opportunity distributed solar, backed by strong policy support, could take off with a likely 55% CAGR in China for 2016-20 (60GW target by 2020). The key advantages of distributed solar include flexible locations (largely rooftops), limited curtailment issues and generous subsidies. During the last few rounds of solar tariff cuts, distributed tariffs stayed intact. We calculate an equity IRR of >15% for distributed solar, higher than ~10% for ground-mounted solar farms. Headwinds for ground-mounted projects Ground-mounted Generally, we expect the demand shift from ground-mounted solar to distributed solar to projects are less be negative for traditional solar farm operators under our coverage (such as Xinyi Solar preferred in China and GCL New Energy), given that >90% of their current capacity exposure is in ground- mounted solar. As a result, we cut our annual capacity additions for those companies by 20-40%. At the same time, the expansion of project bidding (>80% in 2017) has been quicker than our previous expectation, likely to result in lower tariffs and thus lower returns (an equity IRR from 14-10%). Key upside risks could be further industry consolidation and listcos' quicker-than-expected expansion into distributed solar. Stock calls We initiate coverage on the two upstream solar wafer producers: LONGi (OUTPERFORM, TP: Rmb23.0 with 34% upside, and the best positioned in the expanding mono-Si market) with a 25% FY17-19E EPS CAGR and GCL Poly (NEUTRAL, mainly in multi-Si market). For operators, we downgrade XYS to a NEUTRAL with headwinds for ground-mounted farms and upgrade Linyang to a NEUTRAL given exposure to distributed solar. Prefer LONGi over GCL LONGi is well positioned in the mono-Si wafer subsector (a 15% global market share) and still expanding its capacity (a 31% CAGR from 2017-19E) to meet rising mono-Si demand. Strong earnings growth (a 25% FY17-19E EPS CAGR) is supported by stable margins and growing market share. GCL Poly is also a leader in polysilicon (~17% global market share) and multi-Si wafer production (~20% market share). For GCL, the wafer price is more essential for its total profits, which could see pressures due to oversupply and market share loss.

China Solar Sector 3

Sector China Solar Figure 9: Global solar valuation comparison Name Ticker Rating Price TP U/D Mkt P/E (x) P/B (x) Yield (%) ROE (%) EPS CS analyst (LC) (LC) cap CAGR (%) US$ bn 17E 18E 19E 17E 18E 19E 17E 18E 19E 17E 18E 19E 17-19E

Solar component manufacturing Poly GCL Poly 3800.HK N 0.80 0.80 0% 1.9 7.0 7.2 7.1 0.6 0.5 0.5 0.0 0.0 0.0 8.5 7.7 7.2 -0.8 Gary Zhou Wacker Chemie WCHG.DE N 95.2 99.0 4% 5.6 34.4 24.3 n.a. 11.3 11.3 n.a. 2.3 2.9 n.a. n.a. n.a. n.a. n.a. Mathew Hampshire-Waugh OCI Co Ltd 010060.KS NC 92,400 n.a. n.a. 1.9 29.0 22.0 18.2 0.7 0.7 0.6 0.4 0.4 0.5 2.7 3.7 4.7 26.3 Not covered Daqo New Energy DQ NC 20 n.a. n.a. 0.2 25.0 19.8 16.7 1.7 1.6 1.5 3.5 2.7 3.1 7.0 8.6 8.4 22.4 Not covered Weighted average 27.8 20.4 12.9 6.9 6.9 0.6 1.5 1.8 0.4 5.7 5.8 6.1 13.2 Wafer LONGi 601012.SS O 17.11 23.00 34% 5.0 16.9 13.0 10.7 3.8 3.0 2.5 0.9 1.5 2.3 21.1 25.9 25.6 25.4 Gary Zhou GCL Poly 3800.HK N 0.80 0.80 0% 1.9 7.0 7.2 7.1 0.6 0.5 0.5 0.0 0.0 0.0 8.5 7.7 7.2 -0.8 Gary Zhou Weighted average 14.2 11.4 9.7 2.9 2.4 2.0 0.6 1.1 1.7 17.6 20.9 20.6 18.2 Cell & module JA Solar Holdings JASO N 6.41 6.00 -6% 0.3 n.a. 17.8 10.7 1.6 1.6 1.5 0.0 0.0 0.0 0.1 1.8 2.9 n.a. Maheep Mandloi Jinko Solar JKS O 20.22 20.00 -1% 0.6 10.7 7.7 7.3 0.7 0.6 0.6 0.0 0.0 0.0 6.4 8.8 8.4 21.2 Maheep Mandloi Canadian Solar Inc CSIQ.OQ NC 15.39 n.a. n.a. 0.9 11.6 6.9 8.2 0.9 0.8 0.8 0.0 0.0 0.0 10.7 7.9 10.1 19.2 Not covered Weighted average 11.2 9.0 8.3 0.9 0.9 0.8 0.0 0.0 0.0 7.5 7.2 8.3 20.0 Solar farm operators China Solar Farm Xinyi Solar 0968.HK N 2.24 2.40 7% 2.1 6.9 6.8 7.1 2.0 1.7 1.5 6.2 5.9 5.6 31.9 27.5 22.4 -1.4 Gary Zhou GCL New Energy 0451.HK N 0.39 0.40 4% 0.9 7.8 8.9 7.9 1.2 1.0 0.9 0.0 0.0 0.0 16.5 12.5 12.5 -0.5 Gary Zhou Linyang Energy 601222.SS N 7.34 7.50 2% 1.9 17.5 15.4 15.1 1.5 1.4 1.4 2.9 3.3 3.4 8.8 9.5 9.3 7.4 Gary Zhou Concord New Energy 0182.HK O 0.33 0.47 42% 0.4 4.6 3.8 3.4 0.5 0.4 0.4 6.6 8.0 8.8 10.3 11.7 11.8 15.5 Gloria Yan Weighted average 10.7 10.0 9.9 1.6 1.4 1.3 4.0 4.1 4.0 19.5 17.4 15.3 3.1 Note: O = Outperform, N = Neutral, NC = Not covered. Priced as of 28 June 2017. Source: Company data, Bloomberg consensus estimates for non-covered stocks, Credit Suisse estimates for covered stocks

30 June 2017 30 June

4

30 June 2017

Table of contents

The two new trends 1 Mono-Si gaining share on the upstream ...... 3 Rising distributed solar in China ...... 3 Headwinds for ground-mounted projects ...... 3 Stock calls ...... 3

Mono-Si gaining share on the upstream 7 Mild growth in global demand ...... 7 Balanced supply and demand for mono-Si ...... 8 Mono-Si is preferred by solar farm operators ...... 12 Cost efficiency is the key ...... 13 New technology (PERC) also favours mono-Si ...... 14 LONGi is the leader and pioneer in mono-Si ...... 16 Black silicon and diamond wiring may be the silver lining for multi-Si ...... 16 ASP/cost/margin comparison of leading mono-Si/ multi-Si manufacturers (LONGi and GCL) ...... 17

Rising distributed solar in China 20 Superior project return ...... 20 Accelerating capacity additions in 13th FYP ...... 22 Little power curtailment and no subsidy delay ...... 23 Stock implications ...... 24

Headwinds for ground-mounted projects 26 Uncertain quota outlook in 13th FYP ...... 26 Next round of tariff cut discussion in 2H17 ...... 26 Project return to be capped by bidding ...... 27 Rising interest rate is an additional risk ...... 28 Cutting capacity and tariff assumptions for listcos ...... 29

Stock calls 30 Prefer LONGi over GCL Poly ...... 30 Downgrade ground-mounted solar farm operators ...... 32 Summary of stock ideas ...... 32 Key risks ...... 33

Appendix 36 Introduction of mono-Si and multi-Si technologies ...... 36

LONGi Green Energy Technology (601012.SS / 601012 CH) 37

China Solar Sector 5 30 June 2017

Focus charts and table 38

Leader in a growing market 40 Growth driven by market share expansion ...... 41 Cost leader in the industry ...... 43 Mono-Si wafer is the main earnings driver ...... 44 One of the strongest balance sheets ...... 45 Key risks ...... 46 Company background ...... 47

GCL-Poly Energy Holdings Ltd 49

(3800.HK / 3800 HK) 49

Focus charts 50

Clouded outlook 52 Profit capped by weak multi-Si wafer price ...... 53 Poly is more defensive but watch out for low-cost capacity expansion ...... 55 Solar farm remains a growth driver ...... 58 Valuation ...... 59 Key risks ...... 60 Company background ...... 61

Xinyi Solar Holdings (0968.HK / 968 HK) 62

GCL New Energy Holdings (0451.HK / 451 HK) 64

Jiangsu Linyang Energy (601222.SS / 601222 CH) 66

Concord New Energy (0182.HK / 182 HK) 68

China Solar Sector 6 30 June 2017

Mono-Si gaining share on the upstream We expect mono-Si's Despite the global solar market likely growing by single digits in the next few years, we market share to double forecast substantial market share gains in mono-silicon (mono-Si) solar components from during 2016-20 25% in 2016 to 50% by 2020 thanks to its superior efficiency, lower cost and rising support from the China market. This structural change is expected to come at the expense of multi-silicon (multi-Si). As a result, the upstream market leaders in the two competing technologies—LONGi (~15% global share focusing on mono-Si) and GCL Poly (~20% global share on multi-Si)—could face different growth outlooks.

Figure 10: Growth outlook comparison between LONGi and GCL 2014 2015 2016 2017E 2018E 2019E Wafer output growth LONGi 83% 36% 128% 48% 42% 28% GCL 39% 18% 15% 3% 7% 3% Gross profit growth LONGi 124% 94% 162% 27% 27% 18% GCL 107% 16% 22% -11% 6% 7% Net profit growth LONGi 312% 77% 197% 31% 30% 21% GCL n.a. 57% -16% -9% -2% 1% Source: Company data, Credit Suisse estimates Mild growth in global demand Global solar After record-high annual solar installations in 2016 (largely driven by China), global solar demand may grow at demand may expand at a mild 6% CAGR from 2016-20, based on various industry single digits forecasts (e.g. Bloomberg New Energy Finance, GTM Research, etc.). We expect the China market to slow in the next three years, mainly on falling utility-scale ground-mounted solar project quotas, partly offset by rising distributed solar demand. For the US market, the CS Global Solar Research Team forecasts 9 GW demand in 2017 (vs. 15 GW in 2016) due to a demand rush from the pre-ITC (Investment Tax Credit) extension in 2016. On the utility-scale side (for the US), the unknown is Suniva's 201 trade petition, which, if approved, could increase the landed price of modules 2-3x in the US and drive down demand due to higher cost. Solar demand in India is driven by: (1) a strong national policy which aims to install 100 GW by 2022 (or a 55% CAGR over 2016 levels); low-cost solar power purchase agreements (PPA) which are competitive with traditional fuels; and (3) state-level renewable obligations requiring an additional 40GW be installed through 2019.

Figure 11: Global solar demand (industry forecasts) (GW) 100 Others 80 Europe 60 Japan 40 India 20 US 0 China 2011 2012 2013 2014 2015 2016 2017E 2018E 2019E 2020E

Source: GTM Research, Bloomberg New Energy Finance, Credit Suisse estimates

China Solar Sector 7 30 June 2017

Balanced supply and demand for mono-Si The global solar supply value chain has been suffering from a serious oversupply in the past few years. The demand surge in 1H16 offered temporary relief, which has disappeared since 2H16 with continuous price drops. Mono-Si wafers are probably one of the very few along the solar value chain that have had close to 100% capacity utilisation and may continue to see high utilisation in the next few years.

Figure 12: Capacity utilisation forecasts

140% 120% 100% Poly 80% Mono-Si wafer 60% Multi-Si wafer 40% 20% Cell 0% 2011 2012 2013 2014 2015 2016 2017E 2018E

Source: Credit Suisse estimates

China Solar Sector 8 30 June 2017

Figure 13: Global solar supply and demand analysis 2011 2012 2013 2014 2015 2016 2017E 2018E Global solar supply chain Polysilicon (kt) GCL Poly 65 65 65 65 70 70 75 115 Wacker Polysilicon 32 52 52 52 56 80 80 80 OCI 36 42 42 42 52 52 52 52 REC Silicon 18 18 18 20 20 20 30 40 Hemlock Semiconductor 36 36 32 32 32 32 32 32 Tokuyama 9 9 17 22 22 25 28 28 SunEdison 15 9 9 19 23 23 23 23 Daqo 4 4 5 6 12 12 18 18 Others 76 88 96 117 127 133 138 153 Total global poly capacity (kt) 290 323 335 375 413 446 475 540 Total global poly capacity for solar industry 261 293 306 345 382 412 438 500 Polysilicon usage per Watt, gm/Watt 6.7 5.8 5.3 5.1 5.0 4.9 4.8 4.8 Total global poly capacity (GW) 39 51 58 68 76 84 91 104 Time-weight global poly capacity (GW) 32 45 54 63 72 80 88 98 Utilisation ratio 79% 62% 64% 68% 69% 89% 91% 85% Wafer (GW) (1) Multi-Si wafer GCL Poly 7 8 10 13 15 16 18 18 JKS 1 1 2 3 3 4 4 4 Renesola 2 2 2 2 2 3 3 3 Yingli 2 2 3 2 2 2 2 2 Others 39 41 42 40 39 38 41 41 Total multi-Si wafer capacity 51 54 59 60 62 63 68 68 Time-weight multi-Si wafer capacity 49 53 57 59 61 63 66 68 Demand for multi-Si wafer 20 23 29 36 43 56 59 52 Multi-Si's market share 76% 78% 79% 80% 82% 75% 70% 60% Utilisation ratio for multi-Si wafer 42% 43% 51% 61% 71% 90% 89% 77% (2) Mono-Si wafer LONGi 1 1 2 3 5 8 14 14 Zhonghuan Semiconductor 0 0 0 2 3 5 9 9 GCL Poly 0 0 0 0 0 1 2 2 Yinyang New Energy 0 0 0 0 1 1 2 2 Solargiga 1 1 1 1 1 1 1 1 Others 5 5 5 3 2 3 6 9 Total mono-Si wafer capacity (GW) 6 7 7 9 12 18 34 37 Time-weight multi-Si wafer capacity 6 7 7 8 10 15 24 35 Demand for mono-Si wafer 6 6 8 9 9 19 25 35 Mono-Si's market share 24% 22% 21% 20% 18% 25% 30% 40% Utilisation ratio for mono-Si wafer 103% 97% 107% 112% 91% 124% 105% 99% (3) Total global wafer capacity 57 61 66 68 74 82 101 105 Time-weight wafer capacity 52 59 64 67 71 78 92 103 Utilisation ratio 52% 49% 57% 67% 74% 97% 92% 84% Cell (GW) JA Solar 3 3 3 3 4 6 7 7 Trina Solar 2 2 3 3 4 5 5 5 Canadian Solar 2 2 2 2 3 3 5 5 Jinko Solar 1 1 2 2 3 4 5 5 Hanwha Q Cells 1 1 1 2 4 4 5 5 Others 49 56 56 61 66 77 82 83 Total cell capacity 58 65 66 72 83 98 109 110 Time-weight cell capacity 54 61 65 69 78 91 104 109 Utilisation ratio 50% 48% 56% 65% 68% 83% 81% 80% Global solar demand Total demand (industry forecasts) 27 29 36 45 53 75 84 87 - China 3 3 13 11 15 35 25 20 - US 2 3 5 6 7 15 9 10 - India 1 1 1 1 2 5 7 10 - Japan 1 2 6 9 10 7 6 6 - Europe 18 15 8 7 8 7 7 7 Note: Capacity utilisation is calculated based on industry forecasts on solar demand. Source: CS Global Solar Research Team, GTM Research, company data, Credit Suisse estimates

China Solar Sector 9 30 June 2017

Expanding market share of mono-Si ITRPV expects mono-Si demand to account for 44% of the total market by 2020. Latest research (from PV tech) suggests even stronger demand for mono-Si, representing 49% of total cell demand in 2018 and becoming the dominant technology by 2019. Based on our industry research and taking the average estimates from various consulting agencies and industry associations, we expect mono-Si demand to expand from 25% in 2016 to 30%/40%/45%/50% for FY17-20E.

Figure 14: Market demand breakdown by mono-Si and multi-Si

100% 18% 25% 30% 80% 40% 45% 50% 60% Mono-Si 40% 82% 75% 70% Multi-Si 60% 55% 20% 50%

0% 2015 2016 2017E 2018E 2019E 2020E

Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates

Balanced demand and Assuming global solar demand of 84/87/90/96GW for FY17-20E (industry forecasts from supply for mono-Si in GTM Research, etc.), the market demand for mono-Si will be 25/35/41/48GW based on next two years. our market share forecast. Our bottom-up industry research suggests weighted mono-Si wafer capacity to be 24/35/44/54GW during the period. As a result, we expect the mono-Si market to remain balanced from 2017-18E. Starting from 2019, we expect supply could be slightly higher than demand for mono-Si, but it may also change depending on global demand growth and further mono-Si market share expansion.

Figure 15: Global mono-Si demand and supply Figure 16: Mono-Si expansion supported by market forecasts share gains while capacity utilisation remains intact

(GW) Abundant (GW) Balanced 40 140% 70 supply supply and 35 120% 60 54 Supply demand 48 30 100% Mono-Si wafer time- 50 44 41 25 weighted capacity shortage 80% 40 35 35 20 60% Mono-Si wafer's 30 24 25 15 19 market share 20 16 10 40% 9 10 10 5 20% Mono-Si wafer's capacity utilization 0 0 0%

2015 2016 2017E 2018E 2019E 2020E

2013 2014 2015 2016 2011 2012

2017E 2018E Mono-Si supply Mono-Si demand

Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates

LONGi Green Energy Technology (LONGi) is the largest mono-Si wafer manufacturer with ~40% global market share. The second-largest company is Zhonghuan Semiconductor (China) with ~25% global market share. As a result, the largest two players have a dominant combined market share of ~65%, which also helps maintain pricing power.

China Solar Sector 10 30 June 2017

Multi-Si wafer may suffer from market share loss and be forced to reduce capacity The expanding mono-Si market share will come at the expense of multi-Si which currently is the dominant technology. Assuming no industry consolidation, we expect the capacity utilisation of multi-Si wafers may drop to 77% in 2018E (from 86% in 2017E), as a result of both market share loss (contributing ~80% of the utilisation drop) and planned new multi- Si wafer capacity additions from lower-tier manufacturers. High-cost multi-Si Unlike mono-Si, the multi-Si wafer market is quite fragmented, with the largest producer, manufacturers may be GCL Poly, taking ~25% market share while the second and third players have <5% market forced to shut down share each. Looking at the entire wafer industry, we expect multi-Si wafers may come under pressure and high-cost multi-Si wafer producers may have to be shut down in the next few years. At the same time, the effective multi-Si wafer capacity additions (from lower-tier manufacturers) may also be milder than expected, given the uncertain demand outlook for multi-Si. All of these could help bring back a supply-demand balance for the whole wafer industry.

Figure 17: Multi-Si may suffer from market share loss and a capacity utilisation drop (GW) 80 100% Multi-Si wafer time- 80% 60 weighted capacity 60% 40 Multi-Si wafer's 40% market share 20 20% Multi-Si wafer's 0 0%

capacity utilization

2013 2011 2012 2014 2015 2016

2017E 2018E

Source: Company data, Credit Suisse estimates Earnings growth for mono-Si wafer producers is warranted by increasing sales volume Even if the mono-Si wafer price is dragged down by lower multi-Si with a deteriorating oversupply, we expect the earnings growth for mono-Si wafer producers is still warranted by strong sales volume growth and higher cost efficiency. In the worst-case scenario, we assume only a 5% multi-Si gross profit margin for GCL Poly (vs. 11% in our base-case assumption). The implied unit profit for GCL would drop sharply from Rmb0.24/W in 2016 to only Rmb0.04/W in 2017E. As GCL is a cost leader in the multi-Si wafer industry, such a GP margin would probably mean industry-wide losses for its peers. The implied FY17-19E net profit for GCL would be significantly lower than Bloomberg consensus by 39-51%. Due to its higher efficiency, mono-Si wafers are justified in being priced ~10% higher than for multi-Si. We estimate the implied ASP downside for LONGi (vs our base-case assumption) would be 8-10% for FY17-19E. However, driven by strong volume growth (28-42% per year in FY17-19E), we estimate that worst-case FY17E EPS may only drop 2% YoY for LONGi while the FY16-19E EPS CAGR would remain at a decent 16% (vs. 24% in the base case). The implied fair value of LONGi also indicates limited downside versus the current share price. Even if we assume a zero GP margin for GCL, LONGi may still experience a 11% FY16-19E EPS CAGR.

China Solar Sector 11 30 June 2017

Figure 18: LONGi—stress test on mono-Si ASP (Rmb/Watt) 2016A 2017E 2018E 2019E GCL's multi-Si wafer production cost 0.85 0.81 0.73 0.68 (1) Assuming 5% GPM for GCL: Multi-Si wafer ASP 1.09 0.85 0.77 0.72 - Implied multi-Si wafer unit profit 0.24 0.04 0.04 0.03 - Implied GCL's net profit vs. Bloomberg consensus -51% -49% -39% Justified mono-Si wafer ASP (10% premium to multi-Si) 0.93 0.85 0.79 LONGi's mono-Si wafer ASP assumptions (CSe) 1.02 0.93 0.86 ASP downside (mono-Si wafer) -8% -10% -8% LONGi's output growth 39% 42% 28% LONGi's net profit (YoY) -2% 28% 24% LONGi's 2016-19E EPS CAGR 16% LONGi's fair value (Rmb) 16.7 Current share price 17.0 Downside -2% (2) Assuming 0% GPM for GCL Multi-Si wafer ASP 1.09 0.81 0.73 0.68 - Implied multi-Si wafer unit profit 0.24 0.00 0.00 0.00 - Implied GCL's net profit vs. consensus -79% -74% -61% Justified mono-Si wafer ASP (10% premium to multi-Si) 0.89 0.81 0.75 LONGi's mono-Si wafer ASP assumptions (CSe) 1.02 0.93 0.86 ASP downside (mono-Si wafer) -13% -14% -13% LONGi's output growth 39% 42% 28% LONGi's net profit (YoY) -15% 30% 25% LONGi's 2016-19E EPS CAGR 11% LONGi's fair value (Rmb) 14.9 Current share price 17.0 Downside -12% Source: Company data, Credit Suisse estimates Mono-Si is preferred by solar farm operators Extra return by using mono-Si Comparing solar farm According to EnergyTrend (a news agency focusing on solar industry), the cell efficiency IRRs, mono-Si is 100 difference between multi-Si (18.2%) and mono-Si (19.7%) is about 150 bp, or ~8% in terms bp higher than multi-Si of power output within the same area. With continuous cost reduction, the mono-Si module price premium to multi-Si can now be reduced to Rmb0.1/Watt (W). Due to this higher efficiency, the unit Balance of System (BOS) cost could be ~10% lower for mono-Si. As a result, we calculate that the per-W investment cost for a ground-mounted solar farm may be Rmb0.2/W lower by using mono-Si, which also yields higher project returns (equity IRR: mono-Si at 11.0% vs. multi-Si at 10.0%).

China Solar Sector 12 30 June 2017

Figure 19: Ground-mounted solar farm IRR comparison—mono-Si vs multi-Si Multi-Si Mono-Si Wafer area (sq m3) 243 244 Cell efficiency 18.2% 19.7% Cell power output (W/pc) 4.4 4.8 Cell-to-Module loss (%) 0.5 3.0 60-cell module power output (W) 260 280 Module price (Rmb/W) 3.0 3.1 BOS cost (Rmb/W) 3.0 2.7 All-in investment cost (Rmb/W) 6.0 5.8 Equity IRR for a ground-mounted solar farm 10.0% 11.0% Source: Company data, Credit Suisse estimates Larger advantage in distributed solar Mono-Si has 200 bp Due to land limits and the higher BOS cost of distributed solar projects, mono-Si has even higher IRR in more of an advantage in this area. We estimate a 200 bp solar farm IRR advantage by distributed solar using mono-Si comparing to multi-Si. project than multi-Si Figure 20: Distributed solar IRR comparison—mono-Si vs multi-Si Multi-Si Mono-Si Land usage per kW (sq m) 10.0 9.2 Cell efficiency 18.2% 19.7% Land area (sq m) 100,000 100,000 Designed capacity (MW) 10.0 10.8 Roof-top rental fee (Rmb/sq m) 4.0 4.0 Total annual rental cost (Rmb mn) 0.4 0.4 Module price (Rmb/W) 3.0 3.1 BOS cost (Rmb/W) 3.5 3.2 All-in investment cost (Rmb/W) 6.5 6.3 Total investment cost (Rmb mn) 64.7 68.6 Equity IRR for a distributed solar project 15.0% 17.0% Source: Company data, Credit Suisse estimates

To illustrate, a 100,000 sq m rooftop can install around a 10MW multi-Si module or 10.8MW mono-Si module. The annual rental cost is around Rmb0.4 mn (Rmb4.0 per sq m, usually subject to adjustment every five years), the same for both mono-Si and multi-Si. Total BOS cost should also be similar, so that mono-Si would be cheaper in terms of per W cost. As a result, we estimate an all-in unit investment cost of Rmb6.5/W for multi-Si and Rmb6.3/W for mono-Si. The total investment cost would be slightly higher for mono-Si as more capacity is installed (10.0MW vs. 10.8MW). With a lower cost per W, mono-Si can achieve a 17% equity IRR for a typical mono-Si project while multi-Si can achieve 15% (a 200 bp difference, larger than ground-mounted solar). Apart from China, mono-Si is also starting to gain popularity in the global rooftop market. For example, Tesla's solar tiles are also embedded with mono-Si cells (125mm*125mm), according to PV tech. Tesla has developed techniques to make its solar tiles identical to ordinary tiles (solar cells are invisible).

Cost efficiency is the key We believe that mono-Si should further gain market share from multi-Si in the next few years due to its cost and efficiency advantages. In terms of efficiency, as mentioned above, we estimate that solar farms using a mono-Si module can enjoy a higher equity IRR as long as the price premium of mono-Si over multi-Si is within Rmb0.3/W. Such a price gap was achieved in 2H16 (as low as Rmb0.1/W) and then widened again recently (to Rmb0.3/W) driven by a near-term mono-Si supply shortage. We expect mono-Si module's price premium to maintain at Rmb0.1-0.3/W in order to be competitive as well as maintain a decent profitability.

China Solar Sector 13 30 June 2017

Figure 21: Module price history—mono-Si vs multi-Si (Rmb/W) (Rmb/W) 6.0 1.2 5.0 0.9 4.0 3.0 0.6 2.0 0.3 1.0

0.0 0.0

Jul-14 Jul-15 Jul-16

Jan-14 Jan-15 Jan-16 Jan-17

Sep-14 Sep-15 Sep-16

Nov-14 Nov-15 Nov-16

Mar-14 Mar-15 Mar-16 Mar-17

May-14 May-15 May-16 Price difference Multi-Si module Mono-Si module

Source: WIND

More importantly, in terms of unit cost, the mono-Si module is already Rmb0.2/W lower than multi-Si, based on top producers. Mono-Si used to more expensive at the wafer level due to the higher cost at the ingot pulling process. After years of technology improvement for mono-Si, there is now no significant difference at the non-poly wafer cost (both at ~Rmb0.35/W). However, at the cell and module production phases, mono-Si has a clear advantage due to its higher efficiency. As a result, the unit production cost of mono-Si is Rmb0.2/W cheaper than for multi-Si module.

Figure 22: All-in cost comparison between mono-Si and multi-Si Mono-Si Multi-Si (1) Poly Cost (Rmb/kg) 120 120 (2) Wafer 2.1) Poly cost 0.48 0.54 Poly consumption (g/W) 4.0 4.5 2.2) Non-poly cost 0.35 0.35 Total wafer cost (Rmb/W) 0.83 0.89 (3) Cell Cell processing cost (Rmb/W) 0.30 0.35 Total cell cost (Rmb/W) 1.13 1.24 (4) Module Module processing cost (Rmb/W) 0.9 1.0 Total module cost (Rmb/W) 2.0 2.2 Source: Company data, Credit Suisse estimates

With both a Rmb0.3/W advantage in efficiency and Rmb0.2/W advantage in cost, we expect the profitability of the mono-Si value chain to be much more resilient than for multi- Si. We estimate currently the total unit profit per W is around Rmb0.9/W for mono-Si and Rmb0.4/W for multi-Si. Mono-Si may still maintain a Rmb0.5/W profit even if we assume zero profit for multi-Si. New technology (PERC) also favours mono-Si Mono-Si has more Higher efficiency and lower cost are the two main drivers for solar power grid parity. As a efficiency gain from result, new high-efficiency cells (usually with a 50-100 bp higher cell efficiency) are gaining PERC than multi-Si popularity globally. Passivated Emitter Rear Cell (PERC) technology is currently one of the most promising technologies in the high-efficiency cell market. Industry consultants (such as TrendForce) estimate that PERC cell production capacity will grow to 61GW by 2020 (up from only 15GW in 2016) and represent 44% of global PV cell capacity.

China Solar Sector 14 30 June 2017

Figure 23: PERC technology's market share in global solar cell manufacturing

100%

80% 77GW 91GW 86GW 60% 84GW 92GW Non-PERC 40% PERC 20% 61GW 36GW 47GW 15GW 25GW 0% 2016 2017E 2018E 2019E 2020E

Source: TrendForce

Figure 24: Comparison between PERC cell and conventional cell Conventional Cell PERC Cell Light Light Emitter layer Emitter layer

Base layer Base layer (silicon wafer) (silicon wafer)

Dielectric Aluminum metallization Aluminum metallization Passive Layer Light is absorbed by the Reflected light will generate aluminum metallization. additional current

Source: Renewable Energy Corporation, SinoVoltaics, CleanTechnica, Credit Suisse research

Although both mono-Si and multi-Si cells will have higher efficiency after using PERC technology, the efficiency gain is more significant for mono-Si. According to data from Energy Trend, PERC technology could enhance mono-Si cell efficiency by close to 100 bp, while that for multi-Si is only ~50 bp, both with similar additional costs. International Technology Roadmap for Photovoltaic (ITRPV), a solar research arm under Semiconductor Equipment and Materials International Association, also forecasts the gap to maintain, as PERC may increase mono-Si cell efficiency by ~150 bp by 2025E and ~110 bp for multi-Si.

Figure 25: Cell efficiency gain from PERC—multi-S vs mono-Si (bp) 160 140 120 100 80 Multi-Si 60 Mono-Si 40 20 0 2016 2017E 2018E 2020E 2022E 2025E

Source: ITRPV

China Solar Sector 15 30 June 2017

LONGi is the leader and pioneer in mono-Si LONGi started mono-Si wafer production in 2004, one of the earliest among Chinese peers. The development of mono-Si solar products was not smooth in the early years and mono-Si's market share has been under pressure due to the quick expansion of multi-Si products, with larger-scale production lines and lower costs. Given the technology uncertainty, many module manufacturers were reluctant to build new production lines for mono-Si products. One key milestone for LONGi was the acquisition of Zhejiang Lerri Solar (a module manufacturer) in October 2014. With technology knowhow along the whole solar value chain, LONGi started to produce its own mono-Si modules and build its reputation. Its module price is also highly competitive—only Rmb0.1/W higher than multi- Si—and therefore preferred by more downstream developers. In 2016, LONGi accounted for 27% of China's mono-Si module shipments, the highest among peers, according to EnergyTrend. Its gross margin has also been one of the highest among peers, helped by strong cost control and good product mix.

Figure 26: China mono-Si module shipment rankings in 2016

Lerri (LONGi) 27% JA Solar 40% Trina 7.8GW Linyang 12% SolarGiga

5% Others 8% 8%

Source: EnergyTrend

Figure 27: Gross margin comparison of mono-Si manufacturers Gross margin (%) 2012 2013 2014 2015 2016 LONGi 13 12 17 20 27 Zhonghuan Semiconductor (002129.SZ) 10 12 15 15 14 Solargiga (0757.HK) -29 7 13 8 11 Comtec (0712.HK) 8 8 7 -9 -19 Source: Company data, Bloomberg Black silicon and diamond wiring may be the silver lining for multi-Si One of the key drivers for mono-Si cost-cutting has been the adoption of diamond-wire wafer cutting technology in the past 2-3 years. Compared with traditional slurry-wire cutting (currently the most widely used in multi-Si), diamond-wire cutting can reduce the wafer production cost by US$6-10 cents/piece (or US$1.5-2.5 cents/W) with diamond wiring, according to EnergryTrend. ITRPV expects that diamond wiring may become the dominant ingot slicing technology by 2018 (>70% market share, vs ~35% in 2015). Without further treatment, the multi-Si wafer surface cut by the diamond wiring method would be too smooth, so that much of the sunlight would be reflected (instead of absorbed), causing lower conversion efficiency. As a result, a new technology called black silicon surface etching is now used to solve this issue.

China Solar Sector 16 30 June 2017

In our base-case analysis, we expect diamond-wiring to reduce multi-Si wafer prices by Rmb0.1/W, while black silicon treatment would incur an extra Rmb0.05/W cost, so that the net cost saving may be around Rmb0.05/W. If the reduced cost is fully passed through to the end-user, it would help reduce the solar farm IRR premium of using mono-Si from 100 bp to ~50 bp. We still expect mono-Si to be more cost efficient in the next few years, while better-than-expected technology improvements in multi-Si would be a risk.

Figure 29: Percentage of wafer production using Figure 28: Wafer cost efficiency comparison diamond-wiring technology

(Rmb/W) The price advantage of multi-Si wafer may expand by 100% 1.8 Rmb0.05/W after diamond wiring and black silicon treatment, and thus reduce the solar farm IRR gap. 80% 1.7 Rmb Rmb 60% 0.1/W 1.6 0.15/W Mono-Si wafer 40% Multi-Si wafer 1.5 IRR gap: +0.05 -0.1 IRR gap: 100bp 20% ~50bp 1.4 Spot price with diamond wire with black silicon 0% 2016 2017E 2018E 2020E 2022E 2025E Mono-Si wafer price Multi-Si wafer price

Source: Credit Suisse estimates Source: Company data, Credit Suisse estimates ASP/cost/margin comparison of leading mono-Si/ multi-Si manufacturers (LONGi and GCL) We expect an 8% annual average sales price (ASP) drop for mono-Si wafers, largely offset by continuous cost-cutting with the gross margin stable at 32-34%. We expect mono-Si wafers to be the main earnings driver for LONGi, accounting for >90% of its total gross profit (by process) after FY18E. Due to less competitive efficiency, we expect multi- Si wafer to see a larger price drop in the next few years (-10% CAGR for FY17-20E), which would be the main drag of GCL's operating performance.

China Solar Sector 17 30 June 2017

Figure 30: LONGi—key operating metrics 2013 2014 2015 2016 2017E 2018E 2019E 2020E Mono-Si wafer Capacity year-end (MW) 1,600 3,000 5,000 7,500 14,200 14,200 24,200 24,200 Total sales (MW) 1,174 1,971 2,662 7,204 10,000 14,200 18,200 24,200 - YoY 68% 35% 171% 39% 42% 28% 33% (1) External sales (MW) 1,174 1,963 1,929 4,696 5,500 7,200 9,200 13,700 - YoY 67% -2% 143% 17% 31% 28% 49% (2) Internal sales to module segment (MW) 0 9 733 2,508 4,500 7,000 9,000 10,500 Wafer ASP ex. VAT (Rmb/W) 1.58 1.61 1.33 1.08 1.02 0.93 0.86 0.79 - YoY 2% -18% -18% -6% -8% -8% -8% Wafer revenue (Rmb mn) 1,861 3,159 2,557 5,075 5,587 6,729 7,911 10,837 - YoY 70% -19% 98% 10% 20% 18% 37% Wafer unit cost (Rmb/W) 1.39 1.33 1.04 0.78 0.69 0.63 0.57 0.52 - YoY -5% -22% -25% -12% -9% -9% -9% GP margin 12% 17% 22% 28% 32% 33% 34% 34% Unit gross profit (Rmb/W) 0.19 0.28 0.29 0.30 0.33 0.31 0.29 0.27 Gross profit—external sales (Rmb mn) 223 552 551 1,429 1,810 2,217 2,651 3,695 - YoY 148% 0% 160% 27% 22% 20% 39% Mono-Si module Capacity year-end (MW) 1,500 5,000 6,000 8,000 10,000 11,000 Total sales (MW) 760 2,129 4,500 7,000 9,000 10,500 - YoY 180% 111% 56% 29% 17% (1) External sales (MW) 721 1,847 4,400 6,900 8,900 10,400 - YoY 156% 138% 57% 29% 17% (2) Internal sales to solar farm segment (MW) 39 281 100 100 100 100 Module ASP (Rmb/W) 3.49 3.09 2.50 2.27 2.09 1.93 - YoY -12% -19% -9% -8% -8% Unit cost (Rmb/W) 2.82 2.25 2.05 1.90 1.75 1.62 - YoY -20% -9% -8% -8% -8% GP margin 19% 27% 18% 17% 16% 16% Unit gross profit (Rmb/W) 0.67 0.84 0.45 0.38 0.34 0.30 Gross profit - external sales (Rmb mn) 487 1,551 1,960 2,597 3,012 3,163 Gross profit breakdown (1) By external sales (Rmb mn) Wafer 551 1,429 1,810 2,217 2,651 3,695 Module 487 1,551 1,960 2,597 3,012 3,163 Wafer (%) 53% 48% 48% 46% 47% 54% Module (%) 47% 52% 52% 54% 53% 46% (2) By process (Rmb mn) Wafer 756 1,991 3,257 4,341 5,215 6,500 Module 281 988 513 473 448 358 Wafer (%) 73% 67% 86% 90% 92% 95% Module (%) 27% 33% 14% 10% 8% 5% Source: Company data, Credit Suisse estimates

China Solar Sector 18 30 June 2017

Figure 31: GCL—key operating metrics 2014 2015 2016 2017E 2018E 2019E 2020E Polysilicon Year-end capacity (tonnes) 65,000 70,000 70,000 75,000 115,000 115,000 115,000 Output (t) 66,876 74,358 69,345 75,038 87,000 115,000 115,000 External sales (t) 15,443 18,023 9,951 9,951 18,089 26,125 23,875 Internal used (t) 51,433 56,335 59,394 65,087 68,912 88,875 91,125 ASP (US$/kg) 21.6 15.6 15.0 14.3 13.1 12.1 11.5 - YoY 25% -28% -4% -5% -8% -8% -5% Unit cost (US$/kg) 15.5 13.5 11.5 10.5 9.7 8.8 8.5 - YoY -9% -13% -15% -9% -7% -9% -4% Gross margin 28% 13% 23% 26% 26% 27% 26% Unit margin (US$/kg) 6.1 2.1 3.5 3.8 3.4 3.2 2.9 Wafer Year-end capacity (MW) 13,000 15,000 17,000 19,000 19,500 20,000 20,500 - Mono wafer (MW) 0 0 1,000 1,500 2,000 2,500 3,000 - Multi wafer (MW) 13,000 15,000 16,000 17,500 17,500 17,500 17,500 Output (MW) 12,909 15,178 17,518 18,000 19,250 19,750 20,250 ASP (US$/W) 0.215 0.188 0.164 0.133 0.116 0.103 0.096 - YoY 2% -13% -13% -19% -13% -11% -7% Wafer cost (US$/W) 0.175 0.150 0.129 0.118 0.107 0.099 0.093 - YoY -8% -14% -14% -8% -9% -8% -6% Gross margin 19% 20% 22% 11% 8% 4% 3% Unit profit (US$/W) 0.040 0.038 0.035 0.015 0.009 0.004 0.003 Gross profit breakdown by process (Rmb mn) Poly 2,533 967 1,611 1,927 2,022 2,583 2,350 Wafer 3,103 3,394 3,993 1,895 1,220 720 534 Poly (%) 45% 22% 29% 50% 62% 78% 81% Wafer (%) 55% 78% 71% 50% 38% 22% 19% Source: Company data, Credit Suisse estimates

China Solar Sector 19 30 June 2017

Rising distributed solar in China Distributed solar could Despite a likely growth slowdown in the China solar power sector, distributed solar, see 55% capacity backed by strong policy support, could take off with a likely 55% CAGR from 2016-20 (a CAGR in 2016-20 60GW target by 2020). The key advantages of distributed solar include flexible locations (largely rooftops), limited curtailment issues and generous subsidies. During the few rounds of solar tariff cut in the past several years, distributed tariffs were kept intact. We calculate equity IRR at >15%, higher than ~10% for ground-mounted solar farms.

Figure 32: China—accumulated solar capacity Figure 33: China—annual solar capacity additions (GW) (GW) 35 150 4 54 30 40 28 25 100 18 20 8 10 15 1 30 10 1 14 50 99 104 10 12 84 94 2 17 6 67 14 5 5 2 12 10 3 37 9 0 1 2 16 23 2 5 5 0 0 2 4 0 2010 2011 2012 2013 2014 2015 2016 2017E2018E2019E2020E 2012 2013 2014 2015 2016 2017E 2018E 2019E 2020E Ground-mounted Distributed Ground-mounted Distributed Source: NEA, Credit Suisse research Source: NEA, Credit Suisse research Superior project return China introduced the tariff subsidy (Rmb0.42/kWh) for distributed solar in August 2013. Although system costs have been declining over the years, China has kept the subsidy unchanged. In September 2016, there was a proposal from National Development and Reform Commission (NDRC) to cut the distributed solar subsidy to Rmb0.2-0.3/kWh, but it was eventually rejected in the final tariff decision announced by the NDRC on 26 December 2016, in order to better support the growth of distributed solar industry.

Figure 34: Historical tariff/subsidy changes for ground-mounted and distributed solar (Rmb/kWh) 2011 2012-13 2014 to mid-2016 Mid-2016 to mid-2017 After mid-2017 Ground-mounted (total tariff)* 1.15 1.00 0.95 0.88 0.70 Distributed solar (tariff subsidy) n.a. n.a. 0.42 0.42 0.42 Note: * Using Zone II as an example. Source: NDRC

Strong policy support In a typical distributed solar power generation project, the developer takes roof-tops on for distributed solar rent from the landlord (such as a factory, hospital, shopping mall, etc.) and installs the solar panels. Most of the electricity generated is sold to the landlord, usually at a 10-15% discount to the local retail electricity price (Rmb0.7-0.8/kWh).The rest is sold to grid companies at local coal-fired on-grid tariffs (Rmb0.3-0.4/kWh). For the whole amount of electricity generated, the project developer is eligible to receive national renewables subsidy at Rmb0.42/kWh for 20 years, which supports an equity IRR of 12% based on our calculation. Besides, there are also many provinces/cities that have local subsidies at roughly Rmb0.1-0.2/kWh (usually for the first three to five years of operation), which further enhances equity IRR to >15%. This is much better than the ground-mounted solar project with only ~10% IRR.

China Solar Sector 20 30 June 2017

Figure 35: Distributed solar: IRR calculation Key assumptions—distributed solar Capacity (MW) 10 Average tariff (Rmb/kWh): 1.02 Unit capex (Rmb/W) 6.5 % of output sold to end-user 80% - Module 3.0 % of output sold to power grid 20% - Others 3.5 (1) Total tariff sold to roof-top owner: 1.07 Total investment (Rmb mn) 65 Local C&I retail price (Rmb/kWh) 0.77 VAT recoverable (Rmb mn) 6.6 Discount 15% Realised electricity price (Rmb/kWh) 0.65 Leverage 70% National solar subsidy 0.42 Equity investment (Rmb mn) 19.5 (2) Total tariff sold to the grids: 0.78 Total borrowings (Rmb mn) 45.5 On-grid coal-fired tariff 0.36 Interest rate 6.0% National solar subsidy 0.42

Utilisation hours 1,000 Local government subsidy (assuming 5 years, Rmb/kWh) 0.10 Degradation 1% Tax rate 25% Roof-top annual rental cost (per W) 0.04 IRR calculation Project IRR 10% Equity IRR 15% Source: Credit Suisse estimates

Figure 36: Ground-mounted solar—IRR calculation Key assumptions – ground-mounted solar Capacity (MW) 100 Leverage 70% Unit capex (Rmb/W) 6.0 Total borrowings (Rmb mn) 420 1) Module 3.0 Benchmark interest rate 4.9% 2) BOS 3.0 Mark-up 22% - inverter 0.2 Interest rate 6.0% - cable 0.3 Tariff (Rmb/kWh) 0.68 - other equipment 0.8 - on-grid tariff 0.36 - bracket 0.5 - renewables subsidies 0.32 - civil work 1.2 Discount to benchmark tariff due to bidding -10% Total investment (Rmb mn) 600 Utilisation hours 1,300 Capex in PP&E (Rmb mn) 517 Degradation per year 1% Tax receivable (Rmb mn) 83 Income tax rate 25% IRR calculation Project IRR 8% Equity IRR 10% Note: Using Zone II project as an example. Source: Credit Suisse estimates

Based on our industry research, at least 15 provinces in China have announced provincial distributed solar subsidies on top of the Rmb0.42/kWh national subsidy, in the form of (1) additional tariff subsidies at Rmb0.1-0.2/kWh usually covering the first 5-10 years of an operating project; or (2) upfront investment cost subsidy at Rmb1-3/W. Furthermore, there are also many city level subsidies, usually at Rmb0.1-0.3/kWh for the first three to five years, in addition to the national and provincial subsidies.

China Solar Sector 21 30 June 2017

Figure 37: Impact of local government subsidies on equity IRRs of a distributed solar project Additional provincial/ municipal subsidy Equity IRR No additional provincial subsidy 12% Rmb0.1/kWh for 5 years (base case) 15% Rmb0.2/kWh for 5 years 18% Rmb0.3/kWh for 5 years 21% Source: Credit Suisse estimates

Figure 38: Provincial and city/county level subsidies on distributed solar projects Province Distributed solar Provincial subsidy Subsidising Note capacity in 2016 (MW) (Rmb/kWh) period Beijing 190 0.3 5 years Hebei 390 0.2 3 years Shanxi 130 Yes Rmb0.2/kWh plus Rmb3/W for investment cost Jilin 50 0.15 Shanghai 330 0.25/0.4 Jiangsu 1,730 Yes Rmb0.1/kWh for 3 years; Rmb0.2-2/W for investment cost Zhejiang 2,070 0.1 20 years Rmb0.1-0.3/kWh for first 3-5 years of operation Anhui 780 0.25 15 years Rmb2-4/W for investment cost Jiangxi 570 0.2 20 years Rmb3-4/W for investment (provincial); Rmb0.15/W for 5 years (Nanchang) Shandong 1,190 0.05 Henan 360 Yes Rmb0.1/W for investment cost for 3 years Hubei 200 Yes Rmb0.1/kWh for 10 years Hunan 300 0.2 10 years Rmb0.1/kWh for 5 years Guangdong 880 0.1-0.2 10 years Rmb0.02-1/W for investment cost Shaanxi 120 Rmb1/W Initial investment Rmb1/W; or Rmb0.006-0.05/kWh Source: Provincial Development and Reform Commissions (such as Zhejiang, Jiangsu, etc.), PV Trade.

Besides, unlike ground-mounted projects, the national and local subsidies for distributed solar usually have no delay issues, which alleviates cash flow pressure for project owners. Accelerating capacity additions in 13th FYP Despite slower-than-expected development in the past few years (only 1-2GW annual additions before 2015), China has accelerated distributed solar installations since 2017. In 1Q17, China added 2.4GW distributed solar projects, up strongly by 151% YoY, much better than ground-mounted solar which recorded a 23% YoY drop during the same period. We believe this was mainly driven by distributed solar's superior project return, as well as strong central and local government policy support. Unlike the uncertainty for ground-mounted capacity target, distributed solar has a much more clear and aggressive target of >60GW by 2020. Based on the current run-rate (2.4GW in 1Q17), we expect at least 8GW distributed solar capacity addition in 2017, and annual additions should gradually pick up in the next few years in order to achieve NDRC's capacity target by 2020. In our base-case scenario, we expect China to have an accumulated distributed solar capacity of 54GW by 2020 (slightly lower than NDRC's target of 60GW), accounting for 34% of total capacity. With 51% 2016-20E capacity CAGR based on our forecasts, distributed solar is likely to be one of the very few types of power that can still witness high growth, given the overall power oversupply in China. Ground- mounted solar, on the other hand, is likely to slow down going forward compared with its peak installations in 2016-17.

China Solar Sector 22 30 June 2017

Figure 39: China solar capacity forecasts: ground-mounted vs. distributed (2017-20E) (MW) Accumulated capacity Annual capacity additions Ground-mounted Distributed Total Ground-mounted Distributed Total 2009 100 180 28 n.a. n.a. n.a. 2010 420 440 860 320 260 832 2011 2,320 600 3,500 1,900 160 2,640 2012 4,198 2,300 6,498 1,878 1,700 2,998 2013 16,320 3,100 19,420 12,120 800 12,920 2014 23,380 4,670 28,050 8,550 2,050 10,600 2015 37,120 6,060 43,180 13,740 1,390 15,130 2016 67,100 10,320 77,420 30,310 4,240 34,540 2017E 84,100 18,320 102,420 17,000 8,000 25,000 2018E 94,100 28,320 122,420 10,000 10,000 20,000 2019E 99,100 40,320 139,420 5,000 12,000 17,000 2020E 104,100 54,320 158,420 5,000 14,000 19,000 Source: NEA, Credit Suisse estimates

Little power curtailment and no subsidy delay Distributed solar has One of the key reasons that some provinces in China suffer from serious renewable power fewer issues curtailment issues is due to the location mismatch of power supply and demand. For comparing with example, provinces such as Gansu and Xinjiang may not be able to solve the power ground-mounted solar curtailment issue (22-30%) even by 2020E, based on estimates from Northwest China Energy Regulatory Bureau of NEA, due to weak local power demand and inadequate outbound power transmission lines. Distributed solar, on the other hand, would not have any such issues. They have flexible locations (usually built on roof-tops of factories, hospitals, etc.) and closer to electricity end-users. Most of the power generated is sold directly to the roof-top owners (such as factories) and hence there will be no power curtailment. Unlikely ground-mounted projects, there is usually no delay for roof-top distributed solar projects. Otherwise, project IRR and equity IRR may be 100 bp lower based on our estimates. Besides, the additional local government subsidies for distributed solar are also unusually paid on time. As a result, we estimate that a distributed solar project should be able to record positive free cash flow to equity (FCFE) since the first year of operation, much better than ground-mounted projects which usually suffer two to three years of negative FCFE due to subsidy delays.

Figure 40: IRR analysis if there is subsidy delay Project IRR Equity IRR Without subsidy delay 10% 15% With subsidy delay 9% 14% Source: Credit Suisse estimates

China Solar Sector 23 30 June 2017

Figure 41: Projected free cash flow to equity (FCFE) trend for distributed solar and ground-mounted solar projects (Rmbm) 100 50 0 Year 0 Year 1 Year 2 Year 3 Year 4 Year 5 -50 -100 -150 -200 Distributed solar Ground-mounted solar

Source: Credit Suisse estimates Stock implications For most of the listed solar farm operators under our coverage, the shift of focus from ground-mounted projects to distributed solar may not be good news, at least in the near term. Both Xinyi Solar (XYS) and GCL New Energy (GCLNE) had limited exposure (3-5%) to distributed solar in 2016, and Concord New Energy (CNE) had no exposure (while CNE's current main focus is on wind farm). Linyang is the only However, Linyang Energy (Linyang) as a local distributed solar developer (in Jiangsu, stock under our Shandong, etc.) may benefit from expanding distributed solar opportunities (most are likely coverage that has to be built in coastal regions where there is strong power demand). By 2016, Linyang had meaningful exposure to a total installation solar farm capacity of 931MW, of which 85% is distributed solar. In distributed solar 2017, the company plans to add 500MW solar capacity (all distributed solar), of which 70% may be roof-top distributed solar projects mainly in coastal regions.

Figure 42: Listcos' capacity exposure to distributed solar (2016)

100% 15% 80%

60% 97% 95% 100% Ground-mounted solar 40% 85% Distibuted solar 20%

0% 3% 5% 0% GCLNE XYS Linyang CNE

Source: Company data

Given strong government support and superior project returns, we expect distributed solar to account for 32% of China's annual solar capacity addition in 2017 and 50% in 2018, much higher than the mere 12% in 2016. In the long run, we do not rule out the possibility of more traditional solar farm operators expanding into this area. For example, GCLNE is planning to add 100-200MW distributed solar in 2017, accounting for 5-10% of its total addition. We believe XYS may find it more challenging to invest in distributed solar given its more concentrated geographical exposure (most are provincial ground-mounted projects in Anhui Province) in central regions.

China Solar Sector 24 30 June 2017

Figure 43: 2017E solar capacity addition breakdown for listcos (MW) GCLNE XYS Linyang CNE 2017E solar capacity addition target 2,000 600-800 500 50 - Distributed solar 100-200 0 500 0 - Ground-mounted solar 1,800 700 0 50 (1) Provincial quota 1,200 700 0 50 (2) Top Runner 360 0 0 0 (3) Poverty alleviation 285 0 0 0 Source: Company data, Credit Suisse estimates

Figure 44: Geographical exposure of listed solar farm operators under our coverage

Heilongjiang

Jilin

Liaoning Xinjiang Inner Mongolia Beijing Gansu Tianjin Hebei Ningxia Shanxi Shandong

Qinghai Shaanxi Henan Jiangsu

Tibet Shanghai Hubei Anhui Sichuan Jiangxi Zhejiang Hunan Fujian GCL New Energy Guizhou

Xinyi Solar Yunan Guangdong Guangxi Zone I: Rmb0.65/kWh Concord Zone II: Rmb0.75/kWh

Linyang Energy Zone III: Rmb0.85/kWh Hainan Tibet: Rmb1.05/kWh

Source: Company data, Credit Suisse estimates

For solar component manufacturers China constitutes a major portion of global demand and such a shift from ground-mounted to distributed solar could help drive demand for high efficiency products, such as mono-silicon (mono-Si). This is because distributed solar usually has limited land space (such as roof-tops, etc.) so that high efficiency modules would be more economically attractive for those distributed solar projects despite their higher cost. LONGi is a major player in the mono-Si space.

China Solar Sector 25 30 June 2017

Headwinds for ground-mounted projects We turn more cautious on ground-mounted solar farm operators due to declining annual project quotas, continuing tariff cut risks (next round could take place in 2H17 with likely ~15-20% cut), and expanding competitive projects bidding (>80% in 2017). Rising interest rate is an additional risk. With declining IRRs and uncertain project quotas, we expect ground-mounted solar farms to witness a 12% CAGR during 2016-20. Uncertain quota outlook in 13th FYP In the past decade, ground-mounted projects have been dominating China's solar market, helped by strong policy support (highest subsidy among all utilities) and continuously declining investment cost. In 2016 alone, China added 30GW ground-mounted solar farms, much ahead of market expectations and NDRC's plans (21GW in total including provincial, Top Runner and poverty-alleviation quotas). Ground-mounted solar According to China's solar industry 13th FYP, the NEA targets to build more than 110GW is not well supported solar capacity by 2020, of which distributed energy should be >60GW and solar thermal by 13th FYP should be >5GW. The implied capacity target for ground-mounted projects is only >45GW by 2020. However, China has already reached this "minimum" target with 67GW accumulated ground-mounted capacity in 2016. As a result, unlike distributed solar, there is no capacity target support for ground-mounted solar, leaving its growth in uncertainty.

Figure 45: Uncertain 13th FYP capacity outlook for ground-mounted solar (GW) ? 80 67 70 60 60 >45 50 2016 40 30 2020 target 20 10 10 0 Distributed Ground-mounted

Source: Company data, Credit Suisse estimates Next round of tariff cut discussion in 2H17 In the last solar tariff adjustment document issued in late 2016, the NDRC has made it clear that solar tariff for future projects will be reviewed every year based on investment cost changes. The last round of tariff cut discussion started in September 2016 with the final decision announced in December 2016. A similar time table could apply for 2017.

Figure 46: Benchmark solar tariff cut history and forecasts Tariff decisions 2014 to mid-2016 Mid-2016 to mid-2017 Mid-2017 to mid-2018 After mid-2018 (Rmb/kWh) Tariff Implied Tariff Implied Tariff Implied Tariff Implied IRR IRR IRR IRR Zone I 0.90 13% 0.80 17% 0.65 14% 0.52 10% Zone II 0.95 10% 0.88 15% 0.75 14% 0.62 10% Zone III 1.00 7% 0.98 13% 0.85 12% 0.72 10% Source: Company data, Credit Suisse estimates

China Solar Sector 26 30 June 2017

We expect ~15-20% tariff cut in the next round of cuts, implying 10% equity IRR for a typical solar farm, mainly based on (1) further drop in solar farm investment cost (we use Rmb5.5/W); and (2) the implied IRR of ~10% from recent project bidding. Project return to be capped by bidding Expanding project Even if the final tariffs came in better than expected (like what happened in December bidding is a concern. 2016), we believe that the project returns of ground-mounted solar farms may still be capped by expanding project bidding going forward. With a 10% tariff discount, we estimate the equity IRR for new projects would be reduced from 14% (implied by benchmark tariffs) to ~10%. In Top Runner programmes, some aggressive bidders are even willing to accept a <10% equity IRR.

Figure 47: Equity IRR sensitivity to tariff discount Equity IRR Tariff discount 0% -5% -10% -15% -20% 4.5 26% 24% 20% 14% 7% 5.0 21% 19% 15% 10% 4% Unit capex (Rmb/W) 5.5 17% 15% 12% 7% 2% 6.0 14% 12% 10% 5% 0% Source: Company data, Credit Suisse estimates

In China, there are currently three types of ground-mounted projects: (1) provincial projects (total amount decided by NEA but project quota to be allocated by local governments); (2) Top Runner programme; and (3) poverty-alleviation solar projects. We expect the overall ground-mounted project quota to decline from ~31GW in 2016 (including ~10GW additional quota issued at year-end) to ~20GW in 2017. More importantly, our by-province analysis suggests there may be only ~6GW provincial quota (less competitive), and more quota will be allocated to Top Runner and poverty-alleviation projects (most of which are issued through a bidding process).

Figure 48: Ground-mounted project quota breakdown (GW) 2014 2015 2016 2017E Provincial quota 8.0 16.3 12.6 6.0 Additional provincial quota n.a. 5.3 10.0 n.a. Top Runner n.a. 1.0 5.5 8.0 Poverty alleviation (ground-mounted) n.a. 1.5 3.0 6.0 Total ground-mounted quota 8.0 24.1 31.1 20.0 Actual ground-mounted installation 8.6 13.8 30.3 17.0 Note: Given the construction lead time (0.5-1 year) of a solar project, the quota issued in one year may be completed within the year or early the next year. Source: NEA, Credit Suisse estimates

Even for provincial quota, the NEA is now encouraging local governments to issue through project bidding. We expect 50% of provincial quota to be allocated through bidding in 2017, and therefore >80% of total solar project quota in China will likely be issued through bidding.

China Solar Sector 27 30 June 2017

Figure 49: Ground-mounted project quota breakdown: bidding vs. non-bidding

100%

80% 60% Non-bidding quota 40% Bidding quota 20% 0% 2014 2015 2016 2017E

Source: Company data, Credit Suisse estimates Rising interest rate is an additional risk The industry average funding cost for solar companies can be as high as 8-10%, according to China Institute of Energy Economics Research. This means that when IRRs normalise, the funding cost differences may lower incentives for high-cost players. Solar companies The four companies we cover (XYS, GCLNE, Linyang, and CNE) currently have 3-7% usually have high funding cost, with XYS the lowest at 2-3% (given more non-RMB debt) and GCLNE the borrowing cost highest at 7%.

Figure 50: Funding cost comparisons (%) 10.0 8.0 6.0 4.0 2.0 0.0 XYS CNE LYE GCLNE SOE IPPs Solar industry average

Source: Company data, Credit Suisse estimates

Finance cost accounts for ~20% of all-in cost of solar power generation in China. Based on our calculation, a 50 bp interest rate hike may reduce equity IRR by about 0.5-1 pp.

Figure 51: Ground-mounted solar: equity IRR sensitivity to interest rate Tariff (Rmb/kWh) Equity IRR 0.80 0.75 0.70 0.65 0.60 0.55 0.50 3.0% 21% 18% 16% 13% 10% 8% 6% 3.5% 20% 17% 15% 12% 10% 7% 5% 4.0% 19% 17% 14% 12% 9% 7% 5% Interest rate 4.5% 18% 16% 13% 11% 9% 6% 4% 5.0% 18% 15% 13% 10% 8% 6% 4% 5.5% 17% 14% 12% 10% 8% 5% 3% 6.0% 16% 14% 12% 10% 7% 5% 3% Source: Credit Suisse estimates

China Solar Sector 28 30 June 2017

As a result, there has been an increasing trend of adopting the BT (Build and Transfer) business model for private renewables developers. For example, CNE sold a 100MW solar farm in 2016 and plans to sell another 300MW in 2017 (200MW already sold). GCLNE also plans to dispose 1.5-2GW project in the next one to two years, in order to alleviate its balance sheet pressure. Given the large capex requirement for solar farm projects, we believe the BT model will benefit private solar farm developers as it immediately realises the future profits of the project (although to be shared with the project buyer). Also, the cash received from project sales can be used to finance new project constructions.

Figure 52: Solar project sales of CNE and GCLNE 2016A 2017E 2018E 2019E CNE Solar project sold (MW) 100 300 350 400 Profit per W (Rmb) 1.6 0.8 0.8 0.8 GCLNE Solar project sold (MW) n.a. 1,000 500 500 Profit per W (Rmb) n.a. 0.8 0.8 0.8 Source: Company data, Credit Suisse estimates Cutting capacity and tariff assumptions for listcos Due to the uncertain outlook for ground-mounted solar, we have cut our annual capacity additions for listed companies under our coverage by 20-40%. At the same time, the expansion of project bidding has also been quicker than our previous expectation, which may result in lower tariffs for future projects. The key upside risks could be further industry consolidation and quicker-than-expected project development in distributed solar.

Figure 53: Capacity and tariff assumption changes 2017E 2018E 2019E Annual capacity addition assumptions (MW) XYS Prev. 700 700 800 New 700 500 500 GCLNE Prev. 1,500 700 700 New* 1000 500 500 % of projects won through bidding XYS Prev. 30% 30% 30% New 50% 70% 100% GCLNE Prev. 50% 50% 50% New 70% 90% 100% Tariff discount at project bidding XYS Prev. -5% -5% -5% New -10% -10% -10% GCLNE Prev. -15% -15% -10% New -15% -15% -15% Note: * Net capacity additions for GCLNE (assuming annual disposal at 0.5-1GW). Source: Credit Suisse estimates

China Solar Sector 29 30 June 2017

Stock calls We initiate coverage on two large upstream wafer producers, LONGi (market leader in mono-Si) and GCL (multi-Si), with OUTPERFORM and NEUTRAL ratings, respectively, given two different technology prospects. For operators, we upgrade Lingyang to NEUTRAL given rising investments in distributed energy and downgrade XYS to NEUTRAL with headwinds for ground-mounted farms and solar glass.

Figure 54: Rating, target price and EPS change summary Company Ticker Rating Target price (FC) EPS change (%) Prev. EPS (LC) New EPS (LC) New Old New Old FY17E FY18E FY19E FY17E FY18E FY19E FY17E FY18E FY19E LONGi 601012.SS O n.a. 23.00 n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.01 1.32 1.59 GCL 3800.HK N n.a. 0.80 n.a. n.a. n.a. n.a. n.a. n.a. n.a. 0.10 0.10 0.10 XYS 0968.HK N O 2.40 3.87 -14% -21% -31% 0.38 0.42 0.46 0.32 0.33 0.31 GCLNE 0451.HK N N 0.40 0.50 0% -28% -22% 0.04 0.05 0.05 0.04 0.04 0.04 Linyang 601222.SS N U 7.50 6.80 -9% 8% 13% 0.46 0.44 0.43 0.42 0.48 0.48 CNE 0182.HK O O 0.47 0.56 -23% -26% -38% 0.09 0.11 0.14 0.06 0.08 0.08 Source: Company data, Credit Suisse estimates Prefer LONGi over GCL Poly Due to capacity oversupply, most of the manufacturers along the solar value chain have been suffering from depressed margins. Polysilicon production and mono-Si wafer processing are probably the only two subsectors with decent gross margins (23-35%, based on current spot price), due to relatively higher entry barrier and more balanced supply and demand.

Figure 55: Gross margin comparison across solar value chain (based on current spot price) (Rmb/W) Poly Wafer Module Multi Mono Multi Mono Cost 0.38 0.85 0.78 2.52 2.58 ASP 0.48 1.02 1.18 2.70 3.00 Gross margin 21% 17% 34% 7% 14% LONGi's exposure* - - 90% - 10% GCL's exposure* 60% 40% - - - Note: Measured by FY18E gross profit exposure. Source: Bloomberg, WIND, Credit Suisse estimates

LONGi is our top pick LONGi is well positioned in the mono-Si wafer subsector (accounting for ~15% of global in China solar sector wafer capacity) and is still expanding its capacity to meet the rising mono-Si demand. Strong earnings growth (25% FY17-19E EPS CAGR) is supported by stable margin and growing market share. GCL Poly is also a leading company in the polysilicon (~17% global market share) and multi-Si wafer production (~20%). However, as GCL's wafer capacity is larger than its poly, and most of poly is sold internally, the company may not be able to benefit from the relatively more resilient poly price. On the multi-Si wafer side, the market share evasion from mono-Si is continuing, which may put further pressure on multi-Si wafer price.

China Solar Sector 30 30 June 2017

Figure 56: Growth outlook comparison between LONGi and GCL 2014 2015 2016 2017E 2018E 2019E Poly output growth LONGi n.a. n.a. n.a. n.a. n.a. n.a. GCL 33% 11% -7% 8% 16% 32% Wafer output growth LONGi 83% 36% 128% 48% 42% 28% GCL 39% 18% 15% 3% 7% 3% Gross profit growth LONGi 124% 94% 162% 27% 27% 18% GCL 107% 16% 22% -11% 6% 7% Net profit growth LONGi 312% 77% 197% 31% 30% 21% GCL n.a. 57% -16% -9% -2% 1% Source: Company data, Credit Suisse estimates

One of the key reasons why LONGi's mono-Si wafer has been gaining market share is that it achieved much better cost reduction compared to multi-Si wafer in the past few years. Back in 2013-14, LONGi's mono-Si wafer used to be priced ~20% higher than multi-Si wafer, and market demand was weak (1-2GW annual sales). By using diamond wiring technology and increasing its production scale, LONGi has reduced its unit wafer cost by 45% during 2013-16 (better than GCL at 31% based on our estimate). As a result, LONGi was able to price its mono-Si wafer at a similar ASP to multi-wafer in 2016, while retaining a decent gross margin (28%).

Figure 57: Wafer ASP and unit cost comparison FY13 FY14 FY15 FY16 LONGi ASP (Rmb/W) 1.6 1.6 1.3 1.1 Unit cost (Rmb/W) 1.4 1.3 1.0 0.8 Gross margin 12% 17% 22% 28% GCL Poly ASP (Rmb/W) 1.3 1.3 1.2 1.1 Unit cost (Rmb/W) 1.2 1.1 0.9 0.8 Gross margin 10% 19% 20% 27% Source: Company data, Credit Suisse estimates

LONGi is currently trading at 13x FY18E, below its past-five-year average of 17x. GCLNE is trading at a historical low (on both forward P/E and P/B), largely due to uncertain demand outlook for multi-Si products.

Figure 58: LONGi—one-year forward P/E history Figure 59: LONGi—one-year forward P/B history (x) (x) 50.0 8.0 7.0 40.0 6.0 32.1x Avg+2SD 30.0 5.0 4.5x Avg+2SD 24.5x Avg+1SD 4.0 20.0 3.6x Avg+1SD 16.8x Avg 3.0 2.6x Avg 10.0 2.0 9.1x Avg-1SD 1.6x Avg-1SD 1.0 1.5x Avg-2SD 0.6x Avg-2SD 0.0 0.0 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, Company data, Credit Suisse estimates Source: Bloomberg, Company data, Credit Suisse estimates

China Solar Sector 31 30 June 2017

Figure 60: GCL—one-year forward P/E history Figure 61: GCL—one-year forward P/B history (x) (x) 25 2.5 23.8x Avg+2SD 20 2.0 2.1x Avg+2SD 18.1x Avg+1SD 1.7x Avg+1SD 15 1.5 12.4x Avg 1.2x Avg 10 1.0 6.8x Avg-1SD 0.8x Avg-1SD 5 0.5 0.4x Avg-2SD 0 1.1x Avg-2SD 0.0 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, Company data, Credit Suisse estimates Source: Bloomberg, Company data, Credit Suisse estimates Downgrade ground-mounted solar farm operators We are turning Generally speaking, we expect the demand shift from ground-mounted solar to distributed cautious on traditional solar to be negative to traditional solar farm operators under our coverage (such as Xinyi solar farm operators Solar and GCL New Energy). Due to the uncertain outlook for ground-mounted solar, we have cut our annual capacity additions for listed companies under our coverage by 20- 40%. At the same time, the expansion of project bidding has also been quicker than our previous expectation, which may result in lower tariffs for future projects. The key upside risks could be further industry consolidation and quicker-than-expected project development in distributed solar.

Figure 62: Capacity and tariff assumption changes 2017E 2018E 2019E Annual capacity addition assumptions (MW) XYS Prev. 700 700 800 New 700 500 500 GCLNE Prev. 1,500 700 700 New* 1000 500 500 % of projects won through bidding XYS Prev. 30% 30% 30% New 50% 70% 100% GCLNE Prev. 50% 50% 50% New 70% 90% 100% Tariff discount at project bidding XYS Prev. -5% -5% -5% New -10% -10% -10% GCLNE Prev. -15% -15% -10% New -15% -15% -15% Note: * Net capacity additions for GCLNE (assuming annual disposal at 0.8-1GW). Source: Credit Suisse estimates

As a result, we have cut our FY18-19E EPS for XYS and GCLNE by 14-28%, and downgrade XYS to NEUTRAL with headwinds for ground-mounted farms and solar glass. We upgrade Linyang from Underperform to NEUTRAL mainly due to its increasing investment in the high-margin distributed solar business in the coastal regions, despite challenges for its power meter business. Maintain NEUTRAL for GCLNE given uncertainty on ground-mounted solar. Maintain OUTPERFORM for CNE given its cheap valuation and exposure in the wind farm sector. Summary of stock ideas LONGi Green Energy Tech (601012.SS, OUTPERFORM, TP: Rmb23.0) We initiate coverage on LONGi Green Technology (LONGi) with an OUTPERFORM rating. As the world's largest mono-Si wafer producer with ~40% market share, we expect

China Solar Sector 32 30 June 2017

LONGi to be one of the biggest beneficiaries of the expanding mono-Si solar component demand. We estimate a FY17-19 EPS CAGR of 25%, driven by wafer capacity expansion (up 70% in the next three years) in low-cost regions and a stable margin outlook (balanced demand and supply in the next two years). LONGi is our top pick in the China solar sector. GCL-Poly (3800.HK, NEUTRAL, TP: HK$0.80) We initiate coverage on GCL Poly (GCL) with a NEUTRAL rating and a sum-of-the-parts (SOTP) based target price of HK$0.80. As the largest multi-silicon (multi-Si) wafer producer in the world, we expect uncertain demand outlook for GCL due to market share expansion of the competing mono-Si technology. We expect largely flat EPS in FY17-19E, with increasing earnings from solar farm offset by profit drops at wafer manufacturing business.. Xinyi Solar Holdings (0968.HK, NEUTRAL, TP: HK$2.4) We expect XYS' solar glass business to face near-term pressure in 2H17 with potentially lower selling price due to oversupply (>20%). Despite rush installation demand in 1H17, the solar glass price continued to fall, declining 5-7% YTD (worse than our expectation). Besides, the future growth of its solar farm segment may also be impacted by less ground- mounted project quota. Downgrade to NEUTRAL.. GCL New Energy (0451.HK, NEUTRAL, TP: HK$0.40) We expect multiple uncertainties for GCLNE in the next few years, due to possible reduction in ground-mounted solar project quota and expanding low-tariff project bidding. Its high net gearing (>300%) is also a concern, especially given tightening bank lending. Valuation (FY18E P/B at 1.0x) does not seem attractive compared with leading renewable peers trading below book. Maintain NEUTRAL.. Jiangsu Linyang Energy (601222.SS, NEUTRAL, TP: Rmb7.5) Impacted by less power meter tendering and margin pressure, the share price of LYE has been weak since late 2016. Although its traditional power meter business is yet to see the inflection point, we expect earnings growth from its distributed solar business may accelerate in the next three years (CAGR at 30%) driven by strong policy support. Upgrade to NEUTRAL. . Concord New Energy (0182.HK, OUTPERFORM, TP: HK$0.47) We cut Concord New Energy’s (CNE) FY17-18E EPS by 23-38% to reflect lower earnings contribution from solar farms. The stock is trading at an appealing valuation of 0.5x FY17E P/B (the lowest within the utilities universe). Dividend payout expanded to ~30% in FY16 and the company expects a similar payout ratio in the next few years. We believe risk- reward for CNE is attractive with potential utilisation recovery and BT profit expansion as the key catalysts. Maintain OUTPERFORM. . Key risks Roof-top rental At the moment, we believe the key risks for development of distributed solar in China are contract and project (1) difficulty in finding an ideal roof-top and maintaining a 20-year rental contract with the financing are the two landlord; and (2) project financing. key risks for distributed solar Risks associated with roof-top rental contract To find an ideal roof-top to build a distributed solar system is not always easy. A roof-top rental contract is usually signed for 25 years, while the counterparty risk cannot be neglected. It is essential for project developers to ensure legal protection and compensation in case the roof-top owner (such as a factory) relocates or even shuts down. Solar projects require heavy initial investment and we estimate the average payback period for distributed solar can be six to seven years, so that early unexpected termination could be quite damaging to project return (even resulting in losses).

China Solar Sector 33 30 June 2017

Besides, roof-top owners have already enjoyed lower electricity costs; however, (usually 15%), some may also ask for higher roof-top rental fee, as there is now increasing demand on roof-top resources due to its superior project return compared to other solar projects. The average roof-top rental fee is currently around Rmb4.0/sq m per year while some with better solar resources may charge Rmb6.0/sq m. In our base case scenario, we have assumed a 15% rental fee increase every five years. However, the IRR sensitivity to rental cost change is relatively low. A 50% increase in starting rental fee (from Rmb0.04/W to Rmb0.06/W; land requirement per W is 0.01 sq m) may only bring down equity IRR by 100 bp.

Figure 63: IRR sensitivity to roof-top rental cost Electricity price discount Equity IRR 5% 10% 15% 20% 25% 30% 35% 0.04 17% 16% 15% 13% 12% 11% 10% 0.06 16% 15% 13% 12% 11% 10% 9% 0.08 15% 14% 12% 11% 10% 9% 7% Roof-top rental cost 0.10 14% 12% 11% 10% 9% 7% 6% (Rmb/W) 0.12 13% 11% 10% 9% 8% 6% 5% 0.14 11% 10% 9% 8% 6% 5% 4% 0.16 10% 9% 8% 6% 5% 4% 2% Source: Credit Suisse estimates Project financing difficulty Unlike large-scale ground-mounted solar farms, distributed solar projects are usually smaller and may not provide sufficient asset collateral when it comes to borrowing. Besides, their project developers are mostly private companies which are less favoured by banks (banks prefer to lend to large SOEs). As mentioned before, possible roof-top disputes and long pay-back periods are also among the risks that banks would consider. As a result, it can be difficult for distributed solar projects to obtain bank loans which are usually cheaper than other means of financing. Since 2015-16, there has been increasing support from local governments (usually in some coastal provinces) to provide financing for distributed solar projects. According to China Securities Journal, Shanghai issued the first local policy on bank loans for distributed solar projects in December 2015. According to the policy, banks should provide favourable lending rate to distributed solar projects (at par with benchmark interest rate or no higher than 15% mark-up). The loan default risk is to be shared by financial guarantee companies (charging 1% per year) and local governments. Similarly, some banks in Zhejiang, Jiangsu, Shandong and so on are also starting to provide financing for distributed solar projects. We currently assume 6.5% average cost of borrowing for a distributed solar project. A ~50 bp saving in interest rate may raise equity IRR by close to 80 bp, based on our estimate.

Figure 64: IRR sensitivity to roof-top rental cost Weighted average tariff (Rmb/kWh) Equity IRR 1.10 1.05 1.00 0.95 0.90 0.85 0.80 5.0% 20% 18% 15% 13% 11% 9% 7% 5.5% 19% 17% 15% 13% 11% 9% 7% 6.0% 18% 16% 14% 12% 10% 8% 6% Interest rate 6.5% 17% 15% 13% 11% 9% 7% 5% 7.0% 16% 14% 12% 11% 9% 7% 5% 7.5% 16% 14% 12% 10% 8% 6% 4% 8.0% 15% 13% 11% 9% 7% 6% 4% Source: Credit Suisse estimates

China Solar Sector 34 30 June 2017

Many private companies are also seeking other means of financing such as finance lease. The average borrowing cost under finance lease may be ~100 bp higher than bank loans but it is easier to obtain, unlike bank loans which usually require adequate collaterals. Besides, the depreciation period under finance lease is usually shorter than the normal depreciation period for fixed assets, and thus depreciation cost will be higher in the first few years, creating tax benefits.

China Solar Sector 35 30 June 2017

Appendix Introduction of mono-Si and multi-Si technologies Multi-Silicon (multi-Si) and Mono-Silicon (mono-Si) are the two mainstream technologies that dominate >95% of the current solar panel market. Similar to the semiconductor industry, the technology development of the solar component industry is also progressing fast with constant changes. Back in the 1990s, mono-Si used to account for >50% market share but started to lose the market to other technologies, especially during the 2010s, when multi-Si was aggressively taking market share helped by massive cost cuts. Since 2015-16, however, mono-Si has come back into the spotlight by significantly reducing its price premium to the multi-Si module, which makes mono-Si more competitive due to its higher efficiency. Some industry consultants (ITRPV) even forecast a further market share expansion of mono-Si to ~50% by 2020.

Figure 65: Global annual PV production by technology (multi-Si, mono-Si & thin film)

100% 90% 80% 70% 60% 50% 40% 30% 20% 10%

0%

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

2016E 2017E Thin film Multi-Si Mono-Si

Source: IHS, EnergyTrend

Figure 66: Comparison between mono-Si and multi-Si solar panels Mono-Si panel Multi-Si panel Cost per W More expensive Less expensive Efficiency More efficient (due to the structure being made from one large crystal as opposed to many small Less efficient ones. Also performing up to 10% better than polycrystalline panels in high ambient temperatures) Size Smaller size for the same wattage Larger size for the same wattage (due to higher efficiency) Looks Solar cells are a black hue with a more circular cell shape Solar cells have a blue-ish hue and are in squares Longevity 25+ years 25+ years Source: Energy Informative, Civic Solar, Credit Suisse research

Both mono-Si and multi-Si solar panels use polysilicon as the main raw material. The key difference in the manufacturing process is during the ingot formation. Multi-Si ingot is produced simply by melting the polysilicon into bricks which consist of a number of smaller crystals. Mono-Si, on the other hand, is a single crystal silicon with continuous internal structure. To make mono-Si wafer, polysilicon (the raw material) is manufactured into a large cylindrical ingot and then sliced into pieces. The wafering process is similar for both mono-Si and multi-Si, with mainly two technologies: (1) slurry-based (less efficient, more commonly used in multi-Si) and (2) diamond wire sawing (more efficient, more commonly used in mono-Si). The cell and module processing are basically the same for mono-Si and multi-Si. After wiring and coating, wafers are made into cells and cells are assembled into modules. Modules are used in to build the solar power generation systems (solar farms or distributed solar systems), which convert solar energy into electricity.

China Solar Sector 36 30 June 2017

Asia Pacific/China Multi Utilities

LONGi Green Energy Technology

(601012.SS / 601012 CH) Rating OUTPERFORM Price (28-Jun-17, Rmb) 17.11 INITIATION Target price (Rmb) 23.00 Upside/downside (%) 34.4

Mkt cap (Rmb/US$ mn) 34,150 / 5,022 Leader in a growing market Enterprise value (Rmb mn) 32,209 Number of shares (mn) 1,996 ■ Beneficiary of growing mono-Si market. We initiate coverage on LONGi Free float (%) 59.9 Green Technology (LONGi) with an OUTPERFORM rating. As the world's 52-wk price range (Rmb) 17.23-12.86 ADTO-6M (US$ mn) 34.5 largest mono-Si wafer producer with ~40% market share, we expect LONGi Target price is for 12 months. to be one of the biggest beneficiaries of the expanding mono-Si solar

component demand. We estimate a FY17-19 EPS CAGR of 25%, driven by Research Analysts

wafer capacity expansion (up 70% in the next three years) in low-cost regions Gary Zhou, CFA and a stable margin outlook (balanced demand and supply in the next two 852 2101 6648 [email protected] years). LONGi is our top pick in the China solar sector. Dave Dai, CFA ■ Resilient ASP supported by rising demand. Unlike multi-Si wafer which has 852 2101 7358 [email protected] seen 12% price drop YTD in 2017, the ASP for mono-Si wafer has been much Gloria Yan more resilient (up 2%), supported by rising demand for mono-Si components. 852 2101 7369 As a result, LONGi recorded strong core earnings growth of 51% YoY in 1Q17 [email protected] (one of the best among solar peers), and its gross margin expanded further to 31% (vs. 27% in FY16). Due to superior cost efficiency, we forecast mono-Si's global market share to double from 25% in 2016 to 50% by 2020. ■ Cost leader in the wafer industry. LONGi has achieved significant wafer production cost reduction in the past few years (unit cost down 40% from 2013-16), thanks to more efficient diamond-wire cutting technology and LONGi's increasing economy of scale. Based on our calculation, its unit cost is now close to GCL Poly (leader in multi-Si wafer) but provides better efficiency and thus is preferred by the end-users. Further cost reduction may be driven by capacity expansion in low-cost regions such as Ningxia and Yunnan in 2017-19. ■ Valuation. We value the company using DCF model with a TP of Rmb23.0 (WACC at 9%). The stock is currently trading at FY18E P/E of 13x, much lower than its past-five-year average of 17x. Its low net gearing (<10% in FY18-19E) and high ROE (20-25%) also stands out in the China solar sector. Key risks are: (1) lower-than-expected mono-Si wafer and module prices; (2) more-than-expected capacity expansion from peers. Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (Rmb mn) 11,530.5 17,453.1 23,419.8 27,663.6 EBITDA (Rmb mn) 2,540.2 3,251.8 4,264.4 5,252.6 EBIT (Rmb mn) 2,150.2 2,571.7 3,225.8 3,841.7 Net profit (Rmb mn) 1,547.2 2,019.9 2,630.2 3,178.6 EPS (CS adj.) (Rmb) 0.86 1.01 1.32 1.59 Change from previous EPS (%) n.a. - - - Consensus EPS (Rmb) n.a. 0.98 1.25 1.55 EPS growth (%) 177.4 17.7 30.2 20.9 The price relative chart measures performance against the P/E (x) 19.9 16.9 13.0 10.7 Shanghai Shenzhen CSI300 index which closed at Dividend yield (%) 0.6 0.9 1.5 2.3 3,674.72 on 28/06/17. On 28/06/17 the spot exchange rate EV/EBITDA (x) 12.4 10.2 8.1 6.7 was Rmb6.8/US$1 P/B (x) 3.05 3.75 3.05 2.51

Performance 1M 3M 12M ROE (%) 19.7 21.1 25.9 25.6 Absolute (%) 15.4 7.4 31.7 Net debt/equity (%) Net cash Net cash 4.8 9.6

Relative (%) 9.8 0.4 14.6 Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 37 30 June 2017

Focus charts and table

Figure 67: LONGi—capacity expansion plan by Figure 68: LONGi is a key beneficiary of rising business segment demand for mono-Si solar component (GW) 60 60%

40 40%

20 20%

0 0% 2015 2016 2017E 2018E 2019E 2020E LONGi' output Global demand for mono-Si LONGi's market share

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

Figure 69: Wafer price history Figure 70: LONGi—gross profit trend

Source: Bloomberg Source: Company data, Credit Suisse estimates

Figure 71: Decent profitability for mono-Si wafer (on current spot price) Figure 72: Net gearing comparison Wafer Module (Rmb/W) Poly Multi Mono Multi Mono Cost 0.38 0.85 0.78 2.52 2.58 ASP 0.48 1.02 1.18 2.70 3.00 Gross margin 21% 17% 34% 7% 14% LONGi's ex posure* - - 90% - 10% GCL's ex posure* 60% 40% - - -

Note: * Measured by FY18E gross profit exposure. Source: Company data, Credit Suisse estimates Source: Bloomberg, WIND, Credit Suisse estimates

Figure 73: LONGi—one-year forward P/E history Figure 74: LONGi—one-year forward P/B history (x) (x) 50.0 8.0 7.0 40.0 6.0 32.1x Avg+2SD 30.0 5.0 4.5x Avg+2SD 24.5x Avg+1SD 4.0 20.0 3.6x Avg+1SD 16.8x Avg 3.0 2.6x Avg 10.0 2.0 9.1x Avg-1SD 1.6x Avg-1SD 1.0 1.5x Avg-2SD 0.6x Avg-2SD 0.0 0.0 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, company data, Credit Suisse estimates Source: Bloomberg, company data, Credit Suisse estimates

China Solar Sector 38 30 June 2017

LONGi Green Energy Technology (601012.SS / 601012 CH) Price (28 Jun 2017): Rmb17.11; Rating: OUTPERFORM; Target Price: Rmb23.00; Analyst: Gary Zhou Income Statement (Rmb mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 11,531 17,453 23,420 27,664 LONGi Green Energy Technology (LONGi) is the largest mono- Cost of goods sold 8,361 13,444 18,312 21,654 silicon solar wafer manufacturer in the world. Founded in 2000, the EBITDA 2,540 3,252 4,264 5,253 company has been in the mono-silicon industry for more than a EBIT 2,150 2,572 3,226 3,842 decade, and we expect it to maintain its dominant market share. Net interest expense/(inc.) 102 218 315 385 Recurring PBT 2,125 2,540 3,147 3,783 Blue/Grey Sky Scenario Profit after tax 1,551 2,025 2,637 3,186 Reported net profit 1,547 2,020 2,630 3,179 Net profit (Credit Suisse) 1,547 2,020 2,630 3,179 Balance Sheet (Rmb mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 5,817 6,169 6,125 6,435 Current receivables 3,892 5,872 7,870 9,298 Inventories 1,213 1,837 2,465 2,911 Other current assets 1,120 1,533 1,953 2,268 Current assets 12,042 15,412 18,413 20,912 Property, plant & equip. 4,591 7,307 10,861 13,820 Investments 385 152 152 152 Intangibles 216 137 137 137 Other non-current assets 1,939 219 230 241 Total assets 19,172 23,227 29,793 35,263 Current liabilities 6,451 9,218 11,888 13,753 Total liabilities 9,079 14,124 18,579 21,657 Shareholders' equity 10,093 9,097 11,201 13,585 Minority interests 1 6 13 20 Total liabilities & equity 19,172 23,227 29,793 35,263 Cash Flow (Rmb mn) 12/16A 12/17E 12/18E 12/19E EBIT 2,150 2,572 3,226 3,842 Net interest (102) (218) (315) (385) Tax paid (242) (315) (411) (496) Working capital 264 (311) (440) (394) Other cash & non-cash items (1,535) 160 1,226 1,570 Our Blue Sky Scenario (Rmb) 25.00 Operating cash flow 536 1,888 3,286 4,136 Solar wafer sales price to increase by 2% YoY in 2018 Capex (2,360) (3,397) (4,592) (4,370) Free cash flow to the firm (1,824) (1,509) (1,306) (234) Our Grey Sky Scenario (Rmb) 21.00 Investing cash flow (2,152) (3,397) (4,592) (4,370) Solar wafer sales price to drop by 18% YoY in 2018 Equity raised 2,980 0 0 0 Dividends paid (80) (200) (303) (526) Share price performance Financing cash flow 5,001 1,861 1,262 543 Total cash flow 3,385 353 (44) 309 Adjustments 0 0 0 0 Net change in cash 3,385 353 (44) 309 Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 1,799 1,996 1,996 1,996 EPS (Credit Suisse) (Rmb) 0.86 1.01 1.32 1.59 DPS (Rmb) 0.10 0.15 0.26 0.40 Operating CFPS (Rmb) 0.30 0.95 1.65 2.07 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue 93.9 51.4 34.2 18.1 EBIT 171.1 19.6 25.4 19.1 EPS 177.4 17.7 30.2 20.9 Margins (%) The price relative chart measures performance against the Shanghai EBITDA 22.0 18.6 18.2 19.0 Shenzhen CSI300 index which closed at 3,646.17 on 28-Jun-2017 EBIT 18.6 14.7 13.8 13.9 On 28-Jun-2017 the spot exchange rate was Rmb6.8/US$1 Valuation (x) 12/16A 12/17E 12/18E 12/19E P/E 19.9 16.9 13.0 10.7 P/B 3.05 3.75 3.05 2.51 Dividend yield (%) 0.6 0.9 1.5 2.3 EV/sales 2.7 1.9 1.5 1.3 EV/EBITDA 12.4 10.2 8.1 6.7 EV/EBIT 14.6 12.9 10.8 9.2 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 19.7 21.1 25.9 25.6 ROIC 32.9 29.3 28.4 25.0 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) (27.5) (11.8) 4.8 9.6 Net debt/EBITDA (x) (1.09) (0.33) 0.13 0.25

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 39 30 June 2017

Leader in a growing market We initiate coverage on LONGi Green Technology (LONGi) with an OUTPERFORM rating and a target price of Rmb23. Due to superior cost efficiency, we forecast global market share of mono-Si solar component to double from 25% in 2016 to 50% by 2020. As the world's largest mono-Si wafer producer with ~40% market share, we expect LONGi to be one of the biggest beneficiaries. We estimate a FY17-19E EPS CAGR of 25%, driven by its wafer capacity expansion (up 70% in the next three years) in low-cost regions and a stable margin outlook (balanced demand and supply in the next two years). LONGi is our top pick in the China solar sector. Our target price of Rmb23.0 is derived from a DCF model, assuming a WACC of 9% (cost of equity at 12% and cost of debt at 5%). The stock is currently trading at FY18E P/E of 13x, much lower than its past-five-year average of 17x. Our TP of Rmb23.0 implies a target FY18E P/E of 17x, in line with the historical average. LONGi's low net gearing (<10% in FY18-19E) and high ROE (20-25%) also stand out in the China solar sector.

Figure 75: LONGi—DCF valuation method Key Assumptions Risk-free rate 3% Enterprise value 47,054 Equity risk premium 9% Net debt (1,070) Beta 1.0 Total equity 45,984 Cost of equity 12% Deduct minority (6) Cost of debt (after-tax) 4.5% Shareholders' equity 45,978 % of debt 40% No. of shares 1,996 WACC 9% Terminal growth 0% TP (Rmb) 23.0 Source: Company data, Credit Suisse estimates

Figure 76: LONGi—one-year forward P/E history Figure 77: LONGi—one-year forward P/B history (x) (x) 50.0 8.0 7.0 40.0 6.0 32.1x Avg+2SD 30.0 5.0 4.5x Avg+2SD 24.5x Avg+1SD 4.0 20.0 3.6x Avg+1SD 16.8x Avg 3.0 2.6x Avg 10.0 2.0 9.1x Avg-1SD 1.6x Avg-1SD 1.0 1.5x Avg-2SD 0.6x Avg-2SD 0.0 0.0 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17

Source: Bloomberg, Company data, Credit Suisse estimates Source: Bloomberg, Company data, Credit Suisse estimates

China Solar Sector 40 30 June 2017

Figure 78: LONGi—key operating metrics 2013 2014 2015 2016 2017E 2018E 2019E 2020E Mono-Si wafer Capacity year-end (MW) 1,600 3,000 5,000 7,500 14,200 14,200 24,200 24,200 Total sales (MW) 1,174 1,971 2,662 7,204 10,000 14,200 18,200 24,200 - YoY 68% 35% 171% 39% 42% 28% 33% (1) External sales (MW) 1,174 1,963 1,929 4,696 5,500 7,200 9,200 13,700 - YoY 67% -2% 143% 17% 31% 28% 49% (2) Internal sales to module segment (MW) 0 9 733 2,508 4,500 7,000 9,000 10,500 Wafer ASP ex. VAT (Rmb/W) 1.58 1.61 1.33 1.08 1.02 0.93 0.86 0.79 - YoY 2% -18% -18% -6% -8% -8% -8% Wafer revenue (Rmb mn) 1,861 3,159 2,557 5,075 5,587 6,729 7,911 10,837 - YoY 70% -19% 98% 10% 20% 18% 37% Wafer unit cost (Rmb/W) 1.39 1.33 1.04 0.78 0.69 0.63 0.57 0.52 - YoY -5% -22% -25% -12% -9% -9% -9% GP margin 12% 17% 22% 28% 32% 33% 34% 34% Unit gross profit (Rmb/W) 0.19 0.28 0.29 0.30 0.33 0.31 0.29 0.27 Gross profit - external sales (Rmb mn) 223 552 551 1,429 1,810 2,217 2,651 3,695 - YoY 148% 0% 160% 27% 22% 20% 39% Mono-Si module Capacity year-end (MW) 1,500 5,000 6,000 8,000 10,000 11,000 Total sales (MW) 760 2,129 4,500 7,000 9,000 10,500 - YoY 180% 111% 56% 29% 17% 1) External sales (MW) 721 1,847 4,400 6,900 8,900 10,400 - YoY 156% 138% 57% 29% 17% 2) Internal sales to solar farm segment (MW) 39 281 100 100 100 100 Module ASP (Rmb/W) 3.49 3.09 2.50 2.27 2.09 1.93 - YoY -12% -19% -9% -8% -8% Unit cost (Rmb/W) 2.82 2.25 2.05 1.90 1.75 1.62 - YoY -20% -9% -8% -8% -8% GP margin 19% 27% 18% 17% 16% 16% Unit gross profit (Rmb/W) 0.67 0.84 0.45 0.38 0.34 0.30 Gross profit - external sales (Rmb mn) 487 1,551 1,960 2,597 3,012 3,163 Gross profit breakdown (1) By external sales (Rmb mn) Wafer 551 1,429 1,810 2,217 2,651 3,695 Module 487 1,551 1,960 2,597 3,012 3,163 Wafer % 53% 48% 48% 46% 47% 54% Module % 47% 52% 52% 54% 53% 46% (2) By process (Rmb mn) Wafer 756 1,991 3,257 4,341 5,215 6,500 Module 281 988 513 473 448 358 Wafer (%) 73% 67% 86% 90% 92% 95% Module (%) 27% 33% 14% 10% 8% 5% Source: Company data, Credit Suisse estimates Growth driven by market share expansion One of the key reasons why mono-Si is now preferred by end-users is that leading manufacturers (such as LONGi) have managed to reduce mono-Si wafer production costs more quickly than their multi-Si peers. As a result, the price premium of mono-Si wafer reduced from US$0.3 per piece in 2014 to only US$0.02 in early 2016, and mono-Si's market share also expanded from 18% in 2015 to 25% in 2016. As we have highlighted in the sector section, mono-Si now has advantages in both unit cost and cell efficiency, and therefore we expect its market share expansion to continue in 2017-20E (30/40/45/50%).

China Solar Sector 41 30 June 2017

Figure 79: Global annual solar module production volume breakdown by technologies (multi-Si, mono-Si and thin film)

100% 90% 80% 70% 60% 50% 40% 30% 20% 10%

0%

1990 1980 1982 1984 1986 1988 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

2016E 2018E 2020E Thin film Multi-Si Mono-Si

Source: IHS, EnergyTrend, Credit Suisse estimates

As the largest mono-Si wafer manufacturer, we expect LONGi (~40% market share) to be a key beneficiary of rising mono-Si demand. The second-largest manufacturer is also a Chinese company called Zhonghuan Semiconductor (002129.SZ) with ~25% market share. Unlike other solar subsectors, mono-Si wafer industry has high market concentration and therefore is relatively less affected by competition at the supply side. Although both LONGi and Zhonghuan have aggressive capacity expansion plan until 2020 (both may triple from 2016), we expect most of the increased supply to be absorbed by further market share expansion of mono-Si demand and mild global demand growth. To be conservative, we have assumed unit wafer profit to normalise to Rmb0.27/W by 2020, while LONGi's strong gross profit growth (23% CAGR over FY16-20E) is still well supported by growing demand and its capacity expansion.

Figure 80: Mono-Si wafer capacity forecasts (MW) 2014 2015 2016 2017E 2018E 2019E 2020E LONGi 3,000 5,000 7,500 14,200 14,200 24,200 24,200 Zhonghuan Semiconductor 2,000 3,000 5,000 9,000 9,000 15,000 15,000 GCL Poly 0 0 1,000 1,500 2,000 2,500 3,000 Yinyang New Energy 0 1,000 1,000 2,000 2,000 2,000 2,000 Solargiga 900 900 900 900 900 900 900 Comtec 700 700 400 400 400 400 400 Top module manufacturers 0 0 2,500 5,500 8,500 8,500 8,500 Total (year-end) 6,600 10,600 18,300 33,500 37,000 53,500 54,000 LONGi's market share 45% 47% 41% 42% 38% 45% 45% Time-weighted capacity (GW) 8 10 15 24 35 44 54 Market demand for mono-Si (GW) 9 9 19 25 35 41 48 - Global solar demand (industry forecasts) 45 53 75 84 87 90 96 - Mono-Si's market share 20% 18% 25% 30% 40% 45% 50% Capacity utilisation of mono-Si 112% 91% 124% 105% 99% 93% 89% Source: Company data, GTM Research, Credit Suisse estimates

China Solar Sector 42 30 June 2017

Figure 81: LONGi—unit wafer gross profit and total gross profit trend (Rmbm) (Rmb/W) 8,000 0.35

6,000 0.30

4,000 0.25

2,000 0.20

0 0.15 2013 2014 2015 2016 2017E 2018E 2019E 2020E Total gross profit Unit wafer gross profit

Source: Company data, Credit Suisse estimates

Cost leader in the industry Back in 2013-14, LONGi's mono-Si wafer used to be priced ~20% higher price than multi- Si wafer, and the market demand was much smaller (1-2GW annual sales). By using diamond wiring technology and increasing its production scale, LONGi has reduced its unit wafer cost significantly by 45% over 2013-16 (better than GCL at 31%, based on our estimate). As a result, LONGi was able to price its mono-Si wafer at a similar ASP as multi-wafer in 2016, while retaining a decent gross margin (28%).

Figure 82: Wafer ASP and unit cost comparison FY13 FY14 FY15 FY16 LONGi ASP (Rmb/W) 1.6 1.6 1.3 1.1 Unit cost (Rmb/W) 1.4 1.3 1.0 0.8 Gross margin 12% 17% 22% 28% GCL Poly ASP (Rmb/W) 1.3 1.3 1.2 1.1 Unit cost (Rmb/W) 1.2 1.1 0.9 0.8 Gross margin 10% 19% 20% 27% Source: Company data, Credit Suisse estimates

Plant location is another advantage for LONGi. Most of its existing capacity and all of its new capacity under construction are located in Ningxia and Yunnan where both electricity (coal-fired tariff at only Rmb0.26/kWh in Ningxia vs. the national average of Rmb0.36/kWh and cheap hydro power in Yunnan) and labour costs are lower than in other regions.

China Solar Sector 43 30 June 2017

Figure 83: LONGi—plant locations

Ningxia: Ingot: 7GW (+6GW under construction) Wafer: 5GW under construction Module: 0.5GW

Jiangsu: Wafer: 4GW Module: 2GW

Shaanxi: Wafer: 3GW

Zhejiang: Module: 2.5GW

Yunnan: Ingot: 10GW under construction Wafer: 10GW under construction

Source: Company data

As a result, LONGi has been enjoying one of highest gross margins in the mono-Si wafer industry. However, the exact product-by-product margin comparison may be difficult as many peers do not disclose detailed profit breakdown.

Figure 84: Gross margin comparison of mono-Si manufacturers (company level) Gross margin 2012 2013 2014 2015 2016 LONGi 13% 12% 17% 20% 27% Zhonghuan Semiconductor (002129.SZ) 10% 12% 15% 15% 14% Solargiga (0757.HK) -29% 7% 13% 8% 11% Comtec (0712.HK) 8% 8% 7% -9% -19% Source: Company data, Bloomberg Mono-Si wafer is the main earnings driver Although LONGi is an integrated solar company with exposure to mono-Si ingot, wafer, cell, module and solar farm, it has clear focus on mono-Si wafer which is the most profitable segment. The cell and module production is mainly for the purpose of promoting its mono-Si wafer product to downstream cell and module makers. In the past, those downstream manufacturers were reluctant to use mono-Si wafer as it requires some modifications to their existing equipment. In Oct-2014, LONGi acquired Zhejiang Lerri Solar (a module manufacturer) and started to produce its own mono-Si modules. Good product quality and a competitive pricing strategy helped cultivate downstream demand for mono-Si wafer. However, at the company level, LONGi's wafer capacity (24GW by 2020E, high margin) would be much larger than module (10GW by 2020E, low margin), so that its net exposure to mono-Si wafer would assure a decent profit margin for the company. For the solar farm segment, the company is also cautious on new additions and prefers the build-and-transfer model.

China Solar Sector 44 30 June 2017

Figure 85: LONGi—capacity expansion plan by business segment

(GW) 24 25 23 20 14 14 15 13 13 10 10 7 8 8 5 5 5 6 3 3 3 4 4 5 1 2 2 2 3 1 0 0 0 0 0 0 0 0 2012 2013 2014 2015 2016 2017E 2018E 2019E Ingot Wafer Cell Module

Source: Company data, Credit Suisse estimates

As a result, we expect the mono-Si wafer segment to be the main profit driver for the company, accounting for ~90% of its gross profit by 2020E.

Figure 86: LONGi—gross profit breakdown

100%

80%

60% Others

40% 81% 85% 87% 90% Module 65% 63% Wafer 20%

0% 2015 2016 2017E 2018E 2019E 2020E

Source: Company data, Credit Suisse estimates One of the strongest balance sheets Compared to other solar manufacturing companies, LONGi has one of the lowest net gearings, which is helped by its strong operating cash flows from mono-Si wafer business and limited exposure to the capital-intensive solar farm business. For the solar farm investment, we believe the BT (Build and Transfer) model would be more suitable for private solar companies as it immediately realises the future profits of the project (although to be shared with the project buyer) and the cash received from project sales can be used to finance new project constructions.

China Solar Sector 45 30 June 2017

Figure 87: Net gearing comparison

160% 140% 120% 100% 2017E 80% 60% 2018E 40% 2019E 20% 0% -20% LONGi GCL Poly Xinyi Solar Linyang JA Solar Jinko Energy Holdings Solar

Source: Company data, Credit Suisse estimates Key risks The key risks to our OUTPERFORM rating on LONGi would be lower-than-expected mono-Si product ASP and larger-than-expected industry capacity growth. Watch out for capacity expansion Although we are optimistic on the demand outlook for mono-Si module, investors should also closely watch out for the mono-Si capacity expansion. Oversupply is the key issue for the solar manufacturing industry and several times in the past near-term high profitability attracted irrational capacity expansion. Assuming global solar demand of 84/87/90/96GW for FY17-20E (industry forecasts from GTM Research, etc.), the market demand for mono-Si will be 25/35/41/48GW based on our market share forecast. Our bottom-up industry research suggests weighted mono-Si wafer capacity to be 24/35/44/54GW during the period. As a result, we expect the mono-Si market to remain balanced from 2017-18E. Starting from 2019, we expect supply could be slightly higher than demand for mono-Si, but it may also change depending on global demand growth and further mono-Si market share expansion.

Figure 88: Global mono-Si demand and supply Figure 89: Mono-Si expansion supported by market forecasts share gains

(GW) Abundant (GW) Balanced 40 140% 70 supply supply and 35 120% 60 54 Supply demand 48 30 100% Mono-Si wafer time- 50 44 41 25 weighted capacity shortage 80% 40 35 35 20 60% Mono-Si wafer's 30 24 25 15 19 market share 20 16 10 40% 9 10 10 5 20% Mono-Si wafer's capacity utilization 0 0 0%

2015 2016 2017E 2018E 2019E 2020E

2013 2014 2015 2016 2011 2012

2017E 2018E Mono-Si supply Mono-Si demand

Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates

Sensitivity to mono-Si wafer price We estimate 1% lower mono-Si wafer price would affect LONGi's FY18E EPS by 2.4%. However, compared with multi-Si wafer, the mono-Si wafer price has been relatively more stable (+2% YTD while mono-Si down 12%).

China Solar Sector 46 30 June 2017

Figure 90: FY18E EPS sensitivity to mono-Si wafer price changes Mono-Si wafer ASP comparing to -3% -2% -1% 0% 1% 2% 3% our base-case assumption FY18E EPS change -7.2% -4.8% -2.4% 0.0% 2.4% 4.8% 7.2% Source: Company data, Credit Suisse estimates

Figure 91: Wafer price history (US$ per piece) 2.0

1.5

1.0

0.5

0.0 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 156mm Mono Wafer price 156mm Multi Wafer price

Source: Company data, Credit Suisse estimates Company background LONGi was founded in 2000 and its stock was listed on the Shanghai Stock Exchange in April 2012. Driven by strong mono-Si product demand, the company has been rapidly ramping up its wafer capacity, from 3GW in 2014 to 7.5GW in 2016. In October 2014, LONGi acquired a module producer called Zhejiang LERRI Solar Technology Co. The founder of the company, Mr. LI Zhengguo, graduated in physics (major) from Lanzhou University in China. He founded the company in 2000 and is currently its CEO. He (15%) and his wife (5%) together own a 20% stake in the company. Mr. Zhong Baoshen, currently the Chairman of LONGi, joined the company in 2006. He and Mr. LI are both alumni of Lanzhou University. The second-largest shareholder (11%), Mr. LI Chun'an, is an Executive Director of LONGi and the Chairman of Dalian Linton Machine (a private company which also supplies mono-Si manufacturing equipment to LONGi).

Figure 92: LONGi—shareholding structure

Mr. LI Zhenguo (Founder Mr. ZHONG Baoshen Other public Mr. LI Chun'an and CEO) and his wife (Chairman) shareholders 20.28% 10.95% 1.77% 67.00% LONGi Green Energy (601012.SS)

Source: Company data

Figure 93: LONGi—key events Date Event Feb-2000 Founded. Named Xi'an Xinmeng Electronic Technology at that time. Jul-2008 Renamed to Xi'an LONGi Silicon materials Corp. Apr-2012 Listed on the Shanghai Stock Exchange Oct-2014 Acquired LERRI Solar (module producer). Dec-2016 Announced to build 10GW new wafer capacity in Chuxiong, Yunnan. Source: Company data

China Solar Sector 47 30 June 2017

Figure 94: Key management profile Name Title Introduction Mr. LI Zhengguo CEO Born in 1968. Graduated from the physics major at Lanzhou University in China. He (Founder) founded the company in 2000 and is currently the CEO of LONGi. He and his wife currently own a 20% stake in the company. Mr. ZHONG Baoshen Chairman Born in 1967, currently the Chairman of LONGi, joined the company in 2006, and he and Mr. LI are both alumni of Lanzhou University. Source: Company data

Figure 95: LONGi—capacity expansion plan (MW) 2014 2015 2016 2017E 2018E 2019E 2020E Ingot Yinchuan, Ningxia 3,800 5,000 10,000 10,000 10,000 10,000 Zhongning, Ningxia 1,000 2,000 3,000 3,000 3,000 3,000 Lijiang, Yunnan 5,000 5,000 Baoshan, Yunnan 5,000 5,000 Kuching, Malaysia 300 300 300 300 300 Total 3,000 4,800 7,300 13,300 13,300 23,300 23,300 Wafer Xi'an, Shaanxi 2,000 3,150 3,800 3,800 3,800 3,800 Wuxi, Jiangsu 3,000 3,850 4,400 4,400 4,400 4,400 Yinchuan, Ningxia 5,000 5,000 5,000 5,000 Kuching, Malaysia 500 1,000 1,000 1,000 1,000 Chuxiong, Yunnan 10,000 10,000 Total 3,000 5,000 7,500 14,200 14,200 24,200 24,200 Cell Kuching, Malaysia 500 500 500 500 India 500 500 500 Hefei 500 500 500 500 500 Taizhou 2,000 2,000 2,000 2,000 2,000 Total 2,500 3,000 3,500 3,500 3,500 Mono-Si module Lerri Solar 5,000 5,500 7,000 9,000 10,000 - Yinchuan 500 500 500 500 500 - Zhejiang 2,500 2,500 2,500 2,500 2,500 - Taizhou 2,000 2,000 2,000 2,000 2,000 - Xi'an 500 500 500 500 - Others 1,500 3,500 4,500 Kuching, Malaysia 500 500 500 500 India 500 500 500 Total 200 1,500 5,000 6,000 8,000 10,000 11,000 Source: Company data, Credit Suisse estimates

China Solar Sector 48 30 June 2017

Asia Pacific/China Electric Utilities

GCL-Poly Energy Holdings Ltd

(3800.HK / 3800 HK) Rating NEUTRAL [V] Price (28-Jun-17, HK$) 0.80 INITIATION Target price (HK$) 0.80 Upside/downside (%) 0.0

Mkt cap (HK$/US$ mn) 14,870 / 1,906 Clouded outlook Enterprise value (Rmb mn) 45,282 Number of shares (mn) 18,588 ■ Uncertain demand outlook. We initiate coverage on GCL Poly (GCL) with a Free float (%) 69.1 NEUTRAL rating and a sum-of-the-parts (SOTP) based target price of 52-wk price range (HK$) 1.19-0.73 ADTO-6M (US$ mn) 11.2 HK$0.80. As the largest multi-silicon (multi-Si) wafer producer in the world, Target price is for 12 months. we expect an uncertain demand outlook for GCL due to market share [V] = Stock Considered Volatile (see Disclosure Appendix) expansion of the competing mono-Si technology. We expect largely flat EPS

Research Analysts in FY17-19E, with increasing earnings from solar farm offset by profit decline

Gary Zhou, CFA at the wafer manufacturing business. 852 2101 6648 [email protected] ■ Multi-Si wafer may face pressure. GCL had a 17GW wafer production Dave Dai, CFA capacity in 2016, which needed polysilicon input of ~84kt (higher than GCL's 852 2101 7358 own poly production capacity of 70kt). As a result, poly is more of a [email protected] procurement cost for GCL, while wafer is the end product and its price is Gloria Yan essential for the company's total profit. After the significant wafer ASP drop 852 2101 7369 [email protected] (down 27%) in 1Q17 and uncertain 2H17 demand outlook, we expect the overall profit contribution (poly + wafer) from GCL's solar manufacturing business to decline by >30% in 2017E. ■ Poly is relatively more defensive. In April 2017, GCL announced its plan to build 40-60kt polysilicon production facilities in Xinjiang in 2018-20. Xinjiang has significant cost advantage in electricity (compared to its existing capacity in Jiangsu) which accounts for ~35% of poly production cost. However, we are also concerned that further expansion by peers in Xinjiang may pose a threat, as GCL may have only ~35% Xinjiang capacity exposure by FY18E. ■ Valuation. Our TP of HK$0.80 is derived from (1) HK$0.55 for poly and wafer manufacturing business valued at 6x FY18E P/E; and (2) HK$0.25 for its solar farm assets valued by a DCF model. The stock is trading at 7x FY18E P/E, close to its historical low while we expect valuation to remain depressed due to an uncertain outlook. Key risks are: (1) changes in multi-silicon wafer prices; and (2) market share competition between multi-silicon and mono- silicon solar components. Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (Rmb mn) 22,024.5 21,446.5 22,152.2 22,155.1 EBITDA (Rmb mn) 9,482.2 8,520.0 8,656.6 8,901.1 EBIT (Rmb mn) 6,051.0 4,932.3 4,909.7 5,059.5 Net profit (Rmb mn) 2,029.4 1,848.1 1,805.1 1,818.3 EPS (CS adj.) (Rmb) 0.11 0.10 0.10 0.10 Change from previous EPS (%) n.a. - - - Consensus EPS (Rmb) n.a. 0.10 0.11 0.12 EPS growth (%) (28.9) (9.9) (2.3) 0.7 The price relative chart measures performance against the P/E (x) 6.3 7.0 7.2 7.1 MSCI CHINA F IDX which closed at 7,479.41 on 28/06/17. Dividend yield (%) 0.0 0.0 0.0 0.0 On 28/06/17 the spot exchange rate was HK$7.8/US$1 EV/EBITDA (x) 4.6 5.5 6.1 5.7

Performance 1M 3M 12M P/B (x) 0.62 0.57 0.53 0.49 Absolute (%) -4.8 -23.8 -20.0 ROE (%) 11.1 8.5 7.7 7.2 Relative (%) -5.8 -33.3 -56.3 Net debt/equity (%) 131.4 132.5 144.4 126.6

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 49 30 June 2017

Focus charts

Figure 96: GCL—polysilicon ASP, unit cost, unit profit and gross margin trend Figure 97: GCL and Daqo—unit profit comparison (US$/kg) (US$/kg) 28% 25.0 27% 30% 25.0 26% 26% ASP Unit cost Unit profit 23% 20.0 25% 20.0 20% 15.0 15.0 13% 15% 10.0 10.0 7.7 10% 6.1 6.1 5.0 6.1 4.1 3.5 3.5 3.8 3.4 5% 5.0 2.1 2.1 3.2 0.0 0% 0.0 2014 2015 2016 2017E 2018E 2019E 2014 2015 2016 2014 2015 2016 2014 2015 2016 Gross margin ASP Unit cost Unit gross profit GCL Daqo

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

Figure 99: GCL—breakdown of solar manufacturing Figure 98: GCL—poly production cost gross profit (US$/kg) (Rmbm) 15.5 16.0 6,000 13.5 14.0 5,000 11.5 10.5 Gross profit ultimately 12.0 10.0 GCL - Xinjiang 4,000 9.7 9.48.8 from wafer production 10.0 8.2 7.8 GCL - Jiangsu 3,000 8.0 Gross profit ultimately 2,000 6.0 GCL - blended from poly production 4.0 1,000 2.0 0 0.0 2015 2016 2017E 2018E 2019E 2014 2015 2016 2017E 2018E 2019E Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

Figure 101: GCL—core earnings and dividends Figure 100: GCL—total gross profit breakdown history (Rmbm) (Rmbm) 8,000 4,000 60% 50% 7,000 3,000 6,000 40% 2,000 30% 5,000 Solar materials 20% 4,000 1,000 10% Solar farm 3,000 0 0% 2,000 2009 2010 2011 2012 2013 2014 2015 2016 2017E2018E2019E -10% -1,000 1,000 -20% - -2,000 -30% 2014 2015 2016 2017E 2018E 2019E Core earnings Dividends Payout ratio

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

Figure 102: GCL—one-year forward P/E history Figure 103: GCL—one-year forward P/B history (x) (x) 25 2.5 23.8x Avg+2SD 20 2.0 2.1x Avg+2SD 18.1x Avg+1SD 1.7x Avg+1SD 15 1.5 12.4x Avg 1.2x Avg 10 1.0 6.8x Avg-1SD 0.8x Avg-1SD 5 0.5 0.4x Avg-2SD 0 1.1x Avg-2SD 0.0 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Source: Bloomberg, Company data, Credit Suisse estimates Source: Bloomberg, Company data, Credit Suisse estimates

China Solar Sector 50 30 June 2017

GCL-Poly Energy Holdings Ltd (3800.HK / 3800 HK) Price (28 Jun 2017): HK$0.80; Rating: NEUTRAL [V]; Target Price: HK$0.80; Analyst: Gary Zhou Income Statement (Rmb mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 22,025 21,446 22,152 22,155 GCL-Poly Energy Holdings Limited is one of the world’s largest solar Cost of goods sold 14,980 15,194 15,530 15,069 photovoltaic enterprises, specializing in polysilicon and wafer EBITDA 9,482 8,520 8,657 8,901 manufacturing. It also possesses several large-scale solar farms EBIT 6,051 4,932 4,910 5,059 through its subsidiary named GCL New Energy. Net interest expense/(inc.) 2,149 2,210 2,258 2,306 Recurring PBT 3,935 2,756 2,685 2,787 Blue/Grey Sky Scenario Profit after tax 2,195 2,235 2,151 2,205 Reported net profit 2,029 1,848 1,805 1,818 Net profit (Credit Suisse) 2,029 1,848 1,805 1,818 Balance Sheet (Rmb mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 12,189 11,962 7,819 8,963 Current receivables 12,552 12,230 12,624 12,625 Inventories 957 957 957 957 Other current assets 1,414 1,414 1,414 1,414 Current assets 27,112 26,563 22,814 23,959 Property, plant & equip. 52,462 58,033 60,160 60,017 Investments 3,036 3,036 3,036 3,036 Intangibles 302 302 302 302 Other non-current assets 4,107 4,107 4,107 4,107 Total assets 87,019 92,042 90,419 91,421 Current liabilities 34,401 33,721 28,043 27,750 Total liabilities 63,625 66,413 62,638 61,436 Shareholders' equity 20,821 22,669 24,474 26,292 Minority interests 2,573 2,960 3,306 3,692 Total liabilities & equity 87,019 92,042 90,419 91,421 Cash Flow (Rmb mn) 12/16A 12/17E 12/18E 12/19E EBIT 6,051 4,932 4,910 5,059 Net interest (2,149) (2,210) (2,258) (2,306) Tax paid (537) (520) (534) (582) Working capital 7,290 (358) (6,071) (295) Other cash & non-cash items (2,870) 4,089 3,684 3,966 Our Blue Sky Scenario (HK$) 1.00 Operating cash flow 7,785 5,933 (270) 5,843 Average wafer price to drop by 5% in 2018 Capex (14,243) (9,159) (5,873) (3,698) Free cash flow to the firm (6,458) (3,227) (6,143) 2,144 Our Grey Sky Scenario (HK$) 0.60 Investing cash flow (8,150) (9,159) (5,873) (3,698) Average wafer price to drop by 15% in 2018 Equity raised 260 0 0 0 Dividends paid 0 0 0 0 Share price performance Financing cash flow (1,232) 3,000 2,000 (1,000) Total cash flow (1,597) (227) (4,143) 1,144 Adjustments 296 0 0 0 Net change in cash (1,302) (227) (4,143) 1,144 Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 18,393 18,588 18,588 18,588 EPS (Credit Suisse) (Rmb) 0.11 0.10 0.10 0.10 DPS (Rmb) 0.00 0.00 0.00 0.00 Operating CFPS (Rmb) 0.42 0.32 (0.01) 0.31 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue 1.2 (2.6) 3.3 0.0 EBIT 19.4 (18.5) (0.5) 3.1 EPS (28.9) (9.9) (2.3) 0.7 Margins (%) The price relative chart measures performance against the MSCI CHINA F IDX EBITDA 43.1 39.7 39.1 40.2 which closed at 7,476.27 on 28-Jun-2017 EBIT 27.5 23.0 22.2 22.8 On 28-Jun-2017 the spot exchange rate was HK$7.8/US$1 Valuation (x) 12/16A 12/17E 12/18E 12/19E P/E 6.3 7.0 7.2 7.1 P/B 0.62 0.57 0.53 0.49 Dividend yield (%) 0.0 0.0 0.0 0.0 EV/sales 2.0 2.2 2.4 2.3 EV/EBITDA 4.6 5.5 6.1 5.7 EV/EBIT 7.2 9.5 10.8 10.1 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 11.1 8.5 7.7 7.2 ROIC 10.5 7.0 6.2 5.9 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) 131.4 132.5 144.4 126.6 Net debt/EBITDA (x) 3.24 3.99 4.63 4.27

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 51 30 June 2017

Clouded outlook We initiate coverage on GCL Poly (GCL) with a NEUTRAL rating and a sum-of-the-parts (SOTP) based target price of HK$0.80. As the largest multi-silicon (multi-Si) wafer producer in the world, we expect an uncertain demand outlook for GCL due to market share expansion of the competing mono-Si technology. We expect largely flat EPS in FY17-19E, with increasing earnings from solar farm offset by profit decline at the wafer manufacturing business. Our TP of HK$0.80 is derived from (1) HK$0.55 for poly and wafer manufacturing business valued at 6x FY18E P/E; and (2) HK$0.25 for its solar farm assets valued by a DCF model. The stock is trading at 7x FY18E P/E, close to the historical low while we expect valuation to remain depressed due to an uncertain outlook. Key risks are: (1) changes in multi-silicon wafer prices; and (2) market share competition between multi-silicon and mono-silicon solar components.

Figure 104: GCL—SOTP target price calculation SOTP valuation Per share (HK$) Valuation method Key assumptions Solar farm assets 0.25 DCF Through GCLNE (62% owned) Wafer and poly and others 0.55 P/E 6x FY18E P/E TP 0.80 Source: Credit Suisse estimates

Figure 105: GCL—key operating metrics 2014 2015 2016 2017E 2018E 2019E 2020E Polysilicon Year-end capacity (tonne) 65,000 70,000 70,000 75,000 115,000 115,000 115,000 Output (t) 66,876 74,358 69,345 75,038 87,000 115,000 115,000 External sales (t) 15,443 18,023 9,951 9,951 18,089 26,125 23,875 Internal used (t) 51,433 56,335 59,394 65,087 68,912 88,875 91,125 ASP (US$/kg) 21.6 15.6 15.0 14.3 13.1 12.1 11.5 - YoY 25% -28% -4% -5% -8% -8% -5% Unit cost (US$/kg) 15.5 13.5 11.5 10.5 9.7 8.8 8.5 - YoY -9% -13% -15% -9% -7% -9% -4% Gross margin 28% 13% 23% 26% 26% 27% 26% Unit margin (US$/kg) 6.1 2.1 3.5 3.8 3.4 3.2 2.9 Wafer Year-end capacity (MW) 13,000 15,000 17,000 19,000 19,500 20,000 20,500 - Mono wafer (MW) 0 0 1,000 1,500 2,000 2,500 3,000 - Multi wafer (MW) 13,000 15,000 16,000 17,500 17,500 17,500 17,500 Output (MW) 12,909 15,178 17,518 18,000 19,250 19,750 20,250 ASP (US$/W) 0.215 0.188 0.164 0.133 0.120 0.108 0.100 - YoY 2% -13% -13% -19% -10% -10% -7% Wafer cost (US$/W) 0.175 0.150 0.129 0.118 0.107 0.099 0.093 - YoY -8% -14% -14% -8% -9% -8% -6% Gross margin 19% 20% 22% 11% 11% 8% 7% Unit profit (US$/W) 0.040 0.038 0.035 0.015 0.013 0.009 0.007 Gross profit breakdown by process (Rmb mn) Poly 2,533 967 1,611 1,927 2,022 2,583 2,350 Wafer 3,103 3,394 3,993 1,895 1,727 1,347 1,131 Poly (%) 45% 22% 29% 50% 54% 66% 68% Wafer (%) 55% 78% 71% 50% 46% 34% 32% Source: Company data, Credit Suisse estimates

China Solar Sector 52 30 June 2017

Profit capped by weak multi-Si wafer price According to the CS Global Solar Research Team, the multi-Si wafer oversupply may worsen in 2017E, with global wafer capacity growing 15% to 94GW, much higher than the global demand (~6% annual growth in the next few years). This is evidenced by the large spot price drop in multi-Si wafer by 30% YoY in 1Q17. Mono-Si wafer (LONGi is the largest producer), on the other hand, has a more stable price (largely flat YTD) due to stronger demand (such as Top Runner solar projects in China).

Figure 106: Wafer spot price history—mono-Si vs multi-Si (US$/W) 2.00 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17

Source: Bloomberg

Figure 107: Global wafer production capacity forecasts (GW) Company Country 2010 2011 2012 2013 2014 2015 2016 2017E 2018E GCL China 3,500 6,500 8,000 10,000 13,000 15,000 17,000 17,500 18,000 LONGi China 300 650 1,000 1,600 3,000 5,000 7,500 14,200 14,200 Yingli Green Energy China 1,000 1,700 2,450 3,185 2,450 2,450 2,450 2,450 2,450 JinkoSolar China 600 1,200 1,200 2,000 2,500 3,000 5,000 7,000 7,560 Tongwei Solar (acquired LDK) China 3,000 4,000 4,800 4,800 3,200 3,200 3,200 3,200 3,200 Green Energy Technology Taiwan 500 500 1,634 2,274 2,000 3,000 3,000 3,000 3,000 Renesola China 1,310 2,000 2,000 2,000 2,000 2,400 2,900 2,900 2,900 Trina Solar China 750 1,200 1,200 1,400 2,200 2,300 2,300 2,300 2,300 JA Solar China 300 600 1,000 1,000 1,000 1,000 2,500 3,000 3,000 Others 20,323 26,755 27,073 29,523 31,633 33,333 34,403 39,639 42,193 Total 31,583 45,105 50,744 57,781 62,982 69,183 81,753 93,989 97,103 - YoY 66% 43% 13% 14% 9% 10% 18% 15% 3% Source: Company data, CS Global Solar Research team, Credit Suisse estimates

GCL had a 17GW wafer production capacity in 2016, which needed poly input of ~84kt (higher than GCL's own poly production capacity of 70kt). As a result, poly is more of a procurement cost for GCL, while wafer is the end product and its price is essential for the whole company's profitability. Given its significant wafer ASP drop (down 27% YoY in 1Q17), we expect the overall profit contribution (poly + wafer) from GCL's solar manufacturing business to decline by >30% in 2017E. If we use market price for the poly input cost of GCL's wafer business, we estimate a meaningful profit shift from wafer to poly, which is also evidenced by the company's subsidiary 1Q17 results (GCL-Poly Suzhou New Energy, a major wafer production subsidiary of GCL, reported a 92% net profit drop in 1Q17). Market share loss due to expanding mono-Si Further market share gain of mono-Si would be negative to multi-Si wafer producers such as GCL. ITRPV expects mono-Si demand (including both P-type and N-type) to account for 44% of total market by 2020. Latest research (for example, from PV tech) suggests

China Solar Sector 53 30 June 2017

even stronger demand for mono-Si, representing 49% of total cell demand in 2018, and becoming the dominant technology by 2019. Based on our industry research and taking the average estimates from various consulting agencies and industry associations, we expect mono-Si demand to expand from 25% in 2016 to 30%/40%/45%/50% in FY17-20E.

Figure 108: Market demand breakdown by mono-Si and multi-Si

100% 18% 25% 30% 80% 40% 45% 50% 60% Mono-Si 40% 82% 75% 70% Multi-Si 60% 55% 20% 50%

0% 2015 2016 2017E 2018E 2019E 2020E

Source: PV tech, China PV Industry Association, ITRPV, Credit Suisse estimates Diamond wiring technology to be the saviour? On the cost side, we estimate unit wafer production cost (including both poly and non-poly costs) to drop 8% in FY17E (company guidance at 6-8% decline), mainly helped by new diamond-wiring wafer cutting technology which may represent around half of its wafer capacity by end-2017. According to the company, diamond wiring may reduce wafer cost by Rmb0.5-0.8 per piece (or 15-20%). Based on the current run rate (~30% converted), we expect around half of GCL's wafer production lines to be converted to diamond-wire cutting, with the full conversion likely happening in 2017. The key challenge is mainly at the downstream cell makers, as wafer cut by the diamond wiring method requires further treatment (such as creating a black silicon surface) which leads to extra costs. Only small investment in mono-Si Despite the uncertain outlook for mono-Si wafer, its price has been more stable (up 3% YTD) due to increasing demand (such as Top Runner solar projects in China). In May 2015, GCL set up a mono-Si wafer manufacturing plant in Ningxia, and started producing qualified mono-Si wafer in 2016. It currently has around 1GW production capacity (representing only ~6% of its total capacity), but has the potential to expand to 10GW, according to the company. We believe this could provide a strategic hedge for GCL if the market share gain of mono-Si products turns out to be quicker than expected. That said, the expansion plan for GCL's Ningxia plant is still uncertain as the company has not provided a specific timeline. Gross profit shift from wafer to poly As a result, we expect the gross profit from poly production to continue to grow, driven by stable price outlook and further cost reduction. The wafer business could see >50% profit drop in 2017E and we expect further deterioration in the next few years given the unsolved oversupply issue for multi-Si wafer. We expect poly production to contribute ~70% of GCL's total solar manufacturing business by 2019E (up from 30% in 2016).

China Solar Sector 54 30 June 2017

Figure 109: GCL—solar manufacturing business gross profit breakdown (Rmbm) 6,000 5,000 4,000 Gross profit ultimately from wafer production 3,000 Gross profit ultimately 2,000 from poly production 1,000 0 2015 2016 2017E 2018E 2019E

Source: Company data, Credit Suisse estimates Poly is more defensive but watch out for low-cost capacity expansion Along the solar manufacturing value chain, polysilicon production is one of the very few that still enjoys a relatively decent and stable gross margin, thanks to its capacity shortage in China (44% of poly was imported). Despite large price volatilities distorted by rush installations, GCL's poly ASP dropped only slightly by 4% in 2016, much slower than its unit cost decline of ~15% (based on our estimate), which drove gross margin recovery from 13% to 23%. The poly price rally continued in 1Q17 when GCL recorded 24% YoY ASP hike. Although price may turn weaker in 2H17 after the rush order demand in 1H17 (already down 15% since the peak this year), we estimate the full-year price drop to be mild at 5%. Going forward, we expect the gross margin for GCL's poly production to be stable at 26-27% in the next three years, mainly supported by further cost savings (especially due to low-cost new plants in Xinjiang). Unit gross profit may still drop from US$3.5/kg in 2016 to US$3.2/kg in 2019E, which should be more than offset by strong poly capacity expansion (+65%).

Figure 110: Spot poly price in China Figure 111: Polysilicon supply in China (US$/kg) (kt) 30.0 350 25.0 300 20.0 250 Import 15.0 200 150 Domestic 10.0 production 5.0 100 50 0.0 2013-01-01 2014-01-01 2015-01-01 2016-01-01 2017-01-01 0 2009 2010 2011 2012 2013 2014 2015 2016 Poly - import price (ex. tax) Poly - domestic price (incl. tax)

Source: WIND Source: WIND

Figure 112: China's market share in global solar component supply chain Polysilicon Wafer Cell Module China's output (2016E) 194k tons 63GW 49GW 45.8GW As % of global production 47.8% 79.6% 66.0% 69.1% Note: * 2015 data. Source: China PV Industry Association

China Solar Sector 55 30 June 2017

Figure 113: GCL—polysilicon ASP, unit cost, and gross margin history (US$/kg) 28% 25.0 26% 26% 27% 30% 23% 20.0 25% 20% 15.0 13% 15% 10.0 10% 5.0 6.1 3.5 3.8 3.4 5% 2.1 3.2 0.0 0% 2014 2015 2016 2017E 2018E 2019E Gross margin ASP Unit cost Unit gross profit

Source: Company data, Credit Suisse estimates More balanced supply and demand According to the CS Global Solar Research team, the total global poly production capacity may grow slightly by 6.5% to 475kt in 2017E. After deducting demand from semiconductor industry (~8%) and assuming 5g/W poly consumption for solar wafer production, average poly production capacity may meet 87GW solar demand, which is not significantly higher than the global demand forecasts by some leading consulting firms for 2017E (such as 85GW by GTM Research, 78GW by IHS Markit, etc.). Relatively higher unit capex also creates a higher entry barrier for the poly industry.

Figure 114: Global polysilicon production capacity forecasts (tonne) Company 2010 2011 2012 2013 2014 2015 2016 2017E 2018E GCL Silicon 21,000 46,000 65,000 65,000 65,000 70,000 70,000 75,000 115,000 Wacker Polysilicon 31,130 32,000 52,000 52,000 52,000 56,000 80,000 80,000 80,000 OCI 17,000 35,583 42,000 42,000 42,000 52,000 52,000 52,000 52,000 REC Silicon 17,500 17,500 17,500 17,500 20,000 20,000 20,000 30,000 40,000 Hemlock Semiconductor 36,000 36,000 36,000 32,000 32,000 32,000 32,000 32,000 32,000 Tokuyama (Malaysia acquired by OCI) 8,200 9,200 9,200 17,200 22,000 22,200 25,200 28,200 28,200 SunEdison 12,500 15,000 9,000 9,000 19,000 22,500 22,500 22,500 22,500 Daqo 3,300 4,300 4,300 5,000 6,150 12,150 12,150 18,000 18,000 Others 61,760 75,560 87,560 95,560 116,560 121,560 132,560 132,560 157,560 Total polysilicon year-end capacity 208,390 290,143 322,560 335,260 374,710 413,410 446,410 475,260 540,260 % growth YoY 51.9% 39.2% 11.2% 3.9% 11.8% 10.3% 8.0% 6.5% 13.7% Source: Company data, CS Global Solar Research team, Credit Suisse estimates

Figure 115: Poly—higher entry barrier with higher unit capex requirement Unit capex Poly Wafer (incl. ingot) Cell Module (Rmb/W) 0.75 0.65 0.35 0.11 Source: China PV Industry Association, Credit Suisse estimates

Poly remains the only solar subsector that Chinese companies do not have dominant market share in (48% vs. 66-80% for wafer, cell and module), and further production shift to China (especially to Xinjiang where poly cash cost of ~US$7-8/kg is the lowest among the world) is likely, in our view. The aggressive capacity expansion plans of some Chinese players, if fully executed, may be a concern over the long run (such as East Hope with 120kt planned, Sichuan Yongxiang with 70kt, etc.).

China Solar Sector 56 30 June 2017

Figure 116: Cash cost of major global polysilicon producers (US$/kg) 20 18 16 14 12 10 8 6 4 2

0

OCI(Korea)

(Korea)

China)

China)

Hemlock(US)

China)

China)

RECSilicon (US)

Elkem(Norway)

Renesola(China)

Hanwha Chemical

Wacker (Germany)

Yongxiang(Sichuan,

Sino-Silicon(Henan,

Dun'An (I.M.,China)

Asia-Silicon(Qinghai,

Tokuyama (Malaysia)

Other Chinese(China)

XinteEnergy(Xinjiang,

DAQO(Xinjiang, China)

Wacker Tennessee(US)

Hankook Silicon(Korea) GCL-Poly(Jiangsu, China)

Source: Company data, PV News, Credit Suisse estimates GCL's Xinjiang expansion In early April 2017, GCL announced its plan to build a 60kt polysilicon production plant in Xinjiang Province, including 40kt new-built facilities (20kt to be completed by 2Q18 and 20kt by end-2018) and 20kt of existing Xuzhou (in Jiangsu Province) facilities to be relocated to Xinjiang (by end-2020). We believe it is the right strategic move for GCL, as Xinjiang has a significant cost advantage in electricity which accounts for ~35% of GCL's poly production cost.

Figure 117: GCL—Poly capacity expansion plan in Xinjiang Province Phase Capacity (kt) Commission date Note I 20 2Q18 New-built II 20 End-2018 New-built III 20 End-2020 Relocation of existing Jiangsu capacity Source: Company data

Figure 118: GCL—poly production cost breakdown (2016) 15% Raw material (TCS) 28% 6% Electricity 2% Labour Steam 9% Maintenance

6% Others Depreciation

35%

Source: Company data, Credit Suisse estimates

Poly manufacturers (such as Daqo New Energy and Xinte Energy) in Xinjiang have enjoyed quite decent gross profit margin (35% on average) in the past three years. We compare the ASP of GCL and Daqo and there has been no significant difference. However, Daqo's unit cost was roughly US$1.8-2.2/kg lower than GCL (based on our calculation), which has resulted in Daqo's higher unit gross profit (US$6.1/kg in 2016 vs. GCL at US$3.5/kg). We believe most of the cost difference should come from electricity cost. Due to higher seaborne coal price in Jiangsu (Rmb590/t, 5000kcal, in April 2017) vs. cheap mine-mouth coal in Xinjiang (Rmb183/t), we estimate the electricity cost would be around US$2/kg higher in Jiangsu.

China Solar Sector 57 30 June 2017

Figure 120: Daqo New Energy—poly ASP/unit cost/ Figure 119: GCL and Daqo—unit profit comparison gross margin trend (US$/kg) (US$/kg) 25.0 25.00 10.0 Unit cost Unit profit 8.5 8.3 ASP 7.9 7.8 20.0 20.00 7.1 7.2 7.0 8.0

15.0 15.00 5.3 5.0 6.0 3.8 4.1 4.1 10.0 7.7 10.00 3.0 4.0 6.1 6.1 4.1 3.5 5.0 2.1 5.00 2.0 0.0 0.00 0.0 2014 2015 2016 2014 2015 2016 2014 2015 2016 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15 1Q16 2Q16 3Q16 4Q16 1Q17 GCL Daqo Unit gross profit ASP Unit cost

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

With the new Xinjiang poly plant to be commissioned in 2018E, we expect 7-9% per year drop (assuming ~5% annual cost drop for existing capacity) in GCL's blended poly production cost over 2018-19E. By 2019, Xinjiang will account for 35% of GCL's production capacity, while the rest (in Jiangsu) may not be as cost competitive as Xinjiang given higher electricity costs. However, as the largest poly producer in the world, GCL still has strong technology knowhow in this area and may continue to drive down cost through technology upgrades.

Figure 121: GCL—poly production cost (US$/kg) 15.5 16.0 13.5 14.0 11.5 10.5 12.0 10.0 GCL - Xinjiang 9.7 9.48.8 10.0 8.2 7.8 8.0 GCL - Jiangsu 6.0 GCL - blended 4.0 2.0 0.0

2014 2015 2016 2017E 2018E 2019E Source: Company data, Credit Suisse estimates Solar farm remains a growth driver Despite the uncertainty over its solar manufacturing business, we expect GCL's solar farm operations (mainly through its 62% stake holding in GCL New Energy [0451.HK]) to maintain a high growth rate at 25% gross profit CAGR over FY16-19E, and its percentage contribution to total gross profit may also increase from 25% in 2016 to >40% in FY17- 19E. We expect GCL New Energy (GCLNE) to add 2GW new capacity (close to the high end of management guidance of 1.5-2GW) and at the same time dispose/sell the majority stake of 1GW existing capacity. Considering GCLNE's highly leveraged balance sheet (261% net gearing in 2016), we believe such a Build and Transfer (BT) business model is likely to drive higher earnings growth as the disposal immediately realises future earnings and cash flows. It also avoids the renewables subsidies delay issue which has been a major problem for private solar farm companies with high borrowing cost.

China Solar Sector 58 30 June 2017

Figure 122: GCL—total gross profit breakdown (Rmbm) 8,000 7,000 6,000 5,000 Solar materials 4,000 3,000 Solar farm 2,000 1,000 - 2014 2015 2016 2017E 2018E 2019E

Source: Company data, Credit Suisse estimates Valuation The stock is currently trading at an undemanding valuation of 8x FY18E P/E, close to the past-five-year low, but we expect further derating given a deteriorating margins outlook.

Figure 123: GCL—one-year forward P/E history Figure 124: GCL—one-year forward P/B history (x) (x) 25 2.5 23.8x Avg+2SD 20 2.0 2.1x Avg+2SD 18.1x Avg+1SD 1.7x Avg+1SD 15 1.5 12.4x Avg 1.2x Avg 10 1.0 6.8x Avg-1SD 0.8x Avg-1SD 5 0.5 0.4x Avg-2SD 0 1.1x Avg-2SD 0.0 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17

Source: Bloomberg, Company data, Credit Suisse estimates Source: Bloomberg, Company data, Credit Suisse estimates

GCL's earnings have been bumpy in the past eight years since the injection of the solar manufacturing business in 2009, as a result of a volatile solar component market. The company has paid dividends only in three years (2010, 2011 and 2015), and the latest one (in 2015) was a special dividend after the disposal of the non-solar power generation business. Considering the challenging multi-Si wafer market and likely no dividends from its solar farm subsidiary (GCLNE) due to a stretched balance sheet, we have assumed no dividend payout from GCL in the next three years.

China Solar Sector 59 30 June 2017

Figure 125: GCL—core earnings and dividends history (Rmbm) 4,000 60% 50% 3,000 40% 2,000 30% 20% 1,000 10% 0 0% 2009 2010 2011 2012 2013 2014 2015 2016 2017E2018E2019E -10% -1,000 -20% -2,000 -30% Core earnings Dividends Payout ratio

Source: Company data, Credit Suisse estimates Key risks The largest risk for GCL is the wafer price, as we estimate every 1% change may impact FY17E EPS by 6.8% and FY18E EPS by 8.1%. In our base case, we assume 19% YoY wafer price drop in 2017E and 10% drop in 2018E. As a result, earnings visibility could be low in the medium-to-long term given the wafer price volatilities.

Figure 126: FY17/18E EPS sensitivity to wafer ASP Wafer price -3% -2% -1% Base case +1% +2% +3% FY17 EPS change -20.4% -13.6% -6.8% 0.0% 6.8% 13.6% 20.4% FY18 EPS change -24.3% -16.2% -8.1% 0.0% 8.1% 16.2% 24.3% Source: Company data, Credit Suisse estimates

The major upside risk for GCL may come from the cost side. The company is currently modifying its wafer production facilities using diamond-wire cutting technology, which could possibly lower its production cost by 15-20%. A faster-than-expected adoption of diamond-wiring technology may further lower costs, but it would still depend on the demand from downstream cell makers. One percent lower-than-expected wafer cost may increase FY17E EPS by 6.0% and FY18E EPS by 7.5%. In our base case, we assume 8- 9% annual wafer cost decline.

Figure 127: FY17/18E EPS sensitivity to wafer cost Wafer cost -3% -2% -1% Base case +1% +2% +3% FY17 EPS change 18.0% 12.0% 6.0% 0.0% -6.0% -12.0% -18.0% FY18 EPS change 22.5% 15.0% 7.5% 0.0% -7.5% -15.0% -22.5% Source: Company data, Credit Suisse estimates

GCL has a relatively high net gearing ratio (~130% in FY17E), partly due to its consolidated solar farm business through GCLNE (net gearing at >300%). As a result, we estimate a 20 bp interest rate hike may impact its FY17E EPS by 5%, which would be a concern especially given the tighter domestic bank lending in 2017.

Figure 128: FY17/18E EPS sensitivity to interest rate Average cost of 6.3% 5.7% 5.9% 6.1% 6.5% 6.7% 6.9% borrowing (base case) FY17 EPS change 14.7% 9.8% 4.9% 0.0% -4.9% -9.8% -14.7% FY18 EPS change 11.1% 7.4% 3.7% 0.0% -3.7% -7.4% -11.1% Source: Company data, Credit Suisse estimates

China Solar Sector 60 30 June 2017

Company background GCL Poly was listed on the Hong Kong Stock Exchange in Nov-2007, originally with the thermal power generation business. In July 2009, the polysilicon production assets (Jiangsu Zhongneng) were injected into the listco, which is now the main business of GCL. Its founder and chairman, Mr. Zhu Gongshan, has a 34% effective stake (including his family) in GCL through Asia Pacific Energy Fund. GCL was one of the three listcos within Mr. Zhu's solar business, the other two being GCL New Energy (GCL has 62% stake) and GCL System Integration Technology (listed in A share, mainly engaged in module manufacturing and solar system sales).

Figure 129: Shareholding structure of GCL Poly

Zhu Gongshan (Chairman) and his family 100% JP Morgan Chase & Haitong Int'l Other public Asia Pacific Energy Fund Templeton Co. Securities shareholders 34.27% 7.74% 10.09% 6.13% 41.77% GCL Poly (3800.HK) 62.28% 22.4% GCL System Integration GCL New Energy Technology (002506.SZ) (0451.HK)

Source: Company data

Figure 130: Key events Date Event Nov-2007 Listed on Hong Kong Stock Exchange Jul-2009 Injected Jiangsu Zhongneng (polysilicon production) at HK$26.3bn; treated as reserve acquisition. May-2014 Acquired SAME TIME (0451.HK) as a separate listing platform for GCL's solar farm business. Sep-2015 Sold non-solar power generation business at Rmb3.2bn Apr-2017 Purchased some solar materials assets from SunEdison Apr 2017 Announced poly production capacity expansion plan in Xinjiang. Source: Company data

Figure 131: GCL—management profiles Name Title Background Mr. ZHU Gongshan Chairman Aged 59. The founder of the company. He has been an Executive Director since July 2006. He is currently a member of the 12th National Committee of the Chinese People’s Political Consultative Conference, and the co-chairman of Asian Photovoltaic Industry Association. Mr. Zhu is also the Honorary Chairman and an executive director of GCL New Energy (0451.HK). Mr. ZHU Zhanjun Chief Executive Mr. ZHU Zhanjun, aged 47, has been an Executive Director of the company since January 2015 and has been the Chief Executive Officer Officer of the company since April 2016. He joined the company in 2004 as the plant manager of one of its power plants and transferred to Jiangsu Zhongneng Polysilicon Technology Development Co. (a subsidiary of the company) in 2008. Mr. Zhu is now responsible for the overall operation and management of the polysilicon and wafer business of the company. Mr. YEUNG Man Chief Financial Mr. YEUNG Man Chung, Charles, aged 49, has been an Executive Director of the company since September 2014. He was appointed Chung, Charles Officer as the Chief Financial Officer of the company on 30 April 2014. Prior to that, he served as partner of Deloitte Touche Tohmatsu and was a part-time member of the Central Policy Unit of the Government of Hong Kong Special Administrative Region. When he left Deloitte Touche Tohmatsu in March 2014, he was the head of corporate finance advisory services, Southern China. Mr. Yeung has over 20 years of experience in accounting, auditing and financial management. He is responsible for the financial control and reporting, corporate finance, tax and risk management of the company and its subsidiaries. Source: Company data.

China Solar Sector 61 30 June 2017

Asia Pacific/China Electric Utilities

Xinyi Solar Holdings (0968.HK / 968 HK) Rating (from OUTPERFORM) NEUTRAL Price (28-Jun-17, HK$) 2.24 DOWNGRADE RATING Target price (HK$) (from 3.87) 2.40 Upside/downside (%) 7.1

Mkt cap (HK$/US$ mn) 16,629 / 2,132 Pressures for both solar glass and solar farms Enterprise value (HK$ mn) 23,934 Number of shares (mn) 7,424 ■ Downgrade to NEUTRAL. We expect XYS' solar glass business to face Free float (%) 37.9 near-term pressure in 2H17 with potentially lower selling price due to 52-wk price range (HK$) 3.37-2.24 ADTO-6M (US$ mn) 4.6 oversupply (>20%). Despite rush installation demand in 1H17, the solar glass Target price is for 12 months. price continued to fall, declining 5-7% YTD (worse than our expectation).

Besides, the future growth of its solar farm segment may also be impacted by Research Analysts

less ground-mounted project quota. Downgrade to NEUTRAL. Gary Zhou, CFA 852 2101 6648 ■ Deteriorating oversupply for solar glass. We expect solar glass over- [email protected] supply to deteriorate in 2H17, due to likely lower demand after solar farm Dave Dai, CFA tariff cut and further capacity addition from peers (~2 kt/day or 10% of current 852 2101 7358 [email protected] capacity, implying >20% oversupply for the industry). We cut our FY17-19E Gloria Yan solar price assumptions 7-19% to Rmb22-26 per sq m. Despite ~50% time- 852 2101 7369 weighted capacity growth for XYS in FY17E driven by newly commissioned [email protected] production lines, we expect the profit contribution to drop 17%, mainly due to the large decline in unit gross profit. We expect FY18-19E gross margin for solar glass to drop to 16-22%, close to the historical low (19% in 2012). ■ Less project quota for solar farms. We expect overall ground-mounted solar quota to decline going forward from the peak in 2016-17. XYS was selective in solar farm investment (IRR at ~15%) but we are concerned that expanding project bidding is likely to reduce returns. XYS guided 600-800MW new solar farm addition in 2017E (down from 1GW in 2016) and we expect further slowdown to 500MW post-2018. ■ Valuation. We cut our FY17-19E EPS by 14-31% due to both lower solar glass prices and lower return from new solar farm projects. Downgrade to NEUTRAL with our TP reduced to HK$2.4 (from HK$3.87). Our valuation includes HK$1.3 for solar farm (DCF model), HK$0.9 for solar glass (7x FY18E P/E) and HK$0.2 for others (EPC, etc.). The stock is currently trading at 1.7x FY17E P/B, much lower than its past-three-year average of 2.5x but still one of the highest among China peers. Key risks are uncertainty in solar glass prices and solar farm project quota. Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (HK$ mn) 6,007.1 7,870.4 8,791.3 8,937.2 EBITDA (HK$ mn) 2,892.1 3,385.1 3,677.6 3,709.7 EBIT (HK$ mn) 2,440.1 2,819.7 2,985.7 2,925.1 Net profit (HK$ mn) 1,985.6 2,180.7 2,238.2 2,118.5 EPS (CS adj.) (HK$) 0.29 0.32 0.33 0.31 Change from previous EPS (%) n.a. (14.4) (20.7) (31.4) Consensus EPS (HK$) n.a. 0.33 0.38 0.39 EPS growth (%) 58.8 9.8 2.6 (5.3) The price relative chart measures performance against the P/E (x) 7.6 6.9 6.8 7.1 MSCI CHINA F IDX which closed at 7,479.41 on 28/06/17. Dividend yield (%) 6.3 6.2 5.9 5.6 On 28/06/17 the spot exchange rate was HK$7.8/US$1 EV/EBITDA (x) 7.8 7.5 7.1 7.1

Performance 1M 3M 12M P/B (x) 2.43 2.03 1.72 1.50 Absolute (%) -7.1 -9.8 -24.8 ROE (%) 33.2 31.9 27.5 22.4 Relative (%) -8.1 -19.2 -61.1 Net debt/equity (%) 78.4 98.9 90.2 78.5

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 62 30 June 2017

Xinyi Solar Holdings (0968.HK / 968 HK) Price (28 Jun 2017): HK$2.24; Rating: (from OUTPERFORM) NEUTRAL; Target Price: (from HK$3.87) HK$2.40; Analyst: Gary Zhou Income Statement (HK$ mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 6,007 7,870 8,791 8,937 The company has been a cost leader in its major businesses, and Cost of goods sold 3,257 4,745 5,507 5,718 we like its defensive profile with cash-generating solar glass EBITDA 2,892 3,385 3,678 3,710 business supporting high-capex high-return solar farm investment. EBIT 2,440 2,820 2,986 2,925 Net interest expense/(inc.) 95 161 186 196 Blue/Grey Sky Scenario Recurring PBT 2,390 2,747 2,888 2,817 Profit after tax 2,150 2,459 2,587 2,497 Reported net profit 1,986 2,181 2,238 2,119 Net profit (Credit Suisse) 1,986 2,181 2,238 2,119 Balance Sheet (HK$ mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 843 1,820 2,067 2,021 Current receivables 3,261 5,351 6,699 7,179 Inventories 288 443 501 499 Other current assets 212 212 212 212 Current assets 4,604 7,825 9,480 9,910 Property, plant & equip. 11,079 14,400 15,951 17,304 Investments 392 411 432 453 Intangibles 327 327 327 327 Other non-current assets 385 504 563 572 Total assets 16,786 23,467 26,752 28,567 Current liabilities 4,591 6,643 7,693 7,985 Total liabilities 9,359 14,519 16,112 16,277 Shareholders' equity 6,216 7,459 8,802 10,073 Minority interests 1,212 1,490 1,839 2,217 Total liabilities & equity 16,786 23,467 26,752 28,567 Cash Flow (HK$ mn) 12/16A 12/17E 12/18E 12/19E EBIT 2,440 2,820 2,986 2,925 Net interest (95) (161) (186) (196) Tax paid (241) (289) (301) (320) Working capital (1,333) (1,085) (813) (312) Our Blue Sky Scenario (HK$) 2.60 Other cash & non-cash items (246) (132) (6) 87 Our blue sky scenario assumes solar glass price to drop by 3% in Operating cash flow 526 1,153 1,679 2,183 2018. Capex (3,774) (2,752) (1,732) (1,653) Free cash flow to the firm (3,249) (1,599) (53) 530 Our Grey Sky Scenario (HK$) 2.20 Investing cash flow (3,802) (2,780) (1,759) (1,680) Our grey sky scenario assumes solar glass price to drop by 13% in Equity raised 0 0 0 0 2018. Dividends paid (580) (950) (938) (895) Financing cash flow 1,337 2,755 505 (361) Share price performance Total cash flow (1,939) 1,129 425 141 Adjustments 0 0 0 0 Net change in cash (1,939) 1,129 425 141 Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 6,749 6,749 6,749 6,749 EPS (Credit Suisse) (HK$) 0.29 0.32 0.33 0.31 DPS (HK$) 0.14 0.14 0.13 0.13 Operating CFPS (HK$) 0.08 0.17 0.25 0.32 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue 26.5 31.0 11.7 1.7 EBIT 80.9 15.6 5.9 (2.0) EPS 58.8 9.8 2.6 (5.3) Margins (%) EBITDA 48.1 43.0 41.8 41.5 The price relative chart measures performance against the MSCI CHINA F IDX EBIT 40.6 35.8 34.0 32.7 which closed at 7,479.41 on 28-Jun-2017 Valuation (x) 12/16A 12/17E 12/18E 12/19E On 28-Jun-2017 the spot exchange rate was HK$7.8/US$1 P/E 7.6 6.9 6.8 7.1 P/B 2.43 2.03 1.72 1.50 Dividend yield (%) 6.3 6.2 5.9 5.6 EV/sales 3.7 3.2 3.0 2.9 EV/EBITDA 7.8 7.5 7.1 7.1 EV/EBIT 9.2 9.0 8.8 9.0 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 33.2 31.9 27.5 22.4 ROIC 21.0 16.3 14.1 12.3 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) 78.4 98.9 90.2 78.5 Net debt/EBITDA (x) 2.01 2.61 2.61 2.60

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 63 30 June 2017

Asia Pacific/China Electric Utilities

GCL New Energy Holdings (0451.HK / 451 HK) Rating NEUTRAL [V] Price (28-Jun-17, HK$) 0.38 DECREASE TARGET PRICE Target price (HK$) (from 0.50) 0.40 Upside/downside (%) 3.9

Mkt cap (HK$/US$ mn) 7,343 / 941.30 Multiple uncertainties ahead Enterprise value (Rmb mn) 26,321 Number of shares (mn) 19,074 ■ Maintain NEUTRAL. We expect multiple uncertainties for GCLNE in the next Free float (%) 34.0 few years, due to possible reduction in ground-mounted solar project quota 52-wk price range (HK$) 0.52-0.32 ADTO-6M (US$ mn) 0.7 and expanding low-tariff project bidding. Its high net gearing (>300%) is also a Target price is for 12 months. concern, especially given tightening bank lending. Valuation (FY18E P/B at [V] = Stock Considered Volatile (see Disclosure Appendix) 1.0x) does not seem attractive compared with leading renewable peers

Research Analysts trading below book. Maintain NEUTRAL.

Gary Zhou, CFA ■ Deleveraging takes time. Investors have been concerned about GCLNE's 852 2101 6648 [email protected] stretched balance sheet. Despite the planned solar farm disposal (we Dave Dai, CFA estimate 1GW for FY17E, with 50% probability assigned), net gearing may 852 2101 7358 still go up to 320% (from 261% in FY16) given the company's aggressive [email protected] capacity addition plan (~2GW) for this year. We believe solar farm deconsolidation is the right strategy for GCLNE, but deleveraging could take years and therefore the possibility of starting dividend payout still seems remote. ■ More cautious on future growth. Going forward, we cut our post-2018 net annual capacity additions for GCLNE from 0.7GW to 0.5GW, as we are now turning more cautious on China's overall ground-mounted solar project quota (accounting for 97% of GCLNE's existing capacity). Possible expansion into new project types (such as distributed solar and poverty-alleviation projects) could be an upside risk but their contribution is yet to ramp up (accounting for ~15% and 5-10% of planned new capacity additions in 2017E). ■ Valuation. We revise down our FY18-19E EPS estimates by 22-28% mainly due to reduced net capacity additions and lower tariffs with expanding project bidding, and our TP was cut to HK$0.40 (from HK$0.50). Its FY18E P/B of 1.0x is close to a historical low but not attractive compared with leading renewables peers such as Longyuan Power and Huaneng Renewables (0.8x). Key upside risks are (1) uncertainty in competitive project bidding; (2) interest rate changes. Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (Rmb mn) 2,246.4 3,593.3 4,185.8 4,537.7 EBITDA (Rmb mn) 2,018.0 3,000.0 3,528.4 3,885.0 EBIT (Rmb mn) 1,295.9 2,148.4 2,553.2 2,853.4 Net profit (Rmb mn) 299.0 822.1 717.7 814.0 EPS (CS adj.) (Rmb) 0.02 0.04 0.04 0.04 Change from previous EPS (%) n.a. (0.1) (28.1) (22.4) Consensus EPS (Rmb) n.a. 0.03 0.05 0.07 EPS growth (%) 195.9 168.3 (12.7) 13.4 The price relative chart measures performance against the P/E (x) 20.9 7.8 8.9 7.9 MSCI CHINA F IDX which closed at 7,479.41 on 28/06/17. Dividend yield (%) 0.0 0.0 0.0 0.0 On 28/06/17 the spot exchange rate was HK$7.8/US$1 EV/EBITDA (x) 11.5 9.9 10.6 10.1

Performance 1M 3M 12M P/B (x) 1.37 1.19 1.05 0.92 Absolute (%) 13.2 -10.5 14.9 ROE (%) 8.5 16.5 12.5 12.5 Relative (%) 12.2 -19.9 -21.3 Net debt/equity (%) 261.4 319.9 388.8 372.3

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 64 30 June 2017

GCL New Energy Holdings (0451.HK / 451 HK) Price (28 Jun 2017): HK$0.38; Rating: NEUTRAL [V]; Target Price: (from HK$0.50) HK$0.40; Analyst: Gary Zhou Income Statement (Rmb mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 2,246 3,593 4,186 4,538 GCL New Energy is a pure solar farm developer in China, with Cost of goods sold 676 1,033 1,183 1,252 2.7GW installed solar power capacity in 1H16 (ranked second in EBITDA 2,018 3,000 3,528 3,885 China). Despite the short history, the company’s asset portfolio is EBIT 1,296 2,148 2,553 2,853 growing fast with 1GW new solar farm capacity added in 2015. Net interest expense/(inc.) 966 1,464 1,781 1,943 Recurring PBT 331 685 773 912 Blue/Grey Sky Scenario Profit after tax 309 959 855 982 Reported net profit 299 822 718 814 Net profit (Credit Suisse) 299 822 718 814 Balance Sheet (Rmb mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 5,855 5,552 4,844 5,218 Current receivables 3,730 4,292 4,616 4,832 Inventories 0 0 0 0 Other current assets 1,154 1,154 1,154 1,154 Current assets 10,739 10,998 10,614 11,203 Property, plant & equip. 26,755 32,104 34,042 35,750 Investments 152 152 152 152 Intangibles 0 0 0 0 Other non-current assets 3,832 4,478 5,069 5,393 Total assets 41,478 47,731 49,877 52,498 Current liabilities 18,017 17,443 11,866 11,667 Total liabilities 35,059 40,484 41,907 43,708 Shareholders' equity 4,573 5,395 6,113 6,927 Minority interests 47 52 58 63 Total liabilities & equity 41,478 47,731 49,877 52,498 Cash Flow (Rmb mn) 12/16A 12/17E 12/18E 12/19E EBIT 1,296 2,148 2,553 2,853 Net interest (940) (1,438) (1,755) (1,917) Tax paid 42 (126) (118) (130) Working capital 3,920 (1,235) (6,002) (517) Other cash & non-cash items (2,187) 449 427 720 Our Blue Sky Scenario (HK$) 0.50 Operating cash flow 2,131 (201) (4,895) 1,010 Our blue sky scenario assumes net annual capacity additions at Capex (13,284) (6,200) (2,914) (2,739) 1GW after 2018. Free cash flow to the firm (11,153) (6,401) (7,809) (1,729) Investing cash flow (13,284) (6,200) (2,914) (2,739) Our Grey Sky Scenario (HK$) 0.30 Equity raised 1,833 0 0 0 Our grey sky scenario assumes no net annual capacity additions Dividends paid 0 0 0 0 after 2018. Financing cash flow 11,181 6,099 7,101 2,103 Total cash flow 29 (303) (709) 374 Share price performance Adjustments 0 0 0 0 Net change in cash 29 (303) (709) 374 Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 18,616 19,074 19,074 19,074 EPS (Credit Suisse) (Rmb) 0.02 0.04 0.04 0.04 DPS (Rmb) 0.00 0.00 0.00 0.00 Operating CFPS (Rmb) 0.11 (0.02) (0.26) 0.04 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue 226.5 60.0 16.5 8.4 EBIT 314.6 65.8 18.8 11.8 EPS 195.9 168.3 (12.7) 13.4 Margins (%) EBITDA 89.8 83.5 84.3 85.6 EBIT 57.7 59.8 61.0 62.9 The price relative chart measures performance against the MSCI CHINA F IDX Valuation (x) 12/16A 12/17E 12/18E 12/19E which closed at 7,479.41 on 28-Jun-2017 P/E 20.9 7.8 8.9 7.9 On 28-Jun-2017 the spot exchange rate was HK$7.8/US$1 P/B 1.37 1.19 1.05 0.92 Dividend yield (%) 0.0 0.0 0.0 0.0 EV/sales 10.3 8.2 8.9 8.6 EV/EBITDA 11.5 9.9 10.6 10.1 EV/EBIT 17.9 13.8 14.6 13.7 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 8.5 16.5 12.5 12.5 ROIC 8.0 6.5 6.2 6.1 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) 261.4 319.9 388.8 372.3 Net debt/EBITDA (x) 8.32 7.73 8.78 8.42

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 65 30 June 2017

Asia Pacific/China Electric Utilities

Jiangsu Linyang Energy (601222.SS / 601222 CH) Rating (from UNDERPERFORM) NEUTRAL [V] Price (28-Jun-17, Rmb) 7.34 UPGRADE RATING Target price (Rmb) (from 6.80) 7.50 Upside/downside (%) 2.2

Mkt cap (Rmb/US$ mn) 12,948 / 1,904 Improving outlook for distributed solar Enterprise value (Rmb mn) 12,900 Number of shares (mn) 1,764 ■ Upgrade to NEUTRAL. Impacted by less power meter tendering and margin Free float (%) 49.0 pressure, the share price of LYE has been weak since late 2016. Although its 52-wk price range (Rmb) 12.11-6.46 ADTO-6M (US$ mn) 9.7 traditional power meter business is yet to see the inflection point, we expect Target price is for 12 months. earnings growth from its distributed solar business may accelerate in the next [V] = Stock Considered Volatile (see Disclosure Appendix) three years (CAGR at 30%) driven by strong policy support. Upgrade to

Research Analysts NEUTRAL.

Gary Zhou, CFA ■ Distributed solar to drive growth. In 1Q17, Linyang recorded 45% core 852 2101 6648 [email protected] earnings growth, which we believe was mainly due to stronger earnings from Dave Dai, CFA distributed solar. By 2016, distributed solar accounted for 85% of Linyang's 852 2101 7358 total solar capacity (931MW). Unlike ground-mounted project facing return [email protected] pressure from tariff cuts and project bidding, the return for distributed solar Gloria Yan has been intact (IRR>15%), supported by generous tariff subsidy 852 2101 7369 [email protected] (Rmb0.42/kWh unchanged) and healthy cash flows (no subsidy delay issue). We expect THE solar segment to account for ~70% of Linyang's gross profit by FY18-19E (43% in 2016). ■ Power meter business may still be under pressure. The company's weak FY16 results (down 4% YoY) were mainly dragged by both revenue decline (-8%) and margin contraction (from 36% to 31%) for its power meter business. After a 28% drop in power meter tender volume from State Grid in 2016, we expect a further drop in 2017 as State Grid reduces the number of tenders from three to two. However, expanding export sales and more orders from China Southern Grid may help reduce the pressure. ■ Valuation. We cut our FY17E EPS by 9% mainly due to lower sales and margin for power meter business, while FY18-19E EPS is raised by 8-13% with higher earnings from distributed solar business, and therefore our TP is revised up to Rmb7.5 (from Rmb6.8). The stock is currently trading at a FY18E P/E of 15x, slightly lower than its five-year historical average (18x). Key risks are: (1) uncertainty in power meter orders; and (2) change of government policy on distributed solar (such as tariff subsidies, etc.). Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (Rmb mn) 3,093.9 3,247.0 5,536.1 5,850.7 EBITDA (Rmb mn) 728.0 1,248.9 1,589.5 1,810.6 EBIT (Rmb mn) 527.0 848.3 1,032.9 1,108.7 Net profit (Rmb mn) 474.4 731.9 829.1 845.0 EPS (CS adj.) (Rmb) 0.28 0.42 0.48 0.48 Change from previous EPS (%) n.a. (9.2) 7.7 12.9 Consensus EPS (Rmb) n.a. 0.41 0.50 0.62 EPS growth (%) (20.0) 50.0 13.3 1.9 The price relative chart measures performance against the P/E (x) 26.2 17.5 15.4 15.1 Shanghai Shenzhen CSI300 index which closed at Dividend yield (%) 1.9 2.9 3.3 3.4 3,674.72 on 28/06/17. On 28/06/17 the spot exchange rate EV/EBITDA (x) 16.0 11.4 10.0 9.7 was Rmb6.8/US$1 P/B (x) 1.53 1.51 1.44 1.37

Performance 1M 3M 12M ROE (%) 7.2 8.8 9.5 9.3 Absolute (%) 9.1 -1.9 -35.4 Net debt/equity (%) Net cash 14.9 33.0 49.2

Relative (%) 3.5 -8.8 -52.5 Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 66 30 June 2017

Jiangsu Linyang Energy (601222.SS / 601222 CH) Price (28 Jun 2017): Rmb7.34; Rating: (from UNDERPERFORM) NEUTRAL [V]; Target Price: (from Rmb6.80) Rmb7.50; Analyst: Gary Zhou Income Statement (Rmb mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 3,094 3,247 5,536 5,851 We expect Linyang's future growth to be driven by solar operating Cost of goods sold 2,153 1,988 3,887 4,091 business, but partly offset by weak power meter business and EBITDA 728 1,249 1,589 1,811 declining ASP for its EPC business. EBIT 527 848 1,033 1,109 Net interest expense/(inc.) 28 83 168 238 Blue/Grey Sky Scenario Recurring PBT 534 816 925 943 Profit after tax 500 772 874 891 Reported net profit 474 732 829 845 Net profit (Credit Suisse) 474 732 829 845 Balance Sheet (Rmb mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 2,775 2,660 2,482 2,847 Current receivables 1,682 1,764 2,992 3,161 Inventories 299 276 540 569 Other current assets 706 704 733 736 Current assets 5,463 5,404 6,748 7,313 Property, plant & equip. 5,739 8,457 10,803 12,805 Investments 71 71 71 71 Intangibles 63 68 74 79 Other non-current assets 1,318 1,321 1,366 1,372 Total assets 12,653 15,321 19,061 21,639 Current liabilities 3,013 2,855 4,727 4,929 Total liabilities 4,377 6,719 10,091 12,292 Shareholders' equity 8,120 8,486 8,900 9,322 Minority interests 156 116 71 25 Total liabilities & equity 12,653 15,321 19,061 21,639 Cash Flow (Rmb mn) 12/16A 12/17E 12/18E 12/19E EBIT 527 848 1,033 1,109 Net interest 0 0 0 0 Tax paid 0 0 0 0 Working capital 781 (218) 305 (5) Our Blue Sky Scenario (Rmb) 8.00 Other cash & non-cash items 148 284 353 438 Our blue sky scenario assumes net profit margin for power meter Operating cash flow 1,457 915 1,691 1,542 sales at 15%. Capex (4,233) (3,125) (2,909) (2,708) Free cash flow to the firm (2,776) (2,210) (1,218) (1,166) Our Grey Sky Scenario (Rmb) 7.00 Investing cash flow (4,154) (3,125) (2,909) (2,708) Our grey sky scenario assumes net profit margin for power meter Equity raised 1,336 0 0 0 sales at 5%. Dividends paid (241) (366) (415) (423) Financing cash flow 4,040 2,094 1,040 1,531 Share price performance Total cash flow 1,342 (116) (178) 365 Adjustments 0 0 0 0 Net change in cash 1,342 (116) (178) 365 Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 1,694 1,743 1,743 1,743 EPS (Credit Suisse) (Rmb) 0.28 0.42 0.48 0.48 DPS (Rmb) 0.14 0.21 0.24 0.25 Operating CFPS (Rmb) 0.86 0.52 0.97 0.88 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue 14.1 4.9 70.5 5.7 EBIT (2.7) 61.0 21.8 7.3 EPS (20.0) 50.0 13.3 1.9 Margins (%) EBITDA 23.5 38.5 28.7 30.9 The price relative chart measures performance against the Shanghai EBIT 17.0 26.1 18.7 18.9 Shenzhen CSI300 index which closed at 3,674.72 on 28-Jun-2017 Valuation (x) 12/16A 12/17E 12/18E 12/19E On 28-Jun-2017 the spot exchange rate was Rmb6.8/US$1 P/E 26.2 17.5 15.4 15.1 P/B 1.53 1.51 1.44 1.37 Dividend yield (%) 1.9 2.9 3.3 3.4 EV/sales 3.8 4.4 2.9 3.0 EV/EBITDA 16.0 11.4 10.0 9.7 EV/EBIT 22.0 16.8 15.4 15.8 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 7.2 8.8 9.5 9.3 ROIC 8.5 9.1 8.5 7.6 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) (16.1) 14.9 33.0 49.2 Net debt/EBITDA (x) (1.83) 1.03 1.86 2.54

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 67 30 June 2017

Asia Pacific/China Electric Utilities

Concord New Energy (0182.HK / 182 HK) Rating OUTPERFORM Price (28-Jun-17, HK$) 0.33 DECREASE TARGET PRICE Target price (HK$) (from 0.56) 0.47 Upside/downside (%) 42.4

Mkt cap (HK$/US$ mn) 2,863 / 367.03 Solar in pain, but wind in recovery Enterprise value (Rmb mn) 5,055 Number of shares (mn) 8,677 ■ Maintain OUTPERFORM. We cut Concord New Energy’s (CNE) FY17-18E Free float (%) 61.1 EPS by 23-38% to reflect lower earnings contribution from solar farms. Given 52-wk price range (HK$) 0.50-0.32 ADTO-6M (US$ mn) 1.0 worsening solar curtailment (9% in FY16 vs 2% in FY15) and Target price is for 12 months. -22% YoY output decline in 1Q17, we lower our solar utilisation forecast. We

stay positive on wind with curtailment improvement a key trend. Our DCF- Research Analysts

based target price is trimmed to HK$0.47. Maintain OUTPERFORM. Note that Gloria Yan CNE changed its reporting currency from HKD to CNY from the FY16 results. 852 2101 7369 [email protected] ■ More pressure from solar. CNE's solar utilisation was under pressure in Dave Dai, CFA FY16 (-8% YoY) due to worsening curtailment in key locations, such as 852 2101 7358 [email protected] Shaanxi and Yunnan (average curtailment: 9% vs. 2% in FY15). Given the still depressed 1Q17 solar output (declined by -22% YoY), we cut our FY17-18E solar utilisation forecast to 1,400-1,430 hours (from 1,553 hours). We expect that solar utilisation recovery may require more time based on the YTD run- rate. Solar tariff also saw a 5% YoY decline in FY16 mainly due to the lower on-grid tariff for new projects and expanding direct supply in Yunnan. ■ Wind utilisation recovery the key catalyst. Despite utilisation pressure for solar, CNE's wind utilisation is seeing gradual improvement. It delivered 40% YoY wind attributable output growth in 1Q17, achieving 26% of our full-year forecast. Given the capacity expansion of 27% in FY16, we believe the solid output growth was mainly attributable to further utilisation improvement. The company is shifting its location mix with most of the new capacities located in non-curtailment locations. Given the strong policy support (such as the minimum utilisation scheme) and State Grid's cooperation, we expect CNE's wind utilisation could recover to 1,800 hours in FY17E (+6% YoY). ■ Attractive valuation. The stock is trading at an appealing valuation of 0.5x FY17E P/B (the lowest within the utilities universe). Dividend payout expanded to ~30% in FY16 and the company expects a similar payout ratio in the next few years. We believe risk-reward for CNE is attractive with potential utilisation recovery and BT profit expansion as the key catalysts. Share price performance Financial and valuation metrics

Year 12/16A 12/17E 12/18E 12/19E Revenue (Rmb mn) 1,785.2 1,937.6 2,149.7 2,365.3 EBITDA (Rmb mn) 393.0 420.9 519.9 581.1 EBIT (Rmb mn) 393.0 420.9 519.9 581.1 Net profit (Rmb mn) 458.1 541.6 659.0 722.0 EPS (CS adj.) (Rmb) 0.05 0.06 0.08 0.08 Change from previous EPS (%) n.a. (23.0) (26.0) (38.1) Consensus EPS (Rmb) n.a. 0.07 0.09 0.09 EPS growth (%) (7.1) 18.2 21.7 9.6 The price relative chart measures performance against the P/E (x) 5.4 4.6 3.8 3.4 MSCI CHINA F IDX which closed at 7,479.41 on 28/06/17. Dividend yield (%) 6.1 6.6 8.0 8.8 On 28/06/17 the spot exchange rate was HK$7.8/US$1 EV/EBITDA (x) 10.3 14.5 15.5 16.9

Performance 1M 3M 12M P/B (x) 0.49 0.46 0.42 0.39 Absolute (%) -8.3 -16.5 -20.5 ROE (%) 8.6 10.3 11.7 11.8 Relative (%) -9.4 -25.9 -56.8 Net debt/equity (%) 29.4 64.9 92.0 112.6

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 68 30 June 2017

Concord New Energy (0182.HK / 182 HK) Price (28 Jun 2017): HK$0.33; Rating: OUTPERFORM; Target Price: (from HK$0.56) HK$0.47; Analyst: Gloria Yan Income Statement (Rmb mn) 12/16A 12/17E 12/18E 12/19E Company Background Sales revenue 1,785 1,938 2,150 2,365 We expect CNE to deliver 22% FY16-18E EPS CAGR driven by: (1) Cost of goods sold 1,274 1,395 1,505 1,656 balanced expansion of solar and wind farms; (2) wind utilisation EBITDA 393 421 520 581 recovery on policy support and (3) construction margin improvement EBIT 393 421 520 581 with business shift from EPC to BT model. Net interest expense/(inc.) 142 182 209 265 Recurring PBT 475 573 703 782 Blue/Grey Sky Scenario Profit after tax 462 545 663 726 Reported net profit 458 542 659 722 Net profit (Credit Suisse) 458 542 659 722 Balance Sheet (Rmb mn) 12/16A 12/17E 12/18E 12/19E Cash & cash equivalents 1,891 1,723 1,125 1,669 Current receivables 2,086 2,139 2,276 2,416 Inventories 81 81 81 81 Other current assets 2,665 2,665 2,665 2,665 Current assets 6,724 6,609 6,148 6,832 Property, plant & equip. 4,492 6,703 8,748 10,656 Investments 1,580 1,675 1,787 1,933 Intangibles 1,068 1,068 1,068 1,068 Other non-current assets 522 564 623 683 Total assets 14,385 16,618 18,374 21,171 Current liabilities 5,788 5,819 5,863 5,907 Total liabilities 9,114 11,034 12,354 14,674 Shareholders' equity 5,070 5,425 5,857 6,331 Minority interests 155 159 163 167 Total liabilities & equity 14,339 16,618 18,374 21,171 Cash Flow (Rmb mn) 12/16A 12/17E 12/18E 12/19E EBIT 393 421 520 581 Net interest 0 0 0 0 Tax paid (13) (28) (40) (56) Working capital (954) 128 (100) (132) Other cash & non-cash items 200 58 71 55 Our Blue Sky Scenario (HK$) 0.51 Operating cash flow (374) 579 451 448 Our blue sky scenario assumes new solar capacity at 100MW per Capex (4,862) (4,890) (5,235) (5,600) year. Free cash flow to the firm (5,235) (4,311) (4,784) (5,152) Investing cash flow (4,862) (2,490) (2,835) (2,800) Our Grey Sky Scenario (HK$) 0.43 Equity raised (2) 0 0 0 Our grey sky scenario assumes no new solar capacity additions in Dividends paid (175) (187) (228) (249) the future. Financing cash flow 74 1,761 1,133 2,127 Total cash flow (5,162) (150) (1,251) (224) Share price performance Adjustments 0 0 1 2 Net change in cash (5,162) (150) (1,250) (222) Per share 12/16A 12/17E 12/18E 12/19E Shares (wtd avg.) (mn) 8,607 8,607 8,607 8,607 EPS (Credit Suisse) (Rmb) 0.05 0.06 0.08 0.08 DPS (Rmb) 0.02 0.02 0.02 0.03 Operating CFPS (Rmb) (0.04) 0.07 0.05 0.05 Earnings 12/16A 12/17E 12/18E 12/19E Growth (%) Sales revenue (58.7) 8.5 10.9 10.0 EBIT (31.0) 7.1 23.5 11.8 EPS (7.1) 18.2 21.7 9.6 Margins (%) EBITDA 22.0 21.7 24.2 24.6 EBIT 22.0 21.7 24.2 24.6 The price relative chart measures performance against the MSCI CHINA F IDX Valuation (x) 12/16A 12/17E 12/18E 12/19E which closed at 7,479.41 on 28-Jun-2017 P/E 5.4 4.6 3.8 3.4 On 28-Jun-2017 the spot exchange rate was HK$7.8/US$1 P/B 0.49 0.46 0.42 0.39 Dividend yield (%) 6.1 6.6 8.0 8.8 EV/sales 2.3 3.2 3.7 4.1 EV/EBITDA 10.3 14.5 15.5 16.9 EV/EBIT 10.3 14.5 15.5 16.9 ROE analysis (%) 12/16A 12/17E 12/18E 12/19E ROE 8.6 10.3 11.7 11.8 ROIC 5.3 5.0 4.7 4.2 Credit ratios 12/16A 12/17E 12/18E 12/19E Net debt/equity (%) 29.4 64.9 92.0 112.6 Net debt/EBITDA (x) 3.91 8.61 10.66 12.59

Source: Company data, Thomson Reuters, Credit Suisse estimates

China Solar Sector 69 30 June 2017

Companies Mentioned (Price as of 28-Jun-2017) Canadian Solar Inc (CSIQ.OQ, $15.98) Concord New Energy (0182.HK, HK$0.33, OUTPERFORM, TP HK$0.47) Daqo New Energy (DQ.N, $20.77) GCL New Energy Holdings (0451.HK, HK$0.38, NEUTRAL[V], TP HK$0.4) GCL-Poly Energy Holdings Ltd (3800.HK, HK$0.8, NEUTRAL[V], TP HK$0.8) JA Solar Holdings (JASO.OQ, $6.4) Jiangsu Linyang Energy (601222.SS, Rmb7.34, NEUTRAL[V], TP Rmb7.5) Jinko Solar (JKS.N, $20.63) LONGi Green Energy Technology (601012.SS, Rmb17.11, OUTPERFORM, TP Rmb23.0) OCI Company Ltd (010060.KS, W91,200) Tesla Motors Inc. (TSLA.OQ, $371.24) Wacker Chemie (WCHG.DE, €96.2) Xinyi Solar Holdings (0968.HK, HK$2.24, NEUTRAL, TP HK$2.4)

Disclosure Appendix Analyst Certification Gary Zhou, CFA, Dave Dai, CFA and Gloria Yan each certify, with respect to the companies or securities that the individual analyzes, that (1) the views expressed in this report accurately reflect his or her personal views about all of the subject companies and securities and (2) no part of his or her compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report.

3-Year Price and Rating History for Concord New Energy (0182.HK)

0182.HK Closing Price Target Price Date (HK$) (HK$) Rating 01-Nov-16 0.40 0.56 O * * Asterisk signifies initiation or assumption of coverage.

OUTPERFORM

3-Year Price and Rating History for GCL New Energy Holdings (0451.HK)

0451.HK Closing Price Target Price Date (HK$) (HK$) Rating 01-Nov-16 0.45 0.50 N * * Asterisk signifies initiation or assumption of coverage.

NEUTRAL

China Solar Sector 70 30 June 2017

3-Year Price and Rating History for GCL-Poly Energy Holdings Ltd (3800.HK)

3800.HK Closing Price Target Price Date (HK$) (HK$) Rating 29-Aug-14 2.56 2.84 O 01-Dec-14 1.85 2.29 04-Mar-15 1.89 2.20 20-Jul-15 1.58 1.92 31-Aug-15 1.14 1.28 N 08-Sep-15 1.17 1.28 O 29-Oct-15 1.47 2.01 16-Dec-15 1.14 1.08 U 24-Mar-16 1.27 1.45 N 12-Aug-16 1.10 NC OUTPERFORM NEUTRAL * Asterisk signifies initiation or assumption of coverage. UNDERPERFORM NOT COVERED Effective July 3, 2016, NC denotes termination of coverage.

3-Year Price and Rating History for Jiangsu Linyang Energy (601222.SS)

601222.SS Closing Price Target Price Date (Rmb) (Rmb) Rating 21-Oct-14 7.08 8.57 O 27-Apr-15 12.61 9.57 U 02-Jul-15 10.06 9.57 N 28-Oct-15 9.54 9.29 27-Apr-16 10.95 7.71 U 29-Aug-16 11.16 7.14 * 01-Nov-16 9.32 6.80 * * Asterisk signifies initiation or assumption of coverage.

OUTPERFORM UNDERPERFORM NEUTRAL

3-Year Price and Rating History for Xinyi Solar Holdings (0968.HK)

0968.HK Closing Price Target Price Date (HK$) (HK$) Rating 01-Nov-16 2.86 3.97 O * 01-Mar-17 2.69 3.87 * Asterisk signifies initiation or assumption of coverage.

OUTPERFORM

The analyst(s) responsible for preparing this research report received Compensation that is based upon various factors including Credit Suisse's total revenues, a portion of which are generated by Credit Suisse's investment banking activities As of December 10, 2012 Analysts’ stock rating are defined as follows: Outperform (O) : The stock’s total return is expected to outperform the relevant benchmark* over the next 12 months. Neutral (N) : The stock’s total return is expected to be in line with the relevant benchmark* over the next 12 months. Underperform (U) : The stock’s total return is expected to underperform the relevant benchmark* over the next 12 months. *Relevant benchmark by region: As of 10th December 2012, Japanese ratings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ra tings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin American and non-Japan Asia stocks, ratings are based on a stock’s total return relative to the average total return of the relevant country or regional benchmark; prior to 2nd October 2012 U.S. and Canadian ratings were based on (1) a stock’s absolute total return potential to its current share price and (2) the relative attractiveness of a stock’s total return potential within

China Solar Sector 71 30 June 2017 an analyst’s coverage universe. For Australian and New Zealand stocks, the expected total return (ETR) calculation in cludes 12-month rolling dividend yield. An Outperform rating is assigned where an ETR is greater than or equal to 7.5%; Underperform where an ETR less than or equal to 5%. A Neutral may be assigned where the ETR is between -5% and 15%. The overlapping rating range allows analysts to assign a rating that puts ETR in the context of associated risks. Prior to 18 May 2015, ETR ranges for Outperform and Underperform ratings did not overlap with Neutral thresholds between 15% and 7.5%, wh ich was in operation from 7 July 2011. Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances. Not Rated (NR) : Credit Suisse Equity Research does not have an investment rating or view on the stock or any other securities related to the company at this time. Not Covered (NC) : Credit Suisse Equity Research does not provide ongoing coverage of the company or offer an investment rating or investment view on the equity security of the company or related products. Volatility Indicator [V] : A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward. Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation: Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months. Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months. Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months. *An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cover multiple sectors. Credit Suisse's distribution of stock ratings (and banking clients) is:

Global Ratings Distribution Rating Versus universe (%) Of which banking clients (%) Outperform/Buy* 44% (65% banking clients) Neutral/Hold* 40% (60% banking clients) Underperform/Sell* 14% (54% banking clients) Restricted 2% *For purposes of the NYSE and FINRA ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, and Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdings, and other individual factors. Important Global Disclosures Credit Suisse’s research reports are made available to clients through our proprietary research portal on CS PLUS. Credit Suisse research products may also be made available through third-party vendors or alternate electronic means as a convenience. Certain research products are only made available through CS PLUS. The services provided by Credit Suisse’s analysts to clients may depend on a specific client’s preferences regarding the frequency and manner of receiving communications, the client’s risk profile and investment, the size and scope of the overall client relationship with the Firm, as well as legal and regulatory constraints. To access all of Credit Suisse’s research that you are entitled to receive in the most timely manner, please contact your sales representative or go to https://plus.credit-suisse.com . Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein. Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: https://www.credit- suisse.com/sites/disclaimers-ib/en/managing-conflicts.html . Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties.

Target Price and Rating Valuation Methodology and Risks: (12 months) for Concord New Energy (0182.HK) Method: Our target price of HK$0.47 and OUTPERFORM rating are based on a discounted cash-flow method assuming a 7.1% WACC and 4.5% cost of debt. We assume a risk-free rate of 3% and an equity risk premium of 9%. We assume no terminal value. We expect CNE to deliver 15% FY17-19E EPS CAGR driven by: (1) balanced expansion of solar and wind farms; (2) wind utilisation recovery on policy support; and (3) construction margin improvement with business shift from engineering procurement construction (EPC) to build-and- transfer (BT) model. Risk: Risks to our OUTPERFORM rating and HK$0.47 target price are larger-than-expected solar and wind curtailment, worse-than-expected tariff cut and lower-than-expected build-and-transfer earnings. Target Price and Rating Valuation Methodology and Risks: (12 months) for GCL New Energy Holdings (0451.HK) Method: Our DCF-based target price for GCL New Energy (GCLNE) is HK$0.40. We use a WACC of 7% (assuming a debt to equity ratio of 80:20). We are cautious on China solar farm operators going forward given declining ground-mounted project quota and expanding project bidding. Thus, we rate the stock NEUTRAL.

China Solar Sector 72 30 June 2017

Risk: The key risks to our NEUTRAL rating and TP of HK$0.40 are: (1) better- or worse-than-expected tariffs obtained at competitive project bidding; (2) higher- or lower-than-expected interest rate; and (3) better- or worse-than-expected disposal gain realised from its planned solar farm sales. Target Price and Rating Valuation Methodology and Risks: (12 months) for GCL-Poly Energy Holdings Ltd (3800.HK) Method: Our target price of HK$0.80 for GCL Poly is based on 6x FY17E P/E for its solar poly and wafer manufacturing business and DCF model for its solar farm assets. We have a NEUTRAL rating on the stock given near-term earnings pressure for multi-Si wafer and an uncertain long-term market share outlook. Risk: Key risks to our NEUTRAL rating and TP of HK$0.80 for GCL Poly are: (1) higher- or lower-than-expected multi-silicon wafer prices; (2) better- or worse-than-expected market share change of multi-silicon wafer; and (3) higher- or lower-than-expected interest rates. Target Price and Rating Valuation Methodology and Risks: (12 months) for Jiangsu Linyang Energy (601222.SS) Method: Our target price of Rmb7.5 for Jiangsu Linyang Energy is based on a sum-of-the-parts valuation. We assume 8x P/E for its power meter business and we use a discounted cash flow model to value its solar farm business. We rate the stock NEUTRAL given the ongoing pressure on power meter sales while its distributed solar business may help drive earnings growth. Risk: The key risks to our NEUTRAL rating and TP of Rmb7.5 for the company are: (1) better- or worse-than-expected power meter orders and gross margin; (2) export sales of power meters; (3) tariff policies on distributed solar; and (4) capacity growth of the company's distributed solar business. Target Price and Rating Valuation Methodology and Risks: (12 months) for LONGi Green Energy Technology (601012.SS) Method: Our target price of Rmb23.0 is derived from a DCF model, assuming a WACC of 9% (cost of equity at 12% and cost of debt at 5%) with no terminal growth. We have an OUTPERFORM rating on the stock and expect it to benefit from the expanding mono-Si solar component market. Risk: Key risks to our OUTPERFORM rating and TP of Rmb23.0 are: (1) lower-than-expected mono-Si wafer and module prices; (2) less-than- expected cost savings in the next few years; and (3) less-than-expected market share gain of mono-Si. Target Price and Rating Valuation Methodology and Risks: (12 months) for Xinyi Solar Holdings (0968.HK) Method: We have a NEUTRAL rating and target price of HK$2.4 for Xinyi Solar. We expect the company's solar glass manufacturing may be subject to near-term pressures due to industry oversupply, and the solar farm segment may also be constrained by uncertain ground- mounted solar project quota. Our target price is based on sum-of-the-parts valuation method. For solar farm, we use a discounted cash flow model with no terminal value. For the solar glass business, we value it at 7x FY18E P/E multiple. Risk: Risks to our NEUTRAL rating and target price of HK$2.4 for Xinyi Solar Holdings include: (1) higher- or lower-than-expected selling prices of solar glass; and (2) higher- or lower-than-expected annual solar farm capacity addition which depends on overall project quota and industry competition.

Please refer to the firm's disclosure website at https://rave.credit-suisse.com/disclosures/view/selectArchive for the definitions of abbreviations typically used in the target price method and risk sections. See the Companies Mentioned section for full company names Credit Suisse currently has, or had within the past 12 months, the following as investment banking client(s): 601012.SS, 3800.HK, 0451.HK, 0182.HK, CSIQ.OQ, WCHG.DE, JASO.OQ, TSLA.OQ, JKS.N Credit Suisse provided investment banking services to the subject company (0182.HK, CSIQ.OQ, WCHG.DE, TSLA.OQ, JKS.N) within the past 12 months. Credit Suisse currently has, or had within the past 12 months, the following issuer(s) as client(s), and the services provided were non-investment- banking, securities-related: CSIQ.OQ, JKS.N Credit Suisse has managed or co-managed a public offering of securities for the subject company (WCHG.DE, TSLA.OQ) within the past 12 months. Within the past 12 months, Credit Suisse has received compensation for investment banking services from the following issuer(s): 0182.HK, CSIQ.OQ, WCHG.DE, TSLA.OQ, JKS.N Credit Suisse expects to receive or intends to seek investment banking related compensation from the subject company (601012.SS, 3800.HK, 0451.HK, 0182.HK, CSIQ.OQ, WCHG.DE, JASO.OQ, TSLA.OQ, JKS.N) within the next 3 months. Within the last 12 months, Credit Suisse has received compensation for non-investment banking services or products from the following issuer(s): CSIQ.OQ, JKS.N As of the date of this report, Credit Suisse makes a market in the following subject companies (3800.HK, 0968.HK). A member of the Credit Suisse Group is party to an agreement with, or may have provided services set out in sections A and B of Annex I of Directive 2014/65/EU of the European Parliament and Council ("MiFID Services") to, the subject issuer (601012.SS, 601222.SS, 3800.HK, 0451.HK, 0968.HK, CSIQ.OQ, WCHG.DE, JASO.OQ, TSLA.OQ, JKS.N) within the past 12 months.

China Solar Sector 73 30 June 2017

As of the end of the preceding month, Credit Suisse beneficially own 1% or more of a class of common equity securities of (CSIQ.OQ). Credit Suisse has a material conflict of interest with the subject company (TSLA.OQ) . Credit Suisse is acting as financial advisor to SolarCity Corporation (SQTY.OQ) on their sale to Tesla Motors Inc (TSLA.OQ). For other important disclosures concerning companies featured in this report, including price charts, please visit the website at https://rave.credit- suisse.com/disclosures or call +1 (877) 291-2683. For date and time of production, dissemination and history of recommendation for the subject company(ies) featured in this report, disseminated within the past 12 months, please refer to the link: https://rave.credit-suisse.com/disclosures/view/report?i=307413&v=71jpjec4b1qsrftl3lpli8rzx . Important Regional Disclosures Singapore recipients should contact Credit Suisse AG, Singapore Branch for any matters arising from this research report. The analyst(s) involved in the preparation of this report may participate in events hosted by the subject company, including site visits. Credit Suisse does not accept or permit analysts to accept payment or reimbursement for travel expenses associated with these events. Restrictions on certain Canadian securities are indicated by the following abbreviations: NVS--Non-Voting shares; RVS--Restricted Voting Shares; SVS--Subordinate Voting Shares. Individuals receiving this report from a Canadian investment dealer that is not affiliated with Credit Suisse should be advised that this report may not contain regulatory disclosures the non-affiliated Canadian investment dealer would be required to make if this were its own report. For Credit Suisse Securities (Canada), Inc.'s policies and procedures regarding the dissemination of equity research, please visit https://www.credit- suisse.com/sites/disclaimers-ib/en/canada-research-policy.html. Credit Suisse has acted as lead manager or syndicate member in a public offering of securities for the subject company (WCHG.DE, TSLA.OQ) within the past 3 years. Principal is not guaranteed in the case of equities because equity prices are variable. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. This research report is authored by: Credit Suisse (Hong Kong) Limited ...... Gary Zhou, CFA ; Dave Dai, CFA ; Gloria Yan To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important disclosures regarding any non-U.S. analyst contributors: The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts listed below may not be associated persons of CSSU and therefore may not be subject to the FINRA 2241 and NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account. Credit Suisse (Hong Kong) Limited ...... Gary Zhou, CFA ; Dave Dai, CFA ; Gloria Yan Important disclosures regarding companies or other issuers that are the subject of this report are available on Credit Suisse’s disclosure website at https://rave.credit-suisse.com/disclosures or by calling +1 (877) 291-2683.

China Solar Sector 74 30 June 2017

This report is produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who-we-are This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates.The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk. This report is issued and distributed in European Union (except Switzerland): by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. Germany: Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). United States and Canada: Credit Suisse Securities (USA) LLC; Switzerland: Credit Suisse AG; Brazil: Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; Mexico: Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); Japan: by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau ( Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; Hong Kong: Credit Suisse (Hong Kong) Limited; Australia: Credit Suisse Equities (Australia) Limited; Thailand: Credit Suisse Securities (Thailand) Limited, regulated by the Office of the Securities and Exchange Commission, Thailand, having registered address at 990 Abdulrahim Place, 27th Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok10500, Thailand, Tel. +66 2614 6000; Malaysia: Credit Suisse Securities (Malaysia) Sdn Bhd; Singapore: Credit Suisse AG, Singapore Branch; India: Credit Suisse Securities (India) Private Limited (CIN no.U67120MH1996PTC104392) regulated by the Securities and Exchange Board of India as Research Analyst (registration no. INH 000001030) and as Stock Broker (registration no. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777; South Korea: Credit Suisse Securities (Europe) Limited, Seoul Branch; Taiwan: Credit Suisse AG Taipei Securities Branch; Indonesia: PT Credit Suisse Sekuritas Indonesia; Philippines:Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Additional Regional Disclaimers Hong Kong: Credit Suisse (Hong Kong) Limited ("CSHK") is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an Australian financial services licence (AFSL) and is exempt from the requirement to hold an AFSL under the Corporations Act 2001 (the Act) under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Act). Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Australia (to the extent services are offered in Australia): Credit Suisse Securities (Europe) Limited (“CSSEL”) and Credit Suisse International (“CSI”) are authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority (“FCA”) and the Prudential Regulation Authority under UK laws, which differ from Australian Laws. CSSEL and CSI do not hold an Australian Financial Services Licence (“AFSL”) and are exempt from the requirement to hold an AFSL under the Corporations Act (Cth) 2001 (“Corporations Act”) under Class Order 03/1099 published by the Australian Securities and Investments Commission (“ASIC”), in respect of the financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). This material is not for distribution to retail clients and is directed exclusively at Credit Suisse's professional clients and eligible counterparties as defined by the FCA, and wholesale clients as defined under section 761G of the Corporations Act. Credit Suisse (Hong Kong) Limited (“CSHK”) is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Credit Suisse Securities (USA) LLC (CSSU) and Credit Suisse Asset Management LLC (CSAM LLC) are licensed and regulated by the Securities Exchange Commission of the United States under the laws of the United States, which differ from Australian laws. CSSU and CSAM LLC do not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1100 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Malaysia: Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. Singapore: This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore Branch to overseas investors (as defined under the Financial Advisers Regulations). Credit Suisse AG, Singapore Branch may distribute reports produced by its foreign entities or affiliates pursuant to an arrangement under Regulation 32C of the Financial Advisers Regulations. Singapore recipients should contact Credit Suisse AG, Singapore Branch at +65-6212-2000 for matters arising from, or in connection with, this report. By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore Branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the “FAA”), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore Branch may provide to you. UAE: This information is being distributed by Credit Suisse AG (DIFC Branch), duly licensed and regulated by the Dubai Financial Services Authority (“DFSA”). Related financial services or products are only made available to Professional Clients or Market Counterparties, as defined by the DFSA, and are not intended for any other persons. Credit Suisse AG (DIFC Branch) is located on Level 9 East, The Gate Building, DIFC, Dubai, United Arab Emirates. EU: This report has been produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-US customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. US customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the US. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority or in respect of which the protections of the Prudential Regulation Authority and Financial Conduct Authority for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case are available upon request in respect of this report. CS may provide various services to US municipal entities or obligated persons ("municipalities"), including suggesting individual transactions or trades and entering into such transactions. Any services CS provides to municipalities are not viewed as "advice" within the meaning of Section 975 of the Dodd-Frank Wall Street Reform and Consumer Protection Act. CS is providing any such services and related information solely on an arm's length basis and not as an advisor or fiduciary to the municipality. In connection with the provision of the any such services, there is no agreement, direct or indirect, between any municipality (including the officials,management, employees or agents thereof) and CS for CS to provide advice to the municipality. Municipalities should consult with their financial, accounting and legal advisors regarding any such services provided by CS. In addition, CS is not acting for direct or indirect compensation to solicit the municipality on behalf of an unaffiliated broker, dealer, municipal securities dealer, municipal advisor, or investment adviser for the purpose of obtaining or retaining an engagement by the municipality for or in connection with Municipal Financial Products, the issuance of municipal securities, or of an investment adviser to provide investment advisory services to or on behalf of the municipality. If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice by Credit Suisse to the clients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any liability whatsoever for any direct or consequential loss arising from their use of this report or its content. Principal is not guaranteed. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Copyright © 2017 CREDIT SUISSE AG and/or its affiliates. All rights reserved. Investment principal on bonds can be eroded depending on sale price or market price. In addition, there are bonds on which investment principal can be eroded due to changes in redemption amounts. Care is required when investing in such instruments. When you purchase non-listed Japanese fixed income securities (Japanese government bonds, Japanese municipal bonds, Japanese government guaranteed bonds, Japanese corporate bonds) from CS as a seller, you will be requested to pay the purchase price only.

China Solar Sector 75