China Solar Equipment...world’s largest solar market in 2010. China soon took over and solar capacity grew from 19GW in 2013 to 175GW in 2018. But this led to overcapacity and a

www.research.hsbc.com

Disclosures & Disclaimer: This report must be read with the disclosures and the analyst certifications inthe Disclosure appendix, and with the Disclaimer, which forms part of it.

Equity Research Report

Sp

otligh

t: Ch

ina S

olar Equ

ipm

ent

Jun

e 2019Eq

uities // En

ergy Eq

uip

men

t & S

ervices

Energy Equipment & Services

June 2019By: Corey Chan (S1700518100001) and Dun Wang (S1700519060002)

China Solar EquipmentInitiate coverage: Brighter days ahead

We forecast an industry upcycle in 2019-20e, driven by robust export demand

We see upside to stock prices due to market expansion and falling costs

Initiate on Longi, Tongwei, Zhonghuan, and Jingsheng with Buy ratings, and First Applied Material with a Reduce

S P O T L I G H T

http://www.euromoney.com/Brokers

加入智库微信群，每日免费获取报告

1、每日微信群分享 3+最新重磅报告；

2、每日分享当日华尔街日报；

3、每周分享经济学人期刊；

4、每月汇总150+行业报告

扫码加入时请备注“东西智库”

加入VIP小密圈，以上报告合集下载

1、加入小密圈即送3000+专属报告；

2、小密圈提供本站报告合集下载

3、智能制造供应链对接（产需对接）

4、独家产业信息、数据分享

1

Equities ● Energy Equipment & Services June 2019

The key messages

The solar industry is no stranger to volatility. Sharp swings in product prices, overcapacity

issues and, in turn, knock-on effects on stock prices have been common. The root cause is

often subsidies as governments push for cleaner power. Germany led the way and became the

world’s largest solar market in 2010. China soon took over and solar capacity grew from 19GW

in 2013 to 175GW in 2018. But this led to overcapacity and a supply glut, so China has cut back

subsidies and plans to phase them out completely by 2021. China is not alone in cutting

subsidies. Other countries like Britain and Spain have also reduced subsidies as solar gets

closer to competing with fossil fuels on an equal footing.

In this report, we explain why we have a bullish outlook on the Chinese solar companies, given:

1) our global team’s bullish forecasts for global solar installations: and 2) China’s rising

penetration of the global solar supply chain. While we expect costs in the supply chain to keep

falling, we believe the rate of cost decline is likely to subside. We expect future cost reductions

to come in the form of higher cell efficiency, less raw material consumption, and economies of

scale in production. Lower solar costs – they have already fallen 92% in the last decade –

should drive long-term demand and increase photovoltaic (PV) installations.

Where our view differs from consensus

We believe consensus concerns about policy risk are overdone. We think the role policy

incentives play in new solar installations will diminish as the market moves closer to grid

parity, the point where alternative energies like solar match the price of using fossil fuels.

We analyze industry technology trends and see more uncertainty in cell and module

technology than for wafer and polysilicon.

We compare the risks to oversupply of different parts of the supply chain and conclude that

wafer manufacturing is a relatively safe place given its relatively consolidated market and

high investment barriers.

Stocks: As supply-chain cost pressure is likely to persist, we expect cost leaders to stand out; we

prefer Longi – the cost leader in wafer – and Tongwei – the cost leader in polysilicon and cell, both

rated Buy. Compared with other parts of the supply chain, we believe wafer and polysilicon cost

leaders will find it relatively easier to maintain scale advantages, given the high share of fixed cost in

their cost structure. We also have Buy ratings on Zhonghuan and Jingsheng. We have a Reduce

rating on First Applied Material (FAM) as we see pressure on earnings.

Why read this report?

We are bullish on the global photovoltaic product market in 2019-21e

as solar gets closer to competing with fossil fuels on an equal footing

Chinese producers stand to benefit as they increase their dominance

of the global supply chain; we think policy concerns are overdone

We explain why Longi and Tongwei are best positioned to benefit

Corey Chan* (S1700518100001)

* Head of A-share Infrastructure Research

HSBC Qianhai Securities Limited

[email protected]

+86 755 8898 3404

Dun Wang* (S1700519060002)

Analyst, A-share Infrastructure


[email protected]

* Employed by a non-US affiliate of HSBC Securities (USA) Inc, and is not registered/ qualified pursuant to FINRA regulations


2

Why read this report? 1

Facts and figures 3

Related research 4

Entering a new, brighter era 5

Technology upgrades, lower

costs to drive upside 15

Our stock-picking framework 37

Company section 43

Longi Green (601012 CH) 44

Tongwei (600438 CH) 55

Zhonghuan Semiconductor

(002129 CH) 68

Jingsheng M&E (300316 CH) 79

First Applied Materials

(603806 CH) 90

Disclosure appendix 105

Disclaimer 108

Contents

3


Facts and figures

11% Global new PV installations CAGR

in 2018-20e

129GW Global new PV installation in 2020e

We expect a closer levelized cost of electricity (LCOE) gap between solar and competing power sources to drive global new PV installation growth in a secular manner

cRMB270bn Market size of the global PV supply chain in 2020e

c90% China’s share in global

wafer supply in 2018


cell/module supply in 2018


polysilicon production in 2018

26% The share of new global PV installations by countries that achieved solar grid parity in 2018

We expect the share to increase to over 70% by 2021 when China reaches grid parity, making policy incentive a less-important factor for solar demand

92% How much solar costs have dropped in the

last decade

10-40% This is how much solar costs need to drop from the current level for solar grid parity

on a global scale


4

Related research

Recommended reading...

China Power T&D Equipment: Initiate coverage: Time to power up, 11 April 2019

GCL-Poly Energy Holdings (3800 HK), 30 March 2019

Global Upstream Solar, 25 March 2019

https://www.research.hsbc.com/R/10/GNwKW2Nx2gJr

5


Long-term sustainable growth ahead

China supplies 80-90% of the global mid-to-downstream PV products, and its dominance is

expanding upstream to polysilicon. Our global team expects the country to account for 60% of

global polysilicon supply in 2019, up from 53% in 2018 (see HSBC Global Renewable team’s

report titled Global Upstream Solar, 25 March 2019). Given China’s rising penetration of the

global solar supply chain and our global team’s bullish forecasts for global new PV installations

(Exhibit 2), we see upside potential for Chinese producers positioned along the supply chain.

We believe consensus concern about policy risks is overdone. In fact, we think the role policy

incentives play in new solar installations will diminish as the market moves closer to grid parity.

Globally, 26% of PV new installations in 2018 came from the markets which have already

achieved grid parity. With China targeting grid parity by 2021, the share would increase to

c70%. In fact, the share could be higher as China’s solar levelized cost of energy (LCOE)

premium is one of the highest globally due to the low generation cost of its coal-fired power

plants. LCOE describes the cost of the power produced over a period of time.

PV system costs have dropped by 75% since 2010. While we expect this to continue, the rate of

cost decline is likely to subside. We expect another 10-40% cuts in system costs to enable grid

parity on a global scale. Not all cost cuts will pressure margins, as some of the reduction could

come from the improvement in cell efficiency. We expect a 1ppt increase in cell efficiency to

bring about at least 3% reduction in PV system cost.

Given the important role that technology plays in the solar industry, suppliers that have invested

in the right equipment and processes are likely to become tomorrow’s winners. While some

technology outcomes can be difficult to predict, some trends are easy to observe – like mono-Si

over multi-Si in the wafer space. This should give mono-Si market leaders like Longi and

Zhonghuan a material advantage over their multi-Si peers. We see more uncertainties about

cell and module technology trends as there are more alternatives than in the upstream.

Compared with other parts of the supply chain, we believe wafer manufacturing is safer from

oversupply due to its relatively consolidated market, high investment barrier, and relative low

profitability for incumbents. However, module production is more vulnerable to new supply given

relatively low investment levels and technology barriers.

A-share PV product suppliers are trading at a 19x 12-month forward PE, close to the historical

average valuation since 2015. However, in view of lower demand volatility associated with

policy changes with the industry approaching grid parity, we expect valuation to rise to 20-30x,

at the upper-end of the historical range. As the supply-chain cost pressure is likely to persist in

Entering a new, brighter era

We believe there are brighter days ahead for the solar industry,

underpinned by sustainable growth in long-term demand and the move

towards grid parity. With technology advances and cost reductions

underway, we see opportunities for suppliers along the solar supply

chain. Our preferred stocks are Longi and Tongwei.

China supplies 80-90% of the

global mid-to-downstream PV

products

Suppliers that have invested

in the right equipment and

processes are likely to

become tomorrow’s winners


6

the near term, we expect cost leaders to outperform its peers down the road. We therefore have

Longi – the cost leader in wafer – and Tongwei – the cost leader in polysilicon and cell – as our

preferred stocks. Compared with other parts of the supply chain, we believe it should be relative

easy for wafer and polysilicon cost leaders to maintain their scale advantage, given the high

share of fixed cost in their cost structure.

Exhibit 1. A-share PV product suppliers: Industry trading close to the historical average valuation

Exhibit 2. HSBC forecasts global PV installations to rise at a 11% CAGR in 2018-20e

Source: Wind, HSBC Qianhai Securities. Source: Wind, HSBC Global Research estimates

Concerns over the China-US trade dispute overdone

In June 2018, the US started to impose an additional 25% tariff on imported Chinese solar cells

and modules. Many market participants are concerned about the negative impacts of the tariff

on the demand of China’s PV product. We believe such concern is overdone. Exports to the US

accounted for only 4% of Longi’s 2018 sales, and almost none for Tongwei. In addition, China’s

overwhelming share in the producer market implies limited likelihood of US reducing its reliance

on China’s PV products, given a lack of substitute supplies.

In fact, the trade dispute could be a blessing in disguise for Chinese upstream solar producers.

In August 2018, China announced plans to impose an additional 5-25% tariff on polysilicon

imported from the US. We believe this could reduce foreign competition for polysilicon suppliers,

including Tongwei and TBEA. Should the trade dispute deepen to include a trade ban on US-

made production equipment, we expect Chinese equipment makers like Jingsheng to benefit.

-

10.0

20.0

30.0

40.0

50.0

60.0

70.0

Jan-12 Jan-14 Jan-16 Jan-18

PE Mean +1SD -1SD

0

20

40

60

80

100

120

140

2015 2016 2017 2018 2019e 2020e

sola

r in

stal

lati

on

s (G

W)

China ROW

Exports to the US accounted

for only 4% of Longi’s 2018

sales

7


Exhibit 3. China accounts for c30% of global cumulative PV installations…

Exhibit 4. … but more than 50% of the global PV supply chain

Source: Bloomberg, HSBC Qianhai Securities. Source: Company, Wind, HSBC Qianhai Securities.

How much higher can renewable energy penetration go?

The Chinese government has set bold targets for renewable energy, aiming for a 50% share of

power generation from renewable sources (including hydro, solar, and wind) by 2030. Some

market participants doubt the feasibility of such a scenario given the variable nature of power

outputs from renewable sources like solar and wind. We believe this concern is groundless.

Power penetration of wind and solar reached 28.3%/24.5%/23.8%/25.1%/21.2% in

Germany/Netherlands/Portugal/Spain/UK in 2018. The share in China and the US is still low, at

7.8% and 8.2% respectively, implying great growth potential. We believe dispatchable baseload

power such as hydro (18% share of power mix in China) and natural gas (35% share of power

mix in the US), and a smarter grid can help to smooth out the generation cycles of wind and

solar farms.

Exhibit 5. National policy targets for solar capacities

Country Policy Targets

China To increase power generation from renewables from 30% in 2018 to 50% in 2030 France To increase solar capacity from 10GW in 2018 to 20GW in 2023 Germany Renewables' share of power generation to rise from 46% in 2018 to 80% in 2050 India To increase solar capacity from 28GW in 2018 to 100GW in 2022 Italy To increase solar capacity from 20GW in 2018 to 50GW in 2030 Japan Renewables' share of power generation to rise from 16% in 2017 to 22-24% in 2030 Sweden 100% power generation from clean energy sources by 2040 US Investment tax credits on solar at federal level

Source: Bloomberg, HSBC Qianhai Securities.

China32%

Europe28%

North America

14%

Japan12%

Rest of Asia-Pacific & Central

Asia5%

India5%

Latin America & Carribbean

2%

Middle East & Africa

2%

Global cumulative PV installations (2017)

94%84% 82%

53%

68% 68% 68%

14%

0%10%20%30%40%50%60%70%80%90%

100%

Wafer Cell PV Module Polysilicon

2018 2010

China's share of global supply


8

Exhibit 6. China’s power mix (2018): Solar and wind accounted for just an 8% share

Exhibit 7. Germany’s power mix (2018): A 28% share from solar and wind power

Source: Wind, HSBC Qianhai Securities. Source: Bloomberg, HSBC Qianhai Securities.

Exhibit 8. China’s solar tariff: Subsidies reduced for centralized projects after 1 July 2019

RMB/kWh Zone I Zone II Zone III Bidding or Not

Poverty-alleviation projects 0.65 0.75 0.85 No bidding Centralized On grid before 1st Jul. 2019 0.50 0.60 0.70 No bidding On grid after 1st Jul. 2019 0.40 0.45 0.55 Bidding Household Self-use Sold to grid On grid before 1st Jul. 2019 0.18+retail tariff 0.18+coal-fired plant on-grid tariff No bidding On grid after 1st Jul. 2019 0.18+retail tariff 0.18+coal-fired plant on-grid tariff No bidding Other decentralized Self-use Sold to grid On grid before 1 July 2019 0.32+retail tariff 0.32+coal-fired plant on-grid tariff No bidding On grid after 1 July 2019 0.10+retail tariff 0.10+coal-fired plant on-grid tariff Bidding

Source: NEA, NDRC, HSBC Qianhai Securities.

Initiate coverage of five A-share PV product makers

Longi (601012 CH, Buy, TP RMB36.2): As the largest global mono wafer producer, Longi

is well positioned to benefit from the trend of rising share of mono wafers in PV panel

production (up from 45% in 2018 to 69% in 2021e). The company has been a cost leader in

the wafer industry, cutting its non-Si cost by 72% since 2013. We expect further declines of

c40% in 2018-21, driven by economies of scale. This, together with the addition of 37GW

new wafer capacities in 2019-21, should lead to market share gains from 35% in 2018 to

42% in 2021. Valuation at 20x 2019e PE looks attractive, with a 43% earnings CAGR in

2018-21e.

Tongwei (600438 CH, Buy, TP RMB19.4): Tongwei is a global cost leader in both

Passivated Emitter and Rear Cell (PERC) and polysilicon production. Its cell processing

cost at RMB0.23/W in 1Q19 is 34% below the industry average, and we expect that to drop

to RMB0.19/W in 2021. In polysilicon production, the company’s Baotou Ph1 and Leshan

Ph1 projects have a unit overall cost of RMB50,000/t, close to the bottom of the industry

range. Around 60% of the company’s revenue in 2018 came from the agriculture business,

which generates steady cash flow. This should alleviate the capex burden of Tongwei’s PV

business. Valuation at 17x 2019e PE looks attractive, in our view, with a 27% earnings

CAGR in 2018-21e.

Wind5%

Solar3%

Hydro17%

Thermal power70%

Nuclear4%

Biomass and other

1%

Wind21%

Solar8%

Hydro4%Thermal

power41%

Nuclear13%

Biomass and other13%

9


Zhonghuan (002129 CH, Buy, TP RMB14.3): The state-owned company is the second

largest PV mono wafer supplier in the world and a leading semiconductor wafer supplier in

China. We expect its revenue to more than double in 2018-21e with the completion of new

PV and semiconductor wafer capacities. Compared with its private sector peers, access to

funding is relatively easy and the company has a low borrowing cost historically thanks to

its SOE status. On the completion of the RMB5bn private placement, we expect its net

gearing to drop to 27% in 2019 from 55% in 2018. While 2019e PE appears rich at 51x, we

expect that to drop to 24x in 2020e and 16x in 2021e.

Jingsheng (300316 CH, Buy, TP RMB14.1): We view Jingsheng as the best way to play

the wafer capacity upcycle among our coverage, given the company’s market leadership in

wafer-making equipment in China. The company has a c50% market share in third party

made crystal-growing equipment in China. We expect Jingsheng’s crystal-growing

equipment to register a 28% revenue CAGR in 2018-21e, driven by capacity expansions in

China’s PV and semiconductor wafer industry, leading to a 25% earnings CAGR in 2018-

21e. Valuation at 19x 2019e PE is attractive, in our view.

FAM (603806 CH, Reduce, TP RMB25.8): The company is the largest supplier of ethylene

vinyl acetate (EVA) film in the world, with around a 60% market share in 2018. While the

EVA market is likely to grow with rising global new PV installations, we flag higher cell

conversion efficiency and continuous cuts in post-cell fabrication costs as downside risks.

As a result, we expect only 3% earnings CAGR for FAM in 2018-21e. Valuation at 25x

2019e PE looks rich, we believe.

Company-specific catalysts

Longi: 1) Stronger-than-expected wafer orders driven by better-than-expected global solar new

installations; 2) larger-than-expected reduction in wafer non-Si cost; and 3) mono’s share of the

wafer market growing faster than expected.

Tongwei: 1) Stronger-than-expected solar cell orders driven by stronger-than-expected global

solar new installations; 2) ramp-up of the new polysilicon capacity sooner-than-expected; and 3)

better-than-expected polysilicon pricings on industry capacity cuts.

Zhonghuan: 1) Stronger-than-expected wafer orders driven by better-than-expected global

solar new installations; 2) making inroads in the 12’ semi wafer markets; and 3) sooner-than-

expected completion of the new production facilities in Inner Mongolia and Wuxi.

Jingsheng: 1) Better-than-expected demand for crystal-growing equipment, spurred by

stronger-than-expected capacity expansion by wafer makers; 2) new product launches; and 3)

better-than-expected market share, driven by an acceleration of localization of Chinese wafer-

lines owing to heightening China-US trade tensions.

FAM: 1) EVA film demand weaker than expected, driven by lower-than-expected global solar

new installations; 2) slower-than-expected progress made in diversification into 3C and

semiconductor downstream; and 3) weaker-than-expected contribution from POE film on

slower-than-expected adoption of bifacial panels.

Where we are different from consensus

Longi: Our 2020-21 earnings estimates are 4-11% above consensus as we expect faster

earnings growth for its wafer and module businesses.

Tongwei: Our 2021 earnings estimate is 7% above consensus as we expect stronger earnings

growth for its mono cell business.

Zhonghuan: Our 2021 earnings estimate is 9% above consensus as we expect stronger

earnings contribution for its Wuxi semiconductor wafer project, slated for completion in phases

in 2019-22.


10

Jingsheng: Our 2019 earnings estimate is 7% above consensus as we expect stronger crystal-

growing machine demand, driven by aggressive capacity expansion plans of Chinese wafer makers.

FAM: Our 2020-21 earnings estimates are 9-21% below consensus as we are more bearish on

EVA film demand driven by higher cell efficiencies.

Exhibit 9. HSBC Qianhai Securities estimates vs. Consensus

__ HSBC Qianhai estimates___ ___ Consensus estimates ____ ________ Variance _________ Company RMBm 2019 2020 2021 2019 2020 2021 2019 2020 2021

Longi Revenue 29,000 37,451 49,803 28,442 37,314 48,445 2% 0% 3% Net profit 3,592 5,598 7,445 3,595 5,386 6,717 0% 4% 11% Zhonghuan Revenue 18,745 22,105 30,281 17,911 21,835 29,274 5% 1% 3% Net profit 635 1,384 2,094 661 1,370 1,924 -4% 1% 9% FAM Revenue 5,740 6,050 6,243 5,880 7,139 8,399 -2% -15% -26% Net profit 694 764 820 701 839 1,032 -1% -9% -21% Jingsheng Revenue 3,271 3,953 4,637 3,173 3,967 4,393 3% 0% 6% Net profit 801 971 1,146 747 987 1,123 7% -2% 2% Tongwei Revenue 34,926 37,755 40,001 34,669 37,830 38,795 1% 0% 3% Net profit 3,220 3,955 4,166 3,176 3,992 3,891 1% -1% 7%

Source: Wind, HSBC Qianhai Securities. E= HSBC Qianhai Securities estimates

Note: Net profit here is after distribution to preferred shareholders and to perpetual security holders

Sector catalysts and downside risks

Catalysts: 1) Stronger-than-expected PV installations globally underpinned by faster-than-expected

solar grid parity; 2) supportive government policies; 3) vertical integration that could generate

synergies; 4) evolution of technologies in favour of existing players; 5) stronger-than-expected cost

reduction in the supply chain on technology improvement and economies of scale.

Downside risks: 1) Weaker-than-expected PV installations on slower-than-expected solar grid

parity; 2) unfavourable policy shifts; 3) equity dilution risks stemming from heavy capex; 4)

market-shifting new technologies unfavourable to incumbents; 5) hurdles to cost reduction in the

supply chain.

ESG

The five companies have all set detailed guidelines for ethical management, safety

management and social contributions, and also follow China’s market rules on corporate

governance. Average board tenure for each of the five companies are all over two years. All five

companies have independent board members accounting for over 20% of the board.

We believe global governments’ efforts to deliver emission caps are positive for the solar industry.

Over the past decade, global countries have set extensive, clear targets to reduce emissions as

part of economic growth efforts, and/or have provided tax incentives for renewables. In China, solar

capacities reached 175GW in 2018, far ahead of the original targets.

Exhibit 10. ESG indicators, 2018

Longi Zhonghuan FAM Jingsheng Tongwei

No. of board members 12 11 7.0 10.0 11 Average board tenure (years) 4.6 2.5 3.9 5.2 3.4 No. of female board members 3 5 1 2 0 Female board members (%) 25.0% 45.5% 14.3% 20.0% 0.0% No. of independent board members 3 4 3 3 3 Board member independence (%) 25.0% 36.4% 42.9% 30.0% 27.3%

Source: Wind, Company data, HSBC Qianhai Securities

11

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 11.Global Solar: Valuation comps

Company Stock Code Cur Rating TP Closing Market cap. ADTV _____ P/E _______ _____ P/B ______ ___ ROE ___ _____ DY ______ Name 6/14/2019 USbn USm 19e 20e 19e 20e 19e 20e 19e 20e

Polysilicon producer GCL-Poly Energy 3800 HK HKD 0.4 1.03 11 8.6 7.7 0.3 0.3 4% 4% 0.3% 0.0% DAQO New Energy DQ US USD 45.5 0.60 12 4.8 4.3 0.8 0.7 17% 16% NA NA TBEA * 600089 CH CNY Buy 10.0 7.1 3.81 49 10.8 9.1 0.9 0.8 8% 9% 3.1% 3.7% OCI 010060 KS KRW 94200.0 1.90 12 14.8 12.8 0.6 0.6 4% 5% 0.9% 1.0% Wacker Chemie AG WCH GR EUR 70.2 4.12 17 14.2 11.7 1.1 1.0 8% 9% 3.7% 4.2% Tongwei * 600438 CH CNY Buy 19.4 14.2 7.97 88 17.1 14.0 3.2 2.7 19% 19% 1.8% 2.2% Crystalline silicon wafer producer Zhonghuan * 002129 CH CNY Buy 14.3 9.6 3.86 97 50.6 24.1 1.8 1.7 4% 7% 0.2% 0.4% Longi * 601012 CH CNY Buy 36.2 22.3 11.67 146 19.9 14.5 3.5 2.9 18% 20% 0.5% 0.7% Sino-American Silicon products 5483 TT TWD 82.1 1.53 29 6.1 5.0 1.4 1.3 23% 26% 10.1% 12.0% Solar cell and module producer Canadian Solar CSIQ US USD 21.7 1.29 19 8.0 9.1 0.8 0.7 11% 8% 0.0% 0.0% Jinko Solar JKS US USD 24.7 1.07 21 2.0 2.1 0.6 0.6 32% 27% 0.0% 0.0% United Renewable Energy/Taiwan 3576 TT TWD 10.1 0.81 4 NA NA NA NA NA NA NA NA First Solar Inc FSLR US USD 62.2 6.55 81 16.9 17.2 1.1 1.1 7% 6% NA NA Other solar material and equipment First Applied Material * 603806 CH CNY Reduce 25.8 32.8 2.48 9 24.7 22.4 2.9 2.6 12% 12% 1.3% 1.4% Jingsheng * 300316 CH CNY Buy 14.1 12.0 2.23 50 19.3 15.9 3.3 2.8 17% 18% 1.1% 1.4% Sungrow Power Supply 300274 CH CNY 8.8 1.85 64 9.3 7.5 1.3 1.1 14% 15% 1.3% 1.7% Solaredge Technologies SEDG US USD 57.9 2.75 30 19.7 20.3 2.8 2.3 14% 11% NA NA

Source: Bloomberg, Company data, HSBC Qianhai Securities, *HSBC Qianhai Securities estimates. E= Bloomberg consensus estimates for all other companies

12

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 12. Key suppliers along the solar supply chain

Source: Company data, HSBC Qianhai Securities

EVA film：FAM (603806.SH)SveckHiuv

Polysilicon：GCL Poly (3800.HK)Wacker Chemie AG (WCHG.DE)OCI (010060.KS)Tongwei (600438.SH)Xinte Energy (1799.HK) DAQO New Energy (DQ.N)REC Silicon (REC.OL)Hemlock SemiconductorSinosicoEast Hope GroupHanwha Chemical (009830.KS)Hankook Silicon

Module：Jinko Solar (JKS.N)Canadian Solar(CSIQ.O)Trina SolarJA SolarLongi (601012.SH)Hanwha Q CellsZhongli Group (002309.SZ)GCL System (002506.SZ)Yingli Green EnergyCsun SolarRisen Energy (300118.SH)

Cell：Trina SolarTongwei (600438.SH)Longi (601012.SH)Hanwha Q CellsJA SolarCanadian Solar(CSIQ.O)JinkoSolar (JKS.N)Aiko Solar Energy (600732.SH)Risen Energy (300118.SH)Yingli Green EnergyNeo Solar Power (3576.TW)

PV System EPC:TBEA (600089.SH)SunPower (SPWR.O)Blattner EnergyTalesun Solar Chint Electronics (601877.SH)Jinko Solar(JKS.N)Risen Energy (300118.SZ)

Wafer：Longi (Mono-Si)(601012.SH)Zhonghuan (Mono-Si)(002129.SZ)GCL Poly (Multi-Si)(3800.HK)Jinko Solar (JKS.N)JA SolarCanadian Solar (CSIQ.O)LDK solar

Inverter：Sungrow Power (300274.SZ)Solaredge Technology (SEDG.O)Hua Wei

Wafer equipment：Jingsheng (300316.SZ)S.C New Energy Technology (300724.SZ)

其他

Solar glass：Flat Glass (6865.HK)China Southern Glass (000012.SZ)

13


Valuation and risks

Valuation methodology Risks

Longi

601012 CH

Current price:

RMB22.3

Target price:

RMB36.2

Up/downside:

62%

Our TP of RMB36.2 is derived from our discounted cash flow (DCF)

valuation model. Key assumptions in our DCF valuation model

include: 1) Cost of equity (COE): We use a COE of 10.0%. This is

derived from a risk-free rate of 2.5%, a market risk premium of

6.5%, and a beta of 1.16; 2) Cost of debt (COD): We assume the

pre-tax cost of debt to be 5% and after-tax cost of debt to be 4%.

We use our 2020e debt-to-capital ratio of 20% as our long-term

debt-to-capital ratio; 3) Operating cash flow to grow 11% per

annum: We expect operating cash flow (before changes in working

capital) to expand at a CAGR of 11% in 2018-29e, reflecting solid

growth in demand; 4) Capital expenditure (capex): We assume

high capex of around RMB6bn per annum in 2019-21e, reflecting

the expansion of its wafer/cell/module capacities; we assume

steady capex of around RMB1-2bn per annum in 2022-29e,

reflecting steady investment in maintenance; 5) Terminal growth

rate at 2%, and we assume the company reaches a steady growth

period after 2029.

With our TP implying upside of 62%, we initiate with a Buy rating.

Downside risks:

Weaker-than-expected PV demand

Risk of equity dilution from fundraising exercise

Greater-than-expected cuts to wafer ASP

Buy

Corey Chan | [email protected] | +86 755 8898 3404

Tongwei

600438 CH

Current price:

RMB14.2

Target price:

RMB19.4

Up/downside:

36%





6.5%, and a beta of 1.42. 2) Cost of debt (COD): We assume the



debt-to-capital ratio. 3) Operating cash flow to grow 9% per



growth in demand. 4) Capital expenditure (capex): We expect a

capex of RMB13bn in 2019e, driven by cell and solar farm capacity

expansion. Thereafter, we expect capex to drop to around RMB2-

3bn per annum in 2020-29e, reflecting steady maintenance capex.

5) Terminal growth rate at 2%, and we assume the company

reaches a steady growth period after 2029.


Downside risks:

Weaker-than-expected polysilicon prices

Ramp-up of the new polysilicon capacity taking longer-than-

expected


Risk of equity dilution from fundraising exercise.

Evolution of the cell technology which could undermine

investment in PERC

Greater-than-expected cuts to cell ASPs

Buy


Zhonghuan

002129 CH

Current price:

RMB9.60

Target price:

RMB14.3

Up/downside:

49%











growth in demand. 4) Capital expenditure (capex): We assume

steady capex of around RMB5bn per annum in 2019-29e, reflecting

high investment in new projects and capacity expansion. 5)

Terminal growth rate at 2%, and we assume the company reaches

a steady growth period after 2029.


Downside risks:


China-US trade tension which could hinder equipment

procurement from US suppliers

Risk of equity dilution from fundraising exercise

Weaker-than-expected semi demand

Greater-than-expected cuts to wafer ASP

Buy



14

Valuation methodology Risks

Jingsheng

300316 CH

Current price:

RMB12.0

Target price:

RMB14.1

Up/downside:

17%






pre-tax cost of debt to be 5% and the after-tax cost of debt to be

4%. We use our 2020e debt-to-capital ratio of 2% as our long-term



capital) to expand at a CAGR of 13% in 2018-29e, reflecting stable

growth in demand. 4) Capital expenditure (capex): We expect

capex of around RMB700m per annum in 2019-21e, reflecting the

company’s share of investment in the Wuxi semi wafer project.

Thereafter, we expect a steady capex of around RMB200-300m per

annum in 2022-29e, reflecting maintenance capex. 5) Terminal

growth rate at 2%, and we assume the company reaches a steady

growth period after 2029e.

With our TP implying upside of 17%, we initiate with a Buy rating as we

believe the company is the best name to play the upcoming capacity

expansions in China’s PV and semiconductor wafer industry.

Downside risks:

Weaker-than-expected PV and semi demand.

Weaker-than-expected capex of Zhonghuan.

Higher-than-expected receivable provision.

Weaker-than-expected margin on intensified competition

Buy


FAM

603806 CH

Current price:

RMB32.8

Target price:

RMB25.8

Up/downside:

-21%




derived from a risk-free rate of 2.5%, a market risk premium of 6.5%,

and a beta of 1.00. 2) Cost of debt (COD): We assume the pre-tax

cost of debt to be 5% and after-tax cost of debt to be 4.3%. We use

our 2020e debt-to-capital ratio of 0% as our long-term debt-to-capital

ratio. 3) Operating cash flow to grow 4% per annum: We expect

operating cash flow (before changes in working capital) to expand at

a CAGR of 4% in 2018-29e, reflecting steady growth in demand. 4)

Capital expenditure (capex): We expect a steady capex of around

RMB200m per annum in 2019-29e, reflecting steady investment in

maintenance. 5) Terminal growth rate at 2%, and we assume the

company reaches a steady growth period after 2029e.

With our TP implying downside of 21%, we initiate with a Reduce rating.

Upside risks:

Stronger-than-expected contribution from the electronic

material business

Stronger-than-expected PV demand

Better-than-expected margin on lessening competition

Reduce


Priced at 14 June 2019 Source: HSBC estimates * Employed by a non-US affiliate of HSBC Securities (USA) Inc, and is not registered/ qualified pursuant to FINRA regulations

15


Our global team expects a 11% CAGR in the world’s PV installations in 2018-20e, driven by lower PV costs

Solar grid parity on a global scale is closer than ever

Solar cost has dropped by 92% in the last decade, resulting in significant reductions in the gap

between the cost of solar power and that of competing fuel sources. As shown in Exhibit 14, the

levelized cost of electricity (LCOE) of solar is now competitive with that of competing forms of

power generation and wholesale electricity prices in several key markets including India, the

US, and western Europe (Exhibit 15). These markets accounted for around 26% of global PV

installations in 2018.

In India, which accounted for 8% of global PV demand in 2018, the LCOE of solar is 40% below

that of coal thanks to the nation’s abundant solar resources. In France and Germany, the LCOE

of solar is 25% and 50% below that of nuclear and that of coal respectively. In China, which

accounted for around 42% of global PV demand in 2018, the LCOE of solar is still 40% above

that of coal, but the government targets parity by 2021. Since China’s electricity price is one of

the lowest globally (Exhibit 16), having parity would have global implications. As such, we are

bullish over the world’s PV demand in 2019-21.

Technology upgrades, lower

costs to drive upside

We expect a rise in demand for PV products, driven by technology

upgrades and lower system costs. This should benefit China’s

producers, accounting for 50-90% of the global solar supply chain

Wafer manufacturing to be the most resilient part of the supply

chain, given its relatively consolidated market, high investment

barrier, and less risk associated with the technology roadmap

As cost pressures persist, we expect cost leaders like Longi and

Tongwei to be market consolidators, extending their reach both

horizontally and vertically

Solar cost has dropped by

92% in the last decade


16

Exhibit 13. Solar system cost has dropped 71% since 2010

Source: BNEF, HSBC, e= HSBC Global Research estimates

Exhibit 14. PV LCOE: Already on parity with electricity wholesale price in several key markets

Source: Bloomberg, HSBC Qianhai Securities

Note: Major electricity power source for each country: Brazil (hydro), Japan (nuclear), Indonesia (coal), China (coal), Germany (coal), US (natural gas), France (nuclear), Australia (coal), Spain (wind), India (coal)

Exhibit 15. Global solar installation breakdown, 2015-18


3.53

2.80

1.831.60 1.52 1.45

1.231.04

0.89 0.80 0.77 0.74

0%

5%

10%

15%

20%

25%

30%

35%

40%

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019e 2020e 2021e

US

D/W

Module Inverter BoP EPC Other Other y-o-y decline (%)

0%

50%

100%

150%

200%

250%

Brazil Japan Indonesia China Germany US France Australia Spain India

vs LCOE of major electricity power source vs electricity wholesale price Grid parity

PV LCOE / LCOE of competing power source or electricity wholesale price

17%9% 8% 9%

14%19%

12% 11%

3% 6%10% 8%

31%45% 53%

42%

24%11% 6%

6%

11% 10% 11%23%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2015 2016 2017 2018

Europe USA India China Japan ROW

17


Exhibit 16. China has one of the lowest electricity prices in the world


Exhibit 17. HSBC global solar installation forecasts

Source: SolarPower Europe, HSBC, e= HSBC Global Research estimates

China has become a major contributor of the global solar supply chain

China’s share of the global solar supply chain has expanded strongly over the last few years

(Exhibit 18). In 2018, the country accounted for c90% of the global wafer supply, c80% of the

global cell/module supply, and c50% of the global polysilicon supply. In view of our bullish

outlook of the world’s PV demand, we expect significant growth in China’s PV module exports in

the coming years. As shown in Exhibit 19, China’s PV module monthly export reached a record

level of 6.3GW in March 2019, up 81% y-o-y.

330

250 240220

190

130 130100

80 80

0

50

100

150

200

250

300

350

Germany Australia Spain Japan France Brazil US Indonesia China India

Electricity retail prices in 2018 (USD/MWh)

0%

10%

20%

30%

40%

50%

60%

0

20

40

60

80

100

120

140

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019e 2020e

sola

r in

stal

lati

on

s (G

W)

China ROW China as % of total (RHS)

Europe drives demand; overcapacity starts to build

China capacity ramp drives down pricing

Phase 1 Phase 2 Phase 3 Phase 4

China dominates supply and demand

RoW drives rebound


18

Exhibit 18. Global solar: China’s share of production has grown strongly over the last few years

Exhibit 19. China’s module exports reached a new record in March 2019

Source: Company data, Bloomberg, HSBC Qianhai Securities Source: Wind, HSBC Qianhai Securities

Solar supply chain still faces pricing pressure, but not as strong as that in the past

Globally, PV LCOE ranges from USD50/MWh to 120/MWh, depending on the cost of the PV

system and the effective operating hours. As shown in Exhibit 20, the average LCOE of global

coal-firing power plants is around USD70/MWh. Hence, to achieve cost parity with fossil-fuel

burning power plants on a global scale, we believe PV LCOE needs to drop by another 10-40%

from the current level. As shown in Exhibit 21, the relationship between LCOE and PV system

cost is almost linear on stable effective operating hours. Hence, we believe PV system cost

needs to drop by 10-40%, to achieve a similar degree of reduction in LCOE, assuming stable

operating hours. In our view, this is achievable via improvements in cell technology and

production on a mass scale.

Exhibit 20. LCOE of coal-firing power plants (2018)


94%84% 82%

53%

68% 68% 68%

14%

0%10%20%30%40%50%60%70%80%90%

100%

Wafer Cell PV Module Polysilicon

2018 2010

China's share of global supply

-40.0%

-20.0%

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

120.0%

140.0%

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Jan-15 Dec-15 Nov-16 Oct-17 Sep-18

Tho

usan

ds

China Module Export YoY

GW YoY

0

20

40

60

80

100

120

140

Australia US Indonesia Japan Malaysia India Brazil SouthKorea

Canada China

LCOE of coal-firing power plants (USD/MWh)

Global PV LCOE band

19


Exhibit 21. PV LCOE: Sensitivity to PV system cost and effective operating hours

PV system cost (RMB/W)

Effective operating hours -30% -25% -20% -15% -10% -5% 0% 750 -7% 0% 7% 13% 20% 27% 33% 800 -13% -6% 0% 6% 13% 19% 25% 850 -18% -12% -6% 0% 6% 12% 18% 900 -22% -17% -11% -6% 0% 6% 11% 950 -26% -21% -16% -11% -5% 0% 5%

1,000 -30% -25% -20% -15% -10% -5% 0%

1,050 -33% -29% -24% -19% -14% -10% -5% 1,100 -36% -32% -27% -23% -18% -14% -9% 1,150 -39% -35% -30% -26% -22% -17% -13% 1,200 -42% -38% -33% -29% -25% -21% -17% 1,250 -44% -40% -36% -32% -28% -24% -20%

Source: HSBC Qianhai Securities estimates

Exhibit 22. The PV panel manufacturing process: the method of ingot creation is what differentiates mono-Si panels from multi-Si

Source: HSBC Qianhai Securities

BOS and module processing could see more price cuts

BOS is the largest contributor to system cost

The cost of a PV system can be broken down into five key parts – polysilicon, wafer non-Si cost,

cell processing cost, module processing cost, and balance of system (BOS). BOS, including

installation and other non-module related cost items, is the biggest, accounting for c50% of the

system cost in 2018. This makes it an ideal target for cost-saving initiatives by PV system

operators. We estimate a 10% cut in BOS reduces system cost by 5%. This compares with a

2% reduction for a 10% cut in module processing cost, and just 1% reduction for a 10% cut in

polysilicon prices/wafer non-Si cost/cell processing cost.


20

Exhibit 23. PV system cost breakdown (2018)


Exhibit 24. PV system cost is most sensitive to BOS cuts

10% reduction in PV system cost reduction

Polysilicon ASP -1.2% Wafer non-Si cost -1.1% Cell processing cost -1.1% Module processing cost -1.7% BOS -4.9%


The rise in cell efficiency has bigger impact on post-cell fabrication costs

BOS and module processing costs are post-cell fabrication costs as they are incurred after the

cells are made. These costs are largely independent of the kind of cell installed and more to do

with the size of the module. This makes them highly susceptible to the increase in cell

efficiency. We expect a 1ppt increase in cell efficiency to lower BOS and module processing

costs by 4% and PV system cost by 3% (Exhibit 27). Here, we assume no change to the cell

cost per watt as it is difficult to predict the cost trend for different cell technologies. As the

improvement in cell efficiency is likely to be a long-term trend for the industry, we expect the

downside pressure of BOS and module processing cost to linger for the next few years.

Exhibit 25. Module processing cost (2018): Costs like glass, EVA, and frame are relevant to the size of the module

Exhibit 26. BOS cost breakdown (2018): A majority of the costs are fixed

Source: CPIA, HSBC Qianhai Securities Source: CPIA, HSBC Qianhai Securities

-

0.35

0.53

1.03

0.24

2.10

-

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

Polysilicon Wafer Cell Module BOS PV system cost

Cost of sales Gross Margin

Rmb/W 4.25

Glass20%

Back plate14%

EVA11%

Frame28%

Welding belt5%

Wire box9%

Conductive adhesive

3%

Package1%

Labor3%

Others3%

D&A3%

Management fee2%

Secondary equipment

4%Primary

equipment12%

Roof adjustment

12%Roof leasing

2%Grid

connecting cost10%

Cable13%

E&C23%

Structure11%

Inverter11%

21


Exhibit 27. Rise in cell efficiency could bring down post-cell fabrication costs

______ Cell efficiency ________ Module processing cost

BOS Other non-cell related costs

Cell cost PV system cost

Chg RMB/W RMB/W RMB/W RMB/W RMB/W

22% Unchanged 0.56 2.10 0.47 1.12 4.25 23% +1ppt 0.54 2.01 0.45 1.12 4.11 24% +2ppt 0.51 1.92 0.43 1.12 3.98 25% +3ppt 0.49 1.85 0.41 1.12 3.87 26% +4ppt 0.47 1.77 0.39 1.12 3.76 27% +5ppt 0.46 1.71 0.38 1.12 3.66 28% +6ppt 0.44 1.65 0.37 1.12 3.57


The decline in per watt polysilicon usage to continue with rising share of mono

In fact, the system cost savings from a 1ppt increase in cell efficiency could be larger than 3% if

we factor in lower per watt raw material usage. As shown in Exhibit 28, per watt polysilicon

usage declined 52% in 2007-18, during which cell efficiency had risen from c12% to c20%. The

decline in per watt polysilicon usage is likely to continue on a rising share of mono wafer which

uses 0.7g/w less polysilicon than multi wafer does. We estimate that a 10% reduction in per watt

polysilicon usage could reduce PV system cost by 1.2% (Exhibit 29).

Exhibit 28. Per watt polysilicon usage: Down 52% since 2007

Exhibit 29. Sensitivity of PV system cost to per watt polysilicon usage

Polysilicon usage per watt

PV system cost

% changes RMB/W % changes

0% 4.25 0.0% -5.0% 4.22 -0.6% -10.0% 4.20 -1.2% -15.0% 4.17 -1.8% -20.0% 4.15 -2.3% -25.0% 4.12 -2.9% -30.0% 4.10 -3.5%

Source: HSBC Source: HSBC Qianhai securities estimates

Wafer and polysilicon production to benefit more from economies of scale

Besides improvements in cell efficiency, economies of scale in production could also help to

reduce system costs by diluting labour and fixed costs over a larger product base. As shown in

Exhibit 30, wafer manufacturing is most sensitive to economies of scale as 34% of its COGS is

in labour and fixed costs. Next is polysilicon (26%), followed by cell (13%). For module

production, only 3% of its COGS is in labour and fixed costs, and the share in its total cost

(including SG&A) could be even lower as selling expense (a variable cost) is higher for module

compared with the upstream.

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

2007 2009 2011 2013 2015 2017 2019e 2021e

Polysilicon usage (g/w)


22

Exhibit 30. Labour and fixed costs: Higher share in wafer and polysilicon COGS

Exhibit 31. Mono wafer COGS breakdown (2018)

Source: Company data, HSBC Qianhai Securities Source: Company data, HSBC Qianhai Securities

Wafer and polysilicon more defensive against supply expansion

While capacity expansion has become rampant along the supply chain, we believe the wafer

and the polysilicon segments are more defensive compared with other parts of the supply chain.

This is because of their relatively high market concentration, the relatively heavy investment

required for new capacity, and low profitability, all factors that should safeguard the market from

new entrants.

Market concentration highest for wafer and lowest for cell

We believe a consolidated market reduces the risk of technology leakage to industry copy-cats

and, in turn, greater competition. Compared with other parts of the supply chain, market

concentration is high for wafer, particularly mono wafer. As shown in Exhibit 32, the top two players

accounted for 36% of global overall wafer supply and 65% of global mono wafer supply in 2018

(Exhibit 32). We expect this to increase to 42% and 77%, respectively in 2019 (Exhibit 34).

Comparatively, the cell market is relatively fragmented, with the top two accounting for only 14% of

global supply in 2018.

Exhibit 32. Market concentration of various parts of the solar supply chain (2018)

Source: Company, Bloomberg, HSBC Qianhai Securities

14%

41%

20%30%

52%

32%67%

67%

34%26%

13% 3%

0%

20%

40%

60%

80%

100%

Wafer Polysilicon Cell Module

Energy & Consumables Raw material

Labour & Fixed cost

Polysilicon52%

Wafer cutting15%

Ingot pulling33%

Mono wafer COGS = RMB0.46/W

81%

61%53%

40%34%

65%

36%29%

19%14%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Wafer (Mono-Si) Wafer (Overall) Polysilicon Module Cell

Top 5 players' market share Top 2 players' market share

23


Exhibit 33. Polysilicon: Global capacity breakdown (2019e)

Exhibit 34. Mono wafer: Global production breakdown (2019e)

Source: HSBC Qianhai Securities, e = Bloomberg consensus estimates Source: Company data, HSBC Qianhai Securities, e = Bloomberg consensus estimates

Exhibit 35. Cell: Global production breakdown (2019e)

Exhibit 36. PV module: Global production breakdown (2019e)

Source: Company data, Bloomberg, HSBC Qianhai Securities, e = Bloomberg consensus estimates

Source: Company data, Bloomberg, HSBC Qianhai Securities, e = Bloomberg consensus estimates

Mono wafer has high investment and technology entry barriers

Compared with other parts of the supply chain, investment levels and technology entry barriers

for mono wafer are high. As shown in Exhibit 37, the prevailing investment cost for 1GW of

mono wafer production capacity is around RMB500m, higher than that of PERC cell and

polysilicon. The investment cost for PV module capacity is much lower, at around

RMB80m/GW. In addition to its higher investment cost, mono wafer capacity also takes longer

to ramp up. As shown in Exhibit 38, a typical mono wafer line takes up to 12 months to ramp up

to its face-plate capacity. This compares with six months for polysilicon lines and three months

for cell lines.

Daqo Group5%

TBEA10%

GCL-Poly20%

Tongwei10%OCI

10%

Wacker Chemie

AG11%

Others34%

Longi42%

JinkoSolar7%

Zhonghuan35%

Others16%

Others50%

Trina Solar7%

Tongwei7%

LONGi7%

JinkoSolar7%

Hanwha Q Cells

7%

JA Solar6%

Canadian Solar6%

Motech Industries

3%

Others42%

JinkoSolar10%

Canadian Solar8%

Trina Solar8%

LONGi7%

JA Solar7%

Hanwha Q Cells7%

Zhongli Talesun

6%

GCL System

5%


24

Exhibit 37. The Investment costs of mono wafer lines are high compared with other parts of the supply chain


Exhibit 38. New capacity by type: Mono wafer takes the longest to ramp up


Profitability for wafer and polysilicon is low, discouraging new supply

Wafer’s profitability is low, even for market leaders. As shown in Exhibit 39, the two largest players

– Longi and Zhonghuan – generated just 15-16% gross margin for their wafer business in 2018.

With SG&A cost ratios at around 8-10%, operating margin could be as low as 5-8%. Margins are

likely to be even lower for smaller players. This does not give new entrants a lot of incentive to

increase supply. For polysilicon, due to a 53% price decline since early 2018, only a handful of

producers, including Tongwei, TBEA, and Daqo are still profitable while others are loss-making. For

cell, gross margins of industry leaders are still high, at around 20-30% in 1Q19, thanks to the price

recovery y-t-d (Exhibit 44). We believe this could lead to more supply in the near term.

500

420 408

295

80

-

100

200

300

400

500

600

Wafer (Mono-Si) Cell (PERC) Polysilicon Wafer (Multi-Si) PV Module

Investment Cost

RMB mn/GW

12

6

3

0

2

4

6

8

10

12

14

Wafer (Mono) Polysilicon Cell

Capacity ramp-up period (months)

25


Exhibit 39. Wafer: GMs for market leaders at mid-to-low teens in 2019e

Exhibit 40. Wafer: Pricings have stabilized since end-2018

Source: Company data, HSBC Qianhai Securities estimates Source: Wind, HSBC Qianhai Securities

Exhibit 41. Polysilicon: GMs for market leaders down to 20-30% in 2019e on price declines

Exhibit 42. Polysilicon: Prices have retreated 53% since early 2018

Source: Company data, HSBC Qianhai Securities. E= HSBC Qianhai Securities estimates

Source: Wind, HSBC Qianhai Securities

Exhibit 43. Cell: GMs for market leaders around 20-30% in 1Q19

Exhibit 44. Cell: Prices have shown a recovery y-t-d

Source: Company data, HSBC Qianhai Securities Source: Wind, HSBC Qianhai Securities

0%

5%

10%

15%

20%

25%

30%

35%

2015 2016 2017 2018 2019E

Zhonghuan Longi

PV wafer GM

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Jan-17 Jun-17 Nov-17 Apr-18 Sep-18 Feb-19

Multi-Si wafer Mono-Si wafer

Rmb/unit

47%42%

45%

28%

34% 34% 33% 28%

26%

18%

0%5%

10%15%20%25%30%35%40%45%50%

Tongwei TBEA DAQO GCL-Poly

2017 2018 2019E

Polysilicon GM

0

20

40

60

80

100

120

140

160


Rmb/kg

30%

20% 19%

0%

5%

10%

15%

20%

25%

30%

35%

Tongwei Risen Energy Aikosolar

1Q19

PV cell GM

0.00.20.40.60.81.01.21.41.61.82.0


Mono-Si cell Multi-Si cell

Rmb/W


26

Implications of cost pressure for the market landscape

Cost leaders could emerge as market consolidators

As cost pressure is likely to persist in the medium term, we see a chance of deepening market

consolidation that should boost the market positions of cost leaders like Longi and Tongwei. In

wafer production, Longi, the first to introduce diamond wire sawing (DWS) in 2015, has led the

cost curve. As shown in Exhibit 45, Longi’s non-Si costs in wafer production dropped by 72% in

2013-18, and we expect that to decline by a further c40% in 2018-21e, driven by economies of

scale. As non-Si costs account for around 50% of wafer costs, this should give Longi a material

advantage against its competitors. In cell and polysilicon markets, Tongwei’s cost of production

is amongst the lowest (Exhibit 46).

Exhibit 45. Longi: Managed to cut wafer non-Si costs by 72% since 2013

Exhibit 46. Tongwei: Cell processing cost is 30-40% below the industry average

Source: Company data, HSBC Qianhai Securities estimates Source: Company data, CPIA, HSBC Qianhai Securities estimates

Exhibit 47. Tongwei: Average variable cost for polysilicon production at low-end of the industry range

Source: BNEF, HSBC Qianhai Securities

Vertical integration could become a key strategy

We believe vertical integration along the supply chain carries three key benefits:

1. It allows downstream players to secure supplies and upstream players to secure markets;

-

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

2013 2015 2017 2019E 2021E

Wafer non-Si cost

RMB/W

-

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

2017 2018 2019E 2020E 2021E

Tongwei Industry average

RMB/W

20.00

20.00

15.00

10.8

10.0

7.5

7.1

6.5

6.2

0.00 5.00 10.00 15.00 20.00 25.00

REC (US)

Japan

Hemlock (US)

Wacker Chemie (US and Germany)

OCI (Korea)

GCL Poly (China)

TBEA (China)

Daqo (China)

Tongwei (China)

Average variable cost (USD/kg)

27


2. It generates synergy as a lot of the tasks along the supply chain such as marketing and

service are repetitive.

3. It allows upstream players to know and address market demand better, and to develop

business objectives in line with that of the users.

Given these benefits, top solar supply chain players including Longi, Zhonghuan, Tongwei, JA

Solar, Trina Solar, GCL Poly, Canadian Solar, Jinko Solar, and Risen Energy, have been

extending their businesses along the supply chain.

Longi expanded into solar cell and module manufacturing by acquiring Lerri Solar in 2014. As of

end-2018, the company has a module capacity of 8.8GW. It supplied 7.3GW of modules in

2018, or a 7% market share globally. As the largest mono wafer supplier globally, the company

is able to self-supply wafers to its modules, which leads to lower costs and better margins for its

modules compared with that of its peers (Exhibit 48).

Longi’s weak spots, in our view, are polysilicon and cell. The company does not produce polysilicon

and its 6GW cell capacity at end-2018 can hardly satisfy its own wafer supply of 18GW in 2018. As

such, co-operation with polysilicon and cell players becomes critical. On 3 June 2019, Longi

announced a strategic co-operation with Tongwei in polysilicon and wafer. According to the

announcement, Tongwei will take a 30% stake in Longi's 15GW mono wafer project in Yinchuan,

while in return Longi will take a 30% stake in Tongwei's 50,000 tonnes Baotou polysilicon project.

We believe the partnership is beneficial to both parties as it helps Longi to secure upstream

polysilicon supply for its wafers and helps Tongwei to secure a major customer for its polysilicon.

Longi’s Yinchuan project, slated for completion around 2020, will need to consume c50,000 tonnes

of polysilicon per annum, an exact match for the production at Baotou Ph1-2.

Zhonghuan moved into cell and module manufacturing by acquiring a 90% stake of Gdsolar in

July 2018. The RMB644m consideration was settled via issuing 84m shares to the seller at

RMB7.67/share. Zhonghuan will build 5GW of shingled-cell module capacity on Gdsolar’s

existing plant and upgrade Gdsolar’s existing HJT cell line.

Exhibit 48. PV module GM: Longi’s higher than peers thanks to its self-supplied wafers


Solar technology trends

The PV panel manufacturing process can be broken down into four parts: polysilicon, wafer,

cell, and module. There have been competing technologies in each part, like mono-Si vs. multi-

Si in wafer processing, and p-type vs. n-type in cell production. In this section, we analyze the

27%

31%

24%

18% 17% 16%

13%

6%

0%0%

5%

10%

15%

20%

25%

30%

35%

2016 2017 2018

Longi Chint CECEP Solar Energy

PV module GM

Top solar supply chain

players have been extending

their businesses along the

supply chain


28

technology trends of various parts of the manufacturing process as we view the technology

trends as one of the biggest uncertainties for companies in the solar supply chain. We believe

there is more uncertainty concerning the technology trends of cell and module than with wafer

and polysilicon. It is difficult to identify the winner in the cell industry as there are at least five

competing cell technologies to choose from. In our view, in order of technology risk, cell >

module > polysilicon > mono wafer.

How does a solar cell work? When the sunlight hits the surface of a solar cell, photons

are absorbed as energy. This activates the electrons breaking free from the silicon atoms,

creating pairs of negatively-charged electrons and positively-charged electron hole. By

adding tiny amounts of the right elements to the top and bottom sides of a silicon wafer, a

P-N junction can be formed that stops electrons that are knocked below it from going back

to the atoms they came from. The roaming electrons are then attracted to the positively

charged area N and the electron holes to the negatively charged area P, developing an

electric voltage across the cell.

Exhibit 49. An illustration of the working mechanism of a solar cell


Polysilicon purification: modified-Siemens method has become the mainstream

Polysilicon is the upstream of the solar supply chain. Polysilicon can be produced from

metallurgical-grade silicon through chemical purification or physical methods. At present, the

modified-Siemens method (a chemical method), adopted by over 90% of the polysilicon

production facilities globally, is the mainstream production method. We see it as the most

competitive method. The modified-Siemens method produces the high quality polysilicon (9-

11N) required for mono wafers. Compared with the Fluidized bed reactor method, a traditional

chemical method, the modified-Siemens method has lower production cost and higher levels of

safety. We expect the modified-Siemens method to remain the mainstream way of producing

polysilicon in the medium term.

N-type silicon (P+)

P-type silicon (B-)

Back electrode (+)

Anti-reflection coating

Front electrode (-)Sunlight

Current

29


Wafer: Choosing between multi-Si brick and mono-Si rod

The key difference in the production of multi-Si and mono-Si wafer is in ingot processing – a

step before wafer cutting (Exhibit 53). Mono-Si products are pulled into rods under the

Czochralski (CZ) method, while multi-Si products are casted into bricks. The furnaces used for

mono-Si and multi-Si are very different and cannot be shared.

As the average efficiency of a mono-based panel is 10-15% higher for a similar price point,

mono panels are gaining popularity among users. This, however, was not the case back in 2015

when multi-Si products still had more than 80% of the market share thanks to its lower cost. The

market landscape started to shift in 2016 after Longi developed the diamond wire sawing (DWS)

technology, which was adopted industry-wide in 2017. DWS enables thinner slicing and lowered

the cost of mono-Si wafer cutting far more than it did for the multi-Si counterparts. In 2018,

mono-Si wafer accounted for 45% of Si cell production and the China PV Industry Association

forecasts that this will grow to 69% by 2021.

Improvements in technology are far from over. We believe future cost reduction could come from

mono c-Si via the technology that allows multiple-crystals to be pulled from a single crucible. The

HSBC Asia Utilities & Alternative Energy team also noted the shift of technology trends towards

mono wafers in their report, GCL-Poly Energy Holdings (3800 HK), 30 March 2019.

Exhibit 50. Polysilicon purification: modified-Siemens method has become a mainstream way


Chemical purification method

Modified Siemens method

Energy beam method

CP methodOther metallurgical

methodFluidized bed reactor

method

Physical method

Poly-Crystalline Silicon Purification


30

Exhibit 51. Share of mono-Si has risen from 24% in 2013 to 45% in 2018

Exhibit 52. Longi has managed to cut non-Si cost by 72% since 2013, and we expect the cuts to continue

Source: CPIA, HSBC Qianhai Securities estimates Source: Company data, HSBC Qianhai Securities estimates

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2015 2017 2019E 2021EMulti-Si Mono-Si

-

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

2013 2015 2017 2019E 2021E

non-Si cost

RMB/W

Exhibit 53. Mono rods vs multi bricks is the key difference in the production of mono and multi solar panels


Mono-silicon wafer

Polysilicon

Multi-silicon cell

Multi-silicon wafer

Power generation

Multicrystal ingot casting process

Mono-silicon PV module

Monocrystal ingot pulling process

Multi-silicon PV module

Mono-silicon cell

Mono-like crystal (what GCL-Poly is producing)

As multi-si, its appearance and performance are close to that of the mono-si

Higher cost but less impurities; overall higher conversion efficiency from sunlight into electricity

Diamond wire sawing (DWS) technology was adopted industry-wise in 2017 and lowered cost

Texturing and polishing process with alkaline solution has lower cost

Lower cost but more impurities; overall lower conversion efficiency from sunlight into electricity

Complete industry-wise adoption of DWS in 2019; but cost reduction is not as significant as that in mono-si wafer production

Texturing and polishing process with acid solution has higher cost; Mono-like crystal can also be treated with alkaline solution

Production process is similar and equipment highly replaceable

31


Cell: Five key technologies

PERC: PERC (Passivated Emitter and Rear Cell) differs from the standard cell architecture

in that, at the rear surface, it has a passivation layer formed by SiNx or Al2O3 with tiny

pockets or holes (Exhibit 54). This film allows the transmitted photons to be reflected back

to the silicon layer for a second chance to get absorbed. This structure can increase

conversion efficiency by 0.6-1.0ppts for multi-Si cells and 1.0-1.5ppts for mono-Si cells.

Compared with other technology, the adoption of PERC on existing cell lines is relatively

easy as it only requires two extra manufacturing steps – back passivation and laser

grooving. Given these benefits, the PERC cell share of total c-Si cell production has

increased from 9% in 2015 to 34% in 2018. And that could grow to 61% in 2021, according

to the China PV Industry Association’s forecasts.

Exhibit 54. Cell structure comparison: PERC vs Traditional P-type cell


N-PERT: N-PERT (N-type Passivated Emitter Rear Totally Diffused Cell) is an n-type mono

cell. An n-type cell has the boron-doped Si wafer on the front, and the phosphorus-doped

wafer on the back, the opposite to the arrangement of a p-type cell (Exhibit 55). Similar to the

PERC design for p-type cells, N-PERT cells also have passivation layers. While N-PERT cells

are relative easy to produce, their yield relative to the costs is not as good as the double-sided

PERC cells. In 2018, N-PERT accounted for 5% of global c-Si cell production.

P-Si P-Si

N+ emitter (phosphorus)

P+ (Boron BSF)

BSF: Back Surface Field

Ag electrode

Ag electrode

Ag electrode

SiN/SiO2Al2O3 layer

SiN capping layer

Ag electrode


Traditional P-type cell Mono PERC cell

Screen-printed Al-paste

P+ （Boron BSF）

Passivation layer


32

Exhibit 55. Cell structure comparison: P-type cell vs N-type cell


HJT: HJT (Heterojunction) is another n-type mono technology that combines the advantage

of c-Si cells and thin film technologies, enabling solar cells to achieve higher degree of

efficiency. The key differentiating part of this technology is that it applies thin layers of doped

and intrinsic amorphous silicon on both sides of the n-type mono-Si layer. The amorphous

layer has wider bandwidth that catches more photons and reduces the number of electrons

lost on the surface. In addition, a transparent and conductive oxide layers (TCO) is placed

on the surface to better absorb the generated power. We see a promising outlook for this

technology given its high conversion efficiency (26% in laboratory trials). However,

equipment prices are high and that could hinder mass production. Cell manufactures,

including Tongwei, Panasonic, Sunpreme, Jinergy, CIE Power, and GS-solar, are all eyeing

this area. In 2018, HJT accounted for only 1% of global c-Si cell production.

P-Si N-Si


P+ (Boron BSF)


N+ (phosphorus BSF)

Ag electrode

Ag electrode

Ag electrode

Al2O3 layer

SiN/SiO2Al2O3 layer

SiN capping layer

Ag electrode

P+ (Boron)

P-type cell N-type cell

33


Exhibit 56. Cell structure comparison: HJT vs traditional N-type cell


TOPCon: In Topcon (Tunnel Oxide Passivated Contact) cell, a thin oxide layer is

introduced to the back of the cell in addition to a layer of heavily doped Si wafer. These

layers passivate the back of the cell and reduce the recombination loss significantly.

TOPCon cell is very hard to mass produce and the potential for conversion efficiency

improvement is likely to be lower than that of HJT. So far, among the players who had

made production plans, only LG Chemical can mass produce TOPCon cells.

IBC: IBC (Interdigitated Back Contact) is one of the most complicated technologies among

all the viable alternatives but also offers great potential for efficiency improvement. In

traditional solar cells, metal contacts are placed on the front side of the cell. For IBC cells,

the contacts are placed on the rear side of the cell. This allows IBC to achieve higher

efficiency than the traditional cells given reduced shading on the front side. An IBC cell can

achieve a conversion efficiency as high as 23.5-24.5%, but is costly and hard to produce.

Players like Sun Power and LG Chemical are setting foot in this area.

N-Si N-Si

P+ (Boron)

N+ (phosphorus BSF)


N+ (phosphorus BSF)

Ag electrode

Ag electrode

Ag/Al electrode

TCO layers on the back

SiN/SiO2Al2O3 layer

SiN capping layer

Ag electrode

P+ (Boron)

HJT (Heterojunction cell) N-type cell

TCO (transparent and conductive oxide) layers on the front

a thin layer of doped and intrinsic amorphous silicon

a thin layer of doped and intrinsic amorphous silicon


34

Exhibit 57. Cell structure comparison: IBC vs traditional N-type cell


Exhibit 58. Market share of key cell technologies, 2018-21e

Source: CPIA, HSBC Qianhai Securities, E= HSBC Qianhai Securities estimates

N-SiN-Si

N+ N+ （phosphorus BSF）

texture plus anti-reflection coating

negativecontact

Ag/Al electrode

SiN/SiO2Al2O3 layer

SiN capping layer

Ag electrode

P+ （Boron）

IBC (Interdigitated Back Contact) N-type cell

Lightly doped front diffusion layer: reduces recombination loss

P+ N+ P+ N+ P+

positive contact

Passivating SiO2 layer: reduces surface recombination loss

Backside mirror: reduces backlight absorption and traps light

Recombination loss: losses incurred when electrons and electron holes recombine

5% 8% 10%15%

34%

51%56%

61%

60%

39%30%

17%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2018 2019E 2020E 2021E

Others IBC HJT N-PERT PERC BSF

35


Exhibit 59. Solar cell: Roadmap of key technologies


Exhibit 60. Solar cell: Pros and cons of different technologies

P-Mono PERC N-PERT N-TOPCon HJT IBC

Conversion efficiency 21.5-22% 21.5-21.7% 22.5-23% 22.5-23.5% 23.5-24.5% Capacity ~63GW ~2.1GW ~2GW ~3.8GW ~1.5GW Suppliers

Mainstream Jolywood, Linyang LG Panasonic, Sunpreme

LG，SunPower

Technology difficulties Low Medium-to-low High Medium Highest Mass-scale production difficulties

Low Medium Only LG can produce in mass-scale

Medium High

Production procedures Less Relatively less Many Least Most Equipment spending Low Medium-to-low High High Highest

Compatibility with existing lines Compatible Compatible Partly compatible Completely

incompatible Completely incompatible

Conclusion

Weakness Ceiling to efficiency improvement

Not as cost-effective as bifacial p-PERC cell

Difficult to produce in mass-scale

High equipment cost

Highest production difficulty and cost

Strength Cost-effective Can convert from existing line

Partly compatible with existing line

Least production procedures

Highest conversion efficiency

Source: PV InfoLink, HSBC Qianhai Securities

Mono

N-Type

P-Type

IBC

HJT

PERT

PERC

TOPCon

SE

HBC

Multi P-Type PERC Black Silicon Mono-like Crystal?


36

Module: The evolution of the array of cells

Half-cut solar cells: Half-cut solar cells are traditional silicon solar cells that have been cut

in half (Exhibit 61). When solar cells are halved, the current generated from each cell is

also halved, leading to lower energy losses. As a result, power output of a half-cut solar cell

is 2-4% higher than that of a full piece. In 2018, half-cut solar cells accounted for 5% of the

global module production, and the China PV Industry Association forecasts that will grow to

40% by 2028.

Shingled cell module: A shingled cell module breaks a solar cell further into 4-5 pieces

and places them in a shingled manner like roof tiles. This structure requires little need for a

welding belt and reduces the spaces between the solar cells (Exhibit 62). As a result, cell

units on the surface can be increased by 13%. In addition, the removal of the busbar, a

metal wire grid on top of the cell, increases the light-receiving area by 5-6%. Instead of

having wires run on the surface, a conductive adhesive is used to connect cells and make

electrical contacts required for current to flow.

Exhibit 61. Cutting a solar cell in half can increase power by a maximum 4%

Exhibit 62. Shingled cell module increases the light-receiving area

Source: HSBC Qianhai Securities Source: HSBC Qianhai Securities

Exhibit 63. Market share of half-cut and shingled-cell to grow at the expense of full-slice cells (2018-21e)

Exhibit 64. Various layers of a solar module

Source: CPIA, HSBC Qianhai Securities. E= HSBC Qianhai Securities estimates Source: HSBC Qianhai Securities

Traditional module:

Shingled cell module:

1% 4% 9% 14%8%17%

26%

35%

92%80%

66%51%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2018 2019E 2020E 2021E

Shingled-cell Half-cut Full-slice

Glass

EVA

EVA

Cell

Backsheet

37


Summary

We are initiating coverage of Longi, Tongwei, Zhonghuan, Jingsheng and FAM. In this section,

we compare the five companies. We believe that they are comparable as they all generate the

majority of their revenue from various parts of the solar supply chain. We compare them on: 1)

revenue exposure; 2) valuation; 3) growth potential; 4) cash-generating ability; 5) return on

equity; 6) balance sheet strength; 7) receivable impairment risk; and 8) R&D capability.

In order of preference, we like Longi, followed by Tongwei, Zhonghuan, Jingsheng and FAM.

We like Longi’s for its superior ROE, strong balance sheet, and solid growth potential

underpinned by wafer market share gains. We like Tongwei for its attractive valuation, low

account receivable days, and cost leadership in polysilicon and cell.

1. Revenue exposure: We divide PV manufacturing into upstream polysilicon, midstream

wafer and cell, and downstream module and packaging material. In the order of preference,

we like wafer > polysilicon > cell > module > packaging material. We like wafer the most

given its relatively consolidated market, high investment barrier, and less risks associated

with the technology roadmap. It is also easier for wafer leaders to get scale advantage

given high share of fixed cost in the cost structure. We like packaging material the least

given its susceptibility to cell efficiency improvements and to module makers’ cash

situation. As shown in Exhibit 65, Longi, Zhonghuan, and Jingsheng are all well exposed to

wafer manufacturing, while FAM is predominantly a packaging material supplier.

Exhibit 65. China solar peers: Revenue exposure breakdown, 2018

Longi Zhonghuan Tongwei Jingsheng FAM

Upstream (Polysilicon) 12% Midstream (Wafer, crystal-growing equipment) 28% 78% 84% Midstream (Cell) 2% 10% 28% Downstream (Module) 60% Downstream (Packaging material) 97% Power generation 4% 2% 2% 1% Others 7% 10% 57% 16% 2% Total 100% 100% 100% 100% 100%

Source: Company, HSBC Qianhai Securities

2. Valuation: On 2019e PE, the peer group trades at a range of 17-51x with a median of 20x.

Tongwei and Jingsheng are both trading below the median, at 17x and 19x respectively.

Zhonghuan and FAM, however, are valued above the median, at 51x and 25x.

Our stock-picking framework

We compare five PV product suppliers on eight metrics

Longi looks strong overall

Tongwei’s valuation is the lowest among peers

Our order of preference is

Longi, followed by Tongwei,

Zhonghuan, Jingsheng and

FAM


38

Exhibit 66. 2019e PE: Tongwei is the most attractive

Source: Company data, HSBC Qianhai Securities, e= HSBC Qianhai Securities estimates

3. Growth potential: We expect Zhonghuan to register the highest EPS CAGR in 2018-21e

(Exhibit 67), driven by capacity expansion in semiconductor (semi) and PV wafers. Longi

ranks second on EPS CAGR, fuelled by strong market share gains in mono wafer. FAM’s

2018-21e EPS CAGR at 3% is the lowest among peers. This is because we see limited

growth potential for the EVA film market to which the company is exposed.

Exhibit 67. EPS CAGR: Zhonghuan to register the fastest growth in 2018-21e

Source: Company data, HSBC Qianhai Securities, e= HSBC Qianhai Securities estimates

4. Cash-generating ability: We use cash generation days and operating cash flow (OCF) to

net profit to compare the cash flow strength for the peer group. Zhonghuan has the lowest

cash generation days and the highest OCF to net profit multiple among peers in 2018

(Exhibits 68-69). We believe this is due to a better working capital cycle as the result of its

stronger bargaining power in payment terms for suppliers. This is validated by the long

cash generation days of Jinsheng, one of Zhonghuan’s key supplier in crystal-growing

machines. However, Zhonghuan’s FCF to net profit multiple was the lowest among peers in

2018, due to heavy capex.

50.6

24.719.9 19.3 17.1

0.0

10.0

20.0

30.0

40.0

50.0

60.0

Zhonghuan FAM Longi Jingsheng Tongwei

2019 PE

PE

41.8%

30.2%27.3%

24.7%

2.9%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

Zhonghuan Longi Tongwei Jingsheng FAM

EPS CAGR (2018-21)

YoY

39


Exhibit 68. Cash generation days (2018): Zhonghuan is the lowest


Exhibit 69. Operating cash flow to net profit (2018): Zhonghuan is the strongest


5. ROE: On average, the peer group generated 12.3% ROE in 2018. Longi’s ROE was 15.5%

in 2018, the highest among peers. We believe this is due to the company’s higher asset-to-

equity leverage (2.4x in 2018) stemming from its large amounts of transactions (sales and

purchases) with other players in the supply chain. The company supplies wafers to cell

makers and procures the cells back for module production. This results in very similar

amounts of trade receivables and trade payables on its balance sheet.

(103)

(53)

30

144

201

(150)

(100)

(50)

0

50

100

150

200

250

Zhonghuan Tongwei Longi Jingsheng FAM

2018

cash days

2.70

1.54

0.46 0.28 0.23

-6.41

-1.66-0.78 -0.49

0.43

-8.00

-6.00

-4.00

-2.00

0.00

2.00

4.00

Zhonghuan Tongwei Longi Jingsheng FAM

OCF to net profit (2018) FCF to net profit (2018)


40

Exhibit 70. Return on equity (2018): Longi is the highest


6. Balance sheet strength: We use a net debt to total equity ratio to compare balance sheet

strength. As shown in Exhibit 71, Jingsheng, FAM, and Longi all have net cash on their

balance sheets in 2018. Zhonghuan had a net debt to total equity ratio of 55% in 2018, the

highest among peers. However, we expect the ratio to drop to 27% in 2019 upon the

completion of the RMB5bn private placement.

Exhibit 71. Net debt to equity (2018): Jingsheng, FAM, and Longi are all net cash


7. Receivable impairment risk: The earnings risk of PV product suppliers is highly

dependent on the susceptibility of receivables to impairment. As shown in Exhibit 72,

Jingsheng and FAM both had account receivable (AR) days over 200 days in 2018, well

above the average of 90 days for the other three companies. We believe Jingsheng’s

higher AR days is the result of its reliance on Zhonghuan as its key customer. FAM’s higher

AR days is likely driven by delayed payments by downstream module makers, of which the

cash flows are poor.

15.5%14.3% 13.7% 13.5%

4.7%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

Longi Jingsheng Tongwei FAM Zhonghuan

2018

ROE

-10.3% -8.6%-2.5%

22.2%

54.7%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

Jingsheng FAM Longi Tongwei Zhonghuan

2018

Net debt / total equity (2018)

41


Exhibit 72. Account receivable days (2018): Tongwei has the lowest number


Exhibit 73. Receivable provision rate (2018): Longi has the lowest rate


8. R&D capability: As shown in Exhibit 74, Jingsheng’s R&D expense as percentage of

revenue was 7% in 2018, topping that of industry peers. We believe such a high R&D

expense ratio supports Jingsheng’s leading market position in crystal-growing machines.

Tongwei’s R&D expense as percentage of revenue was only 2% in 2018. However, after

excluding the agriculture business which accounted for c60% of revenue and requires

limited R&D, the R&D expense ratio was 5% in 2018.

228

202

140

101

29

0

50

100

150

200

250

Jingsheng FAM Longi Zhonghuan Tongwei2018

AR days

10.2%

6.5% 6.2%

2.6%1.7%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

Jingsheng Tongwei FAM Zhonghuan Longi2018

Receivable provision rate


42

Exhibit 74. R&D expense as % of revenue (2018): Jingsheng is the highest


7.2%

3.7%3.0%

2.2%

0.9%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

Jingsheng FAM Zhonghuan Tongwei Longi

2018

R&D expense as % of revenue

43


Company section


44

Investment summary

Cost leader in mono wafer production

As the largest global mono wafer producer, Longi is well positioned to benefit from the trend of

rising share of mono in PV panel production. The company has been a cost leader in the wafer

industry, cutting its non-Si cost significantly by 72% since 2013. We expect further declines of c40%

in 2018-21e, driven by economies of scale. This, together with the addition of 37GW new wafer

capacity in 2019-21, should lead to its market share rising from 35% in 2018 to 42% in 2021e.

The stock price has rallied 58% y-t-d, outperforming the benchmark index by 36% on strong

export volume. However, we believe the market has yet to factor in a better long-term growth

prospects, driven by lower demand volatility associated with policy changes as the industry

approaches grid parity.

Exhibit 75. Longi: We expect earnings to rise at a 43% CAGR in 2018-21e

Exhibit 76. Longi: Revenue breakdown, 2018



-50%

0%

50%

100%

150%

200%

250%

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

2016 2017 2018 2019E2020E2021E

Net profit YoY

Rmb mn

PV Modules

60%

Mono-crystalline

wafer28%

Power generation

4%

Solar Cell2%

PV system0%

Mono c-Si rods1%

PV EPC3%

Others2%

Longi Green (601012 CH)

The largest mono wafer maker globally, with a 35% market share

Cost leadership and new capacity addition should drive 43% net

earnings CAGR in 2018-21e

Initiate with a Buy rating and a TP of RMB36.20

The company has been a

cost leader in the wafer

industry

45


Investment positives

Dominant mono wafer producer globally

With a 35% market share in 2018, Longi is the largest producer of mono wafer in the world. We

expect its market share to rise to 42% in 2021e in line with rapid capacity expansions. As shown

in Exhibit 80, mono has become a more widely used wafer in the last few years, given higher

conversion efficiencies (Exhibit 79) and the evolution of the process flow that reduced costs

(down 30% in 2016-17). The trend is likely to continue – the China PV Industry Association

forecasts mono to account for 69% of all Si cell production in 2021, up from 45% in 2018.

Exhibit 77. Mono wafer: Longi is the largest supplier with a 35% market share in 2018

Exhibit 78. Longi: To gain market share significantly over 2019-21e

Source: Company data, HSBC Qianhai Securities Source: CPIA ,Company data, HSBC Qianhai Securities，E= HSBC Qianhai

Securities estimates

Exhibit 79. Conversion efficiency: Mono cell is higher

Exhibit 80. Global Si cell production by substrate: Share of mono to rise over 2019-21e

Source: CPIA, HSBC Qianhai Securities Source: CPIA, HSBC Qianhai Securities, E = CPIA estimates

Sales growth rising with capacity expansion in 2019-21e

As of the end 2018, the company has 28GW of wafer production capacity. It plans to increase

this to 36GW in 2019, 45GW in 2020e, and 65GW in 2021e. Of the 17GW additional capacity in

2019-20, 12GW will be from the new capacities in Yunnan (slated for completion in mid-2019)

Longi35%

JinkoSolar11%

Zhonghuan29%

Others25%

0%5%

10%15%20%25%30%35%40%45%50%

2015 2016 2017 2018 2019E 2020E 2021E

Mono-Si wafer Global wafer

Longi's Market share

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

2013 2014 2015 2016 2017

Mono-Si cell Multi-Si cell

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2015 2017 2019E 2021EMulti-Si Mono-Si

Longi is the largest producer

of mono wafer in the world


46

and 5GW from upgrading the existing production facilities. Hence, we expect wafer production

to rise at a CAGR of 35% in 2018-21e.

Exhibit 81. Longi: Capacity by segment, 2016-21e

Exhibit 82. Longi: Wafer production to register a 35% CAGR in 2018-21e



The entry barrier of mono wafer manufacturing is high

While we are concerned about rapid capacity expansion along the solar supply chain, we

believe mono wafer has probably the least to worry about given its relatively high investment

and technology entry barriers. As shown in Exhibit 83, the prevailing investment cost for 1GW of

mono wafer production capacity is around RMB500m, higher than that of PERC cell and almost

6x that of PV module. In addition, the market is relatively consolidated with the top three

producers – Longi, Zhonghuan, and JinkoSolar – accounting for an aggregate c80% share of

the market. We believe this reduces the risk of technology leakage to new entrants.

Exhibit 83. The Investment cost of mono wafer is high compared with that of other parts of the solar supply chain


Strong balance sheet to support future growth

As of end 2018, the company had a net cash balance of RMB421m on hand. While capex could

rise in 2019 on capacity expansion, the company raised RMB3.9bn from a rights issue in April

2019, alleviating the balance sheet pressure. Hence, we expect its net gearing to remain

negative in 2019-21e. Of the RMB3.9bn proceeds raised, RMB2.5bn will be used to fund the

8

15

28

36

45

65

2 4 6 11

15 20

5 8 9

16

25 30

-

10

20

30

40

50

60

70

2016 2017 2018 2019E 2020E 2021E

Tho

usan

ds

Mono-Si Wafer Cell PV Module

GW

18

32 36

45

0%

10%

20%

30%

40%

50%

60%

70%

80%

-

5

10

15

20

25

30

35

40

45

50

2018 2019E 2020E 2021E

Tho

usan

ds

Mono-Si wafer sales YoY

GW

1,500

500 420 408

295

80

-

200

400

600

800

1,000

1,200

1,400

1,600

Cell (HJT) Wafer (Mono-Si) Cell (PERC) Polysilicon Wafer (Multi-Si) PV Module

Investment Cost

RMB mn/GW

47


investment in a 5GW mono cell project in Ningxia, RMB1.1bn to fund the 5GW PV module

project in Chuzhou, and the remainder to replenish working capital.

Exhibit 84. Longi: Details of the RMB3.9bn rights issue

Announcement date Aug-18

Issuance date Apr-19 Proceeds (RMBm) m shares RMB/share Rights issue 3,893 837 4.65 Uses (RMBm): Total investment Amt. raised Ningxia 5GW Mono-Si cell project 3,050 2,540 Chuzhou 5GW Mono-Si PV module project 2,262 1,060 Replenish working capital 300 300 Total 5,611 3,900


Investment concerns

Heavy capital commitment in 2019-21e

We expect the company to incur heavy capex of RMB8bn in 2019, and RMB7bn in both 2020

and 2021 due to the expansion of its wafer/cell/module capacities. This is likely to result in

significant investing cash outflows. We forecast its net cash / equity to reduce in 2020-21e after

an increase in 2019 (Exhibit 86).

Exhibit 85. Longi: FCF has been negative since 2015

Exhibit 86. Longi: Net cash / equity to reduce in 2020-21e after an increase in 2019e



Equity financing has been frequent since listing

Equity raising has been a key part of the company’s financing since its listing in 2012. This

inevitably dilutes the interests of existing shareholders. In general, equity financing is more

favourable than debt financing when the project risk is high, and vice versa. Due to reduced

policy risk as we move to grid parity, we expect debt raising to become more common.

(3.0)

(2.5)

(2.0)

(1.5)

(1.0)

(0.5)

0.0

0.5

1.0

1.5

2015 2016 2017 2018 2019E 2020E 2021E

Rm

b bn

FCF

-35%

-30%

-25%

-20%

-15%

-10%

-5%

0%

2015 2016 2017 2018 2019E2020E2021E

Net debt / Equity


48

Exhibit 87. Longi: Historical equity financing

Announcement date Financing via Total proceeds (RMBm)

2019-04-03 Right Issue 3,875 2017-10-31 Convertible bond 2,800 2016-09-10 Private placement 2,980 2015-06-26 Private placement 1,960 2012-03-26 IPO 1,575


The company’s cell operation has been a weak spot

PV modules consist of solar cells which are made from wafers. Hence cells are important

industrial links between wafers and modules. In terms of Longi’s position in wafer/cell/module,

the cell business has been a weak spot, achieving only 9% gross margin in 2018. As shown in

Exhibit 88, the gross margin of the cell business has underperformed that of the wafer and the

module businesses since 2016. The capacity of the cell operation at 6GW as of end-2018 is

also much smaller than that of wafer (28GW) and module (9GW). This means Longi will need to

rely on external cell makers like Tongwei for cell supply (for module) and for wafer procurement.

Exhibit 88. Longi: Solar cell achieved only 9% gross margin in 2018

Source: Company data

Financial forecasts

Earnings forecasts

We expect a 43% earnings CAGR in 2018-21e, compared with a 70% earnings CAGR in 2015-

18. We base our forecasts on the following key assumptions:

Revenue: We forecast a revenue CAGR of 31% in 2018-21e, driven by the strong growth

of the PV module segment (45% revenue CAGR in 2018-21e).

Gross margin: We forecast the gross margin to improve from 22.2% in 2018 to 27.5% in

2021e driven by rapid cost reduction in the wafer segment.

0%

5%

10%

15%

20%

25%

30%

35%

2015 2016 2017 2018

PV module Wafer Cell

GM

49


Exhibit 89. Longi: Segments and full P&L forecasts

RMBm 2018 2019e 2020e 2021e 2018-21e CAGR

Turnover 21988 29000 37451 49803 31% PV Modules 13091 15648 26232 39592 45% Mono-crystalline wafer 6116 11090 8981 8049 10% Power generation 797 800 776 700 -4% Solar Cell 522 0 0 0 -100% PV system 72 72 72 72 0% Mono c-Si rods 318 318 318 318 0% EMS 97 97 97 97 0% Polysilicon 82 82 82 82 0% Others 212 212 212 212 0% PV EPC 682 682 682 682 0% Gross Profit 4892 6551 10222 13687 41% PV Modules 3119 4294 7721 11429 54% Mono-crystalline wafer 995 1527 1786 1590 17% Power generation 503 505 489 442 -4% Solar Cell 49 0 0 0 -112% PV system 16 16 16 16 0% Mono c-Si rods 40 40 40 40 0% EMS 30 30 30 30 0% Polysilicon 4 4 4 4 0% Others 68 68 68 68 0% PV EPC 68 68 68 68 0% Gross Margin 22.2% 22.6% 27.3% 27.5% Business tax (117) (155) (200) (265) 31% Selling expenses (1,017) (1,342) (1,733) (2,304) 31% Admin. expenses (825) (1,088) (1,405) (1,868) 31% Asset impairment losses / Fair value changes (728) (304) (279) (437) -16% Other gain / (losses) 867 800 200 200 -39% Operating profit 3072 4463 6806 9012 43% Net finance charges (267) (295) (270) (296) 4% Share of JCE 62 62 62 62 0% Profit before taxes 2,867 4,230 6,597 8,778 45% Tax (301) (625) (980) (1,307) 63% Minorities (9) (12) (19) (26) 44% Pre-exceptional profit 2,558 3,592 5,598 7,445 43% Dividend to preferred shareholders and perpetual capital securities 0 0 0 0 Exceptionals 0 0 0 0 Net profit 2,558 3,592 5,598 7,445 43%

Source: Wind, company data, HSBC Qianhai Securities. E= HSBC Qianhai Securities estimates

Balance sheet and cash flow forecasts

We forecast the company to remain in a net cash position in 2019-21e, supported by the RMB3.9bn

rights issue in 2019. We expect the increase in OCF to offset higher capex in 2019-21e.

Exhibit 90. Longi: Net debt and cash flow forecasts

RMBm 2018 2019e 2020e 2021e

Net debt/(cash) (421) (2,045) (326) (612) Net debt to equity -2% -9% -1% -2% Cash from Operations 1,173 3,981 5,311 7,542 Cash from Investing (3,169) (5,600) (6,400) (6,400) FCF (1,996) (1,619) (1,089) 1,142



50

Valuation and risks

Target price of RMB36.20

Our TP of RMB36.20 is derived from our discounted cash flow (DCF) valuation model. It implies

a 2020e PE of 24x, slightly above the historical average valuation of 21x. We believe this is

reasonable in view of lower demand volatility associated with policy changes as the industry

approaches grid parity. Key assumptions in our DCF valuation model include:

Cost of equity (COE): We use a COE of 10.0%. This is derived from a risk-free rate of

2.5%, a market risk premium of 6.5%, and a beta of 1.16.

Cost of debt (COD): We assume the pre-tax cost of debt to be 5% and after-tax cost of debt

to be 4%. We use our 2020e debt-to-capital ratio of 20% as our long-term debt-to-capital ratio.

Operating cash flow to grow 11% per annum: We expect operating cash flow (before

changes in working capital) to expand at a CAGR of 11% in 2018-29e, reflecting solid growth in

demand.

Capital expenditure: We assume high capex of around RMB6bn per annum in 2019-21e,

reflecting the expansion of its wafer/cell/module capacities. We assume steady capex of

around RMB1-2bn per annum in 2022-29e, reflecting steady investment in maintenance.

Terminal growth rate at 2%, and we assume the company reaches a steady growth

period after 2029.

51


Exhibit 91. Longi: Discounted cash flow valuation

Rmbm 2015 2016 2017 2018 2019e 2020e 2021e 2022e

Profit after tax 521 1,551 3,549 2,567 3,604 5,617 7,471 8,068 YoY growth 198% 129% -28% 40% 56% 33% 8% Add: Depreciation & amortization 285 426 748 1,206 761 868 1,174 1,239 Net finance expense 88 83 326 371 295 270 296 313 Operating cash flow before W/C changes

893 2,060 4,623 4,144 4,660 6,755 8,941 9,621

Changes in working capital (656) (1,766) (2,860) (2,917) (617) (1,382) (1,337) Net operating cash flow 237 294 1,763 1,227 4,043 5,373 7,604 9,621 CAPEX (1,216) (2,152) (3,774) (3,169) (5,600) (6,400) (6,400) (1,500) Free Cash Flow (979) (1,858) (2,011) (1,942) (1,557) (1,027) 1,204 8,121 Discount Factor 1.00 0.92 0.84 Gross PPE 3,617 5,981 12,814 16,239 14,552 20,495 26,803 28,303 Depreciation Rate 8% 7% 6% 7% 5% 4% 4% 4% PV of FCF (1,027) 1,106 6,849 RMBn 2023E 2024E 2025E 2026E 2027E 2028E 2029E Terminal

Value Profit after tax 8,714 9,150 9,607 10,087 10,390 10,702 11,023 YoY growth 8% 5% 5% 5% 3% 3% 3% Add: Depreciation & amortization 1,306 1,373 1,441 1,509 1,578 1,648 1,718 Net finance expense 330 346 364 381 398 416 434 Operating cash flow before W/C changes

10,349 10,869 11,411 11,977 12,366 12,765 13,174

Changes in working capital 0 0 0 0 0 0 0 Net operating cash flow 10,349 10,869 11,411 11,977 12,366 12,765 13,174 CAPEX (1,515) (1,530) (1,545) (1,561) (1,577) (1,592) (1,608) Free Cash Flow 8,834 9,339 9,866 10,416 10,790 11,173 11,566 171,366 Discount Factor 0.77 0.71 0.65 0.60 0.55 0.51 0.46 0.46 Gross PPE 29,818 31,348 32,894 34,455 36,031 37,623 39,232 Depreciation Rate 4% 4% 4% 4% 4% 4% 4% PV of FCF 6,843 6,644 6,446 6,250 5,946 5,655 5,376 79,659 Summary of PV (Enterprise Value) 129,748 Less: Net debt (incl. perpetual) 2,045 Equity value 131,793 Less: Minority interest (385) Shareholder Equity Value 131,408 Total shares issued by year-end 2019 3,628 Per Share Value - RMB 36.2 Assumptions Risk free rate 2.5% ERPch 6.5% Beta 1.16 Cost of equity = RFR + BETA*ERPch 10.0% Cost of debt 5.0% Income tax 15% After tax cost of debt 4.3% Debt/Capital 20% WACC 8.9% Terminal Growth 2%



52

Exhibit 92. Longi forward PE: Trading close to the historical average

Exhibit 93. Longi forward PB: Trading above the historical average

Source: Wind, HSBC Qianhai Securities Source: Wind, HSBC Qianhai Securities

Exhibit 94. Longi: Earnings sensitivity to gross margin and revenue changes, 2020e

_____________________________________ Gross Margin ______________________________________ Revenue -3% -2% -1% 0% 1% 2% 3%

20% 0% 7% 14% 21% 28% 34% 41% 15% -4% 3% 9% 16% 22% 29% 35% 10% -8% -2% 4% 10% 17% 23% 29% 5% -13% -7% -1% 5% 11% 17% 23%

0% -17% -11% -6% 0% 6% 11% 17%

-5% -21% -16% -11% -5% 0% 6% 11% -10% -26% -21% -16% -10% -5% 0% 5% -15% -30% -25% -20% -16% -11% -6% -1% -20% -34% -30% -25% -21% -16% -12% -7%


Downside risks:

Weaker-than-expected PV demand: PV module and wafer accounted for 87% of Longi’s

revenue in 2018. Should global PV installations slow, we see downside risks to our

earnings forecasts.

Risk of equity dilution from fundraising exercise: Equity raising has been a key part of

the Longi’s financing since its listing in 2012. We see risks of equity dilution should the

company fail to fulfill its capex needs via debt financing.

Greater-than-expected cuts to wafer ASP: We expect mono wafer ASP to decline by

20%/10%/10% in 2019/20/21e, respectively. We see risks of larger-than-expected margin

erosion if the ASP cuts are greater than expected.

-

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0


PE Mean +1SD -1SD

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0


PB Mean +1SD -1SD

53


Financial statements

Year to 12/2018a 12/2019e 12/2020e 12/2021e

Profit & loss summary (CNYm)

Revenue 21,988 29,000 37,451 49,803

EBITDA 4,340 5,285 7,735 10,248

Depreciation & amortisation -1,206 -761 -868 -1,174

Operating profit/EBIT 3,134 4,524 6,868 9,074

Net interest -267 -295 -270 -296

PBT 2,867 4,230 6,597 8,778

HSBC Qianhai PBT 2,867 4,230 6,597 8,778

Taxation -301 -625 -980 -1,307

Net profit 2,558 3,592 5,598 7,445

HSBC Qianhai net profit 2,558 3,592 5,598 7,445

Cash flow summary (CNYm)

Cash flow from operations 1,173 3,981 5,311 7,542

Capex -3,169 -5,600 -6,400 -6,400

Cash flow from investment -3,169 -5,600 -6,400 -6,400

Dividends -669 -362 -359 -560

Change in net debt 1,112 -1,624 1,719 -286

FCF equity 604 -1,234 84 2,244

Balance sheet summary (CNYm)

Intangible fixed assets 237 232 226 220

Tangible fixed assets 14,116 18,960 24,498 29,730

Current assets 22,901 28,125 30,439 36,992

Cash & others 7,708 9,332 7,613 7,899

Total assets 39,659 49,784 57,693 69,533

Operating liabilities 15,896 18,879 21,531 26,460

Gross debt 6,938 6,938 6,938 6,938

Net debt -770 -2,394 -675 -961

Shareholders' funds 16,452 23,581 28,820 35,705

Invested capital 13,649 19,106 26,020 32,583

Ratio, growth and per share analysis

Year to 12/2018a 12/2019e 12/2020e 12/2021e

Y-o-y % change

Revenue 34.4 31.9 29.1 33.0

EBITDA -12.6 21.8 46.3 32.5

Operating profit -25.6 44.4 51.8 32.1

PBT -28.6 47.5 56.0 33.1

HSBC Qianhai EPS -48.6 21.7 37.9 33.0

Ratios (%)

Revenue/IC (x) 1.8 1.8 1.7 1.7

ROIC 24.0 23.6 25.9 26.4

ROE 16.7 17.9 21.4 23.1

ROA 7.1 8.1 10.5 11.7

EBITDA margin 19.7 18.2 20.7 20.6

Operating profit margin 14.3 15.6 18.3 18.2

EBITDA/net interest (x) 16.3 17.9 28.6 34.6

Net debt/equity -4.6 -10.0 -2.3 -2.7

Net debt/EBITDA (x) -0.2 -0.5 -0.1 -0.1

CF from operations/net debt

Per share data (CNY)

EPS Rep (diluted) 0.92 1.12 1.54 2.05

HSBC Qianhai EPS (diluted) 0.92 1.12 1.54 2.05

DPS 0.10 0.11 0.15 0.21

Book value 5.89 6.50 7.94 9.84

Valuation data

Year to 12/2018a 12/2019e 12/2020e 12/2021e

EV/sales 3.7 2.7 2.2 1.6

EV/EBITDA 18.5 14.9 10.4 7.8

EV/IC 5.9 4.1 3.1 2.5

PE* 24.2 19.9 14.5 10.9

PB 3.8 3.4 2.8 2.3

FCF yield (%) 0.7 -1.5 0.1 2.8

Dividend yield (%) 0.4 0.5 0.7 0.9

* Based on HSBC Qianhai EPS (diluted)

ESG metrics

Environmental Indicators [n/a] Governance Indicators 12/2018a

GHG emission intensity* [n/a] No. of board members 12

Energy intensity* [n/a] Average board tenure (years) 4.6

CO2 reduction policy [Yes] Female board members (%) 25

Social Indicators Board members independence (%) 25

Employee costs as % of revenues [n/a]

Employee turnover (%) [n/a]

Diversity policy [Yes]


* GHG intensity and energy intensity are measured in kg and kWh respectively against revenue in USD ‘000s

Issuer information

Share price (CNY) 22.30 Free float 73%

Target price (CNY) 36.20 Sector Energy Equipment

RIC (Equity) 601012.SS Country China

Bloomberg (Equity) 601012 CH Analyst Corey Chan

Market cap (USDm) 11,685 Contact +86 755 8898 3404

Price relative

Source: HSBC Qianhai Securities Note: Priced at close of 14 Jun 2019

3.80

8.80

13.80

18.80

23.80

3.80

8.80

13.80

18.80

23.80

2017 2018 2019

LONGI GREEN Rel to CSI 300 Index

Financials & valuation: LONGI GREEN Buy

54

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 95. Longi green: Company structure, April 2019


Public

Tianjin Zhonghuan Semiconductor

（002129.SZ)

EVA film

Public

Longi (601012 CH)

Other PV-relatedbusiness

Mono-crystalline wafer

68.65%

PV Modules Other

Li Chun'an Li Xiyan

11% 5.35%

Power generation

Li Zhenguo

15%

55


Investment summary

Cost advantage in cell and polysilicon to drive long-term growth

Tongwei is a global cost leader in both PERC cell and polysilicon production. Its cell processing

cost – at RMB0.23/W in 1Q19 – is 34% below the industry average, and we expect that to drop

to RMB0.19/W in 2021e. In polysilicon production, the company’s Baotou Ph1 and Leshan Ph1

projects have a unit overall cost of RMB50k/t, close to the bottom of the industry range. Around

60% of the company’s revenue in 2018 was from the agriculture business, which generates

steady cash flow. This should alleviate the capex burden of Tongwei’s PV business. We expect

the company to register a 27% earnings CAGR in 2018-21e, driven by cost reduction efforts

and capacity expansion in cell and polysilicon.

Exhibit 96. Tongwei: Revenue breakdown, 2018

Exhibit 97. Tongwei: We expect earnings to grow at a 27% CAGR in 2018-21e

Source: Wind, company data, HSBC Qianhai Securities Source: Bloomberg, company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates


Leading solar cell producer globally

As of end-2018, the company had 12GW capacity in solar cells. It plans to increase its capacity

to 20GW in 2019 with the completion of its new production facility in Meishan (6GW) and

Chengdu (2GW). The ramp-up of a new cell production line normally takes 2-3 months, much

shorter than that of the production line of polysilicon. Hence, we expect cell sales to see a 45%

PV cell and modules,

27%

Polysilicon and

chemicals, 12%

Power generation,

2%

Feed, 53%

Food processing, livestock & poultry

breeding, 5%

Other business, 1%

0%

20%

40%

60%

80%

100%

120%

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

2016 2017 2018 2019E 2020E 2021E

Net profit YoY

Rmb mn

Tongwei (600438 CH)

Cost-leader in PERC cell and polysilicon production

The agricultural business is a cash cow, supplementary to PV


We expect the company to

register a 27% earnings

CAGR in 2018-21e


56

CAGR in 2018-21e. Upon completion of the new production facility, Tongwei will have 15GW

capacity in PERC mono cell, or a 16% share of global capacity (Exhibit 99). We expect the

market share of PERC mono cell to rise from 34% in 2018 to 61% in 2021e (Exhibit 100), given

better light capture near the rear surface.

Exhibit 98. Tongwei: Solar cell sales volume to rise on a 45% CAGR in 2018-21e

Exhibit 99. Tongwei: PERC mono cell capacity to reach 15GW by end 2019

Source: Company, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates


Exhibit 100. Global cell market share: PERC to rise from 34% in 2018 to 61% in 2021e

Source: CPIA, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Cell processing cost at low-end of the industry range

Tongwei recorded around 30% gross margin for its solar cell in 1Q19, well above industry peers’

10-20%. We believe this stems from its lower cell processing cost. Cell processing costs

account for 44% of the production cost of solar cells. The company’s cell processing costs at

RMB0.23/W in 1Q19 are 34% below the industry average at RMB0.35/W (Exhibit 101). This

gives the company a significant cost advantage in solar cells. We expect Tongwei’s cell

processing costs to drop to RMB0.20/W in 2020e on economies of scale (Exhibit 102).

0

5

10

15

20

25

2016 2017 2018 2019E 2020E 2021E

Tho

usan

ds

Mono-Si Multi-Si

GW

0%

5%

10%

15%

20%

0

2

4

6

8

10

12

14

16

2018 2019E

Tongwei's capacity % of the global capacity

GW

5% 8% 10%15%

34%

51%56%

61%

60%

39%30%

17%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2018 2019E 2020E 2021E

Others IBC HJT N-PERT PERC BSF

57


Exhibit 101. Cell processing cost: Tongwei at low-end of the industry range

Exhibit 102. Cell processing cost: We expect further declines in 2019-21e

Source: Company data, HSBC Qianhai Securities Source: Company data, CPIA, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Polysilicon sales volume to see a major uplift in 2019 on capacity expansion

Polysilicon sales accounted for around 5-7% of Tongwei’s revenue in 2017-18. In 2018, the

company completed the Baotou Ph1 and Leshan Ph1 projects, adding 60,000 tonnes of new

capacity to its earlier capacity of 20,000 tonnes. In view of this, we expect polysilicon sales to

quadruple from 19,000 tonnes in 2018 to 70,000 tonnes in 2019. The company will consider

whether to invest in Baotou Ph2 and Leshan Ph2 projects later in 3Q19 (60,000 tonnes

capacity). The company expects the new capacity of Baotou Ph1 and Leshan Ph1 to achieve an

overall unit cost of RMB50k/t, 10% below that of the old capacity.

Exhibit 103. Tongwei: We expect polysilicon sales to reach 70,000 tonnes in 2019e

Exhibit 104. Polysilicon: Tongwei’s new capacity has a much lower production cost than most of the existing capacities

Source: Company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates


Funding difficulties easing with the RMB5bn convertible bond issuance

In April 2019, Tongwei issued an RMB5bn convertible bond, with a conversion price of

RMB12.44/share. The bond can be converted into 402m shares, or 10.4% of the total shares

outstanding. The conversion rights will become effective in September 2019. RMB2.7bn of the

RMB5bn proceeds will be used to fund the Baotou Ph1 project (25,000 tonnes), and the

remainder to fund the Leshan Ph1 project (25,000 tonnes).

0.270.30 0.30

0.40

0.23

0.28 0.27

0.35

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

Tongwei Aikosolar Risen Energy Industryaverage2018 1Q19

RMB/W

-

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

2017 2018 2019E 2020E 2021E

Tongwei Industry average

RMB/W

12 16 19

70

80

90

-

10

20

30

40

50

60

70

80

90

100

2016 2017 2018 2019E 2020E 2021E

Tho

usan

ds

Sales volume

k ton

52 52 48 48

40

-

10

20

30

40

50

60

DAQO GCL-Poly TBEA Tongwei(old

capacity)

Tongwei(new

capacity)

Tho

usa

nd

s

Polysilicon production cost'000 RMB/t


58

Exhibit 105. Tongwei: Details of the RMB5bn convertible bond issuance

Announcement date 16/12/2017

Issuance date 05/04/2019 CB Issuance (RMBm) 5000 Conversion price (RMB) 12.44 Shares (m) 402 Project (RMBm) Total Inv. Amount raised from the CB issuance

Baotou polysilicon project Ph1 3,229 2,650 Leshan polysilicon project Ph1 3,184 2,350 Total 6,413 5,000


Exhibit 106. Tongwei: Net gearing to drop in 2020-21e

Exhibit 107. Tongwei: FCF to improve in 2020-21e



Investment concerns

Polysilicon could face overcapacity in the medium term

HSBC’s China solar team expects annual global polysilicon production to reach 626,000 tonnes

by 2020e, based on the current expansion plans of global polysilicon suppliers. This will be

higher than the expected global demand of 491,000 tonnes at the time. Hence, we expect

pricing pressure on Tongwei’s polysilicon business in 2019-20e. In our model, we assume

Tongwei’s polysilicon ASP drops 15% in 2019e, and drops 5% in both 2020e and 2021e.

0%

10%

20%

30%

40%

50%

60%

70%

2015 2016 2017 2018 2019E 2020E 2021E

Net debt / Equity

-8

-6

-4

-2

-

2

4

6

2015 2016 2017 2018 2019E 2020E 2021E

Rm

b bn

FCF

59


Exhibit 108. HSBC global demand & supply forecast for polysilicon

Source: Wind, Solarzoom, e = HSBC Global Research estimates

Exhibit 109. Polysilicon: ASP dropped 53% since early 2018

Exhibit 110. Tongwei: Polysilicon gross margin to decline in 2019 on price drop

Source: Wind, HSBC Qianhai Securities Source: Company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

HJT cell coating technology has a brighter future than that of PERC

Except for 200MW capacity in HJT (Heterojunction) cell, the majority of Tongwei’s 20GW cell

manufacturing capacity by end-2019 will be in PERC. While PERC (Passivated Emitter Rear

Contact) cell is likely to be the mainstream solar cell technology in the near term, there are

certain limitations with the technology which affect its potential for further increases in

conversion efficiency. Currently, the conversion efficiency of PERC cells is around 21-22%.

Comparatively, HJT (Heterojunction) cells achieve higher conversion efficiencies of 22-23%.

Hence, we see a brighter future for the HJT cell coating technology than that of PERC. The

production methodologies of HJT and PERC cells are very different and conversion between

the production lines is almost impossible. Hence, we see risks to Tongwei’s investment in PERC

should HJT becomes a mainstream cell coating technology.

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019e2020e2021e

To

nn

es

Total production Total demand

0

20

40

60

80

100

120

140

160


Rmb/kg

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

2017 2018 2019E 2020E 2021E

Polysilicon

GM


60

Exhibit 111. The conversion efficiency of multi-junction cell (incl. HJT) is higher than mono/multi-Si cell (incl. PERC)

Exhibit 112. The conversion efficiency premium of HJT cell over PERC is likely to sustain over time

Source: CPIA, HSBC Qianhai Securities Source: CPIA, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Asset-heavy model entails high D&A expenses

Tongwei has an asset-heavy business model which requires continuous capital outlays to

maintain its market position. As a result, depreciation and amortization (D&A) expenses have

been high in the last three years, about 4% of the company’s revenue.

Financial forecasts

Earnings forecasts

We expect a 27% earnings CAGR in 2018-21e. We base our forecasts on the following key

assumptions:

Revenue: We forecast a revenue CAGR of 13% in 2018-21e, driven by strong growth of

the PV cell and module segment (29% revenue CAGR in 2018-21e).

Gross margin: We forecast the gross margin to improve from 19% in 2018 to 23% in

2021e on rapid cost reductions of the PV cell and modules segment.

46.0%

33.3%

26.7%22.3%

0%5%

10%15%20%25%30%35%40%45%50%

III-V Multi-Junction

ConcentratorSolar Cell

III-V on Si (2terminal)

Mono-Si Multi-Si

Cell Efficiencies (2017)

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

2018 2019E 2020E 2021E

PERC mono-Si cell HJT cell

61


Exhibit 113. Tongwei: Segments and full P&L forecasts

RMBm 2018 2019e 2020e 2021e 2018-21e CAGR

Turnover 27,535 34,926 37,755 40,001 13% PV cell and modules 7,642 12,868 14,982 16,547 29% Polysilicon and chemicals 3,317 6,229 6,598 6,920 28% Power generation 620 1,108 1,355 1,506 34% Solar: Inter-segment elimination (1,575) (2,748) (3,119) (3,396) 29% Feed 15,236 15,693 16,164 16,649 3% Food processing, livestock & poultry breeding 1,397 1,397 1,397 1,397 0% Other business 377 377 377 377 0% Inter-segment elimination (131) 0 0 0 -100% Gross Profit 5,208 7,568 8,684 9,124 21% PV cell and modules 1,429 2,939 3,810 4,074 42% Polysilicon and chemicals 1,183 1,805 1,882 1,935 18% Power generation 381 681 833 926 34% Solar: Inter-segment elimination (224) (390) (443) (482) 29% Feed 2,225 2,292 2,361 2,431 3% Food processing, livestock & poultry breeding 69 69 69 69 0% Other business 172 172 172 172 0% Inter-segment elimination (28) 0 0 0 -100% Gross Margin 19% 22% 23% 23% Business tax (111) (141) (153) (162) 13% Selling expenses (863) (1,094) (1,183) (1,253) 13% Admin. expenses (1,637) (2,076) (2,244) (2,377) 13% Asset impairment losses / Fair value changes (48) (41) (16) (13) -36% Other gain / (losses) 148 59 59 59 -26% Operating profit 2,697 4,275 5,147 5,379 26% Net finance charges (316) (484) (486) (468) 14% Share of JCE 17 17 17 17 0% Profit before taxes 2,398 3,808 4,678 4,928 27% Tax (367) (569) (699) (737) 26% Minorities (12) (20) (24) (25) 27% Pre-exceptional profit 2,019 3,220 3,955 4,166 27% Dividend to preferred shareholders and perpetual capital securities

0 0 0 0 NA

Exceptionals 0 0 0 0 NA Net profit 2,019 3,220 3,955 4,166 27%

Source: Wind, company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates


We forecast net gearing to increase to 65% in 2019e from 22% in 2018, driven by the capex for 8GW

new capacity in PERC-cell in Meishan and Chengdu and for 540MW new capacity in solar farms.

Exhibit 114. Tongwei: Net debt and cash flow forecasts

RMBm 2018 2019e 2020e 2021e

Net debt/(cash) 3,382 11,617 9,793 7,525 Net debt to equity 22% 65% 47% 32% Cash from Operations 3,100 5,870 6,500 7,153 Cash from Investing (6,442) (13,000) (3,200) (3,200) FCF (3,342) (7,130) 3,300 3,953



62

Valuation and risks



a 2020e PE of 19x, above the historical average valuation of 16x. We believe this is reasonable

in view of lower demand volatility associated with policy changes as the industry approaches

grid parity. Key assumptions in our DCF valuation model include:



Cost of debt (COD): We assume the pre-tax cost of debt to be 5% and after-tax cost of

debt to be 4%. We use our 2020e debt-to-capital ratio of 42% as our long-term debt-to-

capital ratio.


changes in working capital) to expand at a CAGR of 9% in 2018-29e, reflecting solid growth

in demand.

Capital expenditure: We expect a capex of RMB13bn in 2019e, driven by cell and solar

farm capacity expansion. Thereafter, we expect capex to drop to around RMB2-3bn per

annum in 2020-29e, reflecting steady maintenance capex.


period after 2029.

63


Exhibit 115. Tongwei: Discounted cash flow valuation

RMBm 2015 2016 2017 2018 2019e 2020e 2021e 2022e

Profit after tax 808 1,023 2,038 2,031 3,239 3,979 4,191 4,401 YoY growth 27% 99% 0% 59% 23% 5% 5% Add: Depreciation & amortization 611 841 969 1,218 1,309 1,770 2,187 2,278 Net finance expense 360 247 241 403 484 486 468 488 Operating cash flow before W/C changes 1,779 2,111 3,248 3,652 5,033 6,236 6,846 7,167 Changes in working capital 38 323 (264) (512) 855 282 324 Net operating cash flow 1,817 2,435 2,984 3,140 5,887 6,517 7,170 7,167 Capex (1,464) (4,393) (4,099) (6,442) (13,000) (3,200) (3,200) (2,000) Free Cash Flow 353 (1,959) (1,115) (3,302) (7,113) 3,317 3,970 5,167 Discount Factor 1.00 0.92 0.85 Gross PPE 11,307 14,035 18,481 24,292 30,996 42,737 47,645 49,645 Depreciation Rate 5% 6% 5% 5% 4% 4% 5% 5% PV of FCF 3,317 3,654 4,378 RMBm 2023e 2024e 2025e 2026e 2027e 2028e 2029e Terminal

Value Profit after tax 4,621 4,852 5,094 5,349 5,616 5,785 5,958 YoY growth 5% 5% 5% 5% 5% 3% 3% Add: Depreciation & amortization 2,375 2,476 2,582 2,694 2,811 2,934 3,063 Net finance expense 508 530 553 577 602 628 656 Operating cash flow before W/C changes 7,504 7,858 8,229 8,619 9,029 9,347 9,677 Changes in working capital 0 0 0 0 0 0 0 Net operating cash flow 7,504 7,858 8,229 8,619 9,029 9,347 9,677 Capex (2,100) (2,205) (2,315) (2,431) (2,553) (2,680) (2,814) Free Cash Flow 5,404 5,653 5,914 6,188 6,477 6,667 6,863 105,441 Discount Factor 0.78 0.72 0.66 0.61 0.56 0.52 0.47 0.47 Gross PPE 51,745 53,950 56,265 58,696 61,249 63,929 66,743 Depreciation Rate 5% 5% 5% 5% 5% 5% 5% PV of FCF 4,214 4,058 3,908 3,764 3,626 3,436 3,256 50,019 Summary of PV (Enterprise Value) 87,630 Less: Net debt (incl. perpetual) (11,617) Equity value 76,013 Less: Minority interest (509) Shareholder Equity Value 75,505 Total shares issued by year-end 2019 (mn) 3,882

Per Share Value - Rmb 19.4 Upside 28% Assumptions Risk-free rate 2.5% ERPch 6.5% Beta 1.42 Cost of equity = RFR + BETA*ERPch 11.8% Cost of debt 5.0% Income tax 15% After tax cost of debt 4.3% Debt/Capital 42% WACC 8.6% Terminal Growth 2%



64

Exhibit 116. Tongwei forward PE: Trading in line with the historical average

Exhibit 117. Tongwei forward PB: Trading above the historical average

Source: Wind, company data, HSBC Qianhai Securities Source: Wind, company data, HSBC Qianhai Securities

Exhibit 118. Tongwei: Earnings sensitivity to gross margin and revenue changes, 2020e


20% -7% 2% 12% 22% 31% 41% 51% 15% -11% -2% 7% 16% 26% 35% 44% 10% -16% -7% 2% 11% 20% 29% 38% 5% -20% -11% -3% 5% 14% 22% 31%

0% -24% -16% -8% 0% 8% 16% 24%

-5% -28% -21% -13% -5% 2% 10% 18% -10% -33% -25% -18% -11% -4% 4% 11% -15% -37% -30% -23% -16% -9% -3% 4% -20% -41% -35% -28% -22% -15% -9% -2%


Downside risks:

Weaker-than-expected polysilicon prices: Polysilicon business accounted for 10% of

Tongwei’s gross profit in 2018. Greater-than-expected declines in polysilicon prices due to

a supply glut could pose downside risk to our earnings forecasts.

Ramp-up of the new polysilicon capacity taking longer than expected: In our model,

we expect the new polysilicon project ramp-up to its face-plate capacity of 60,000 tonnes

by end-2019. Should the ramp-up take longer than expected, we see downside risks to our

2019 earnings forecasts.

Weaker-than-expected PV demand: Solar cells accounted for 28% of Tongwei’s earnings

in 2018. Should global PV installations slow, we see downside risks to our earnings

forecasts.


Tongwei’s financing since its listing in 2013. As the capex burden is likely to remain high at

around RMB13bn in 2019, we see risks of equity dilution should the company fail to fulfill its

capex needs via debt financing.

Evolution of the cell technology could undermine investment in PERC: A majority of

Tongwei’s existing cell capacity is in PERC, which is not compatible with competing

technologies like HJT. Hence a shift of technology trends from PERC to HJT could render

Tongwei’s existing PERC capacity useless.

-

5.0

10.0

15.0

20.0

25.0

30.0

35.0


PE Mean +1SD -1SD

0.0

1.0

2.0

3.0

4.0

5.0


PB Mean +1SD -1SD

65


Greater-than-expected cuts to cell ASPs: We expect solar cell ASPs to decline by

25%/10%/10% in 2019/20/21e. We see risks of larger-than-expected margin erosion if the

ASP cuts are greater than expected.


66


Year to 12/2018a 12/2019e 12/2020e 12/2021e


Revenue 27,535 34,926 37,755 40,001

EBITDA 3,932 5,601 6,935 7,582

Depreciation & amortisation -1,218 -1,309 -1,770 -2,187


Net interest -316 -484 -486 -468

PBT 2,398 3,808 4,678 4,928

HSBC Qianhai PBT 2,398 3,808 4,678 4,928

Taxation -367 -569 -699 -737

Net profit 2,019 3,220 3,955 4,166

HSBC Qianhai net profit 2,019 3,220 3,955 4,166



Capex -6,442 -13,000 -3,200 -3,200


Dividends -936 -621 -991 -1,217

Change in net debt 412 8,235 -1,824 -2,268

FCF equity -3,193 -8,451 2,549 3,178


Intangible fixed assets 2,029 2,013 1,997 1,980


Current assets 8,745 9,395 11,542 14,092

Cash & others 3,412 3,177 5,001 7,269

Total assets 38,484 50,841 54,435 58,015


Gross debt 6,503 14,503 14,503 14,503

Net debt 3,090 11,325 9,502 7,234




Year to 12/2018a 12/2019e 12/2020e 12/2021e

Y-o-y % change

Revenue 5.5 26.8 8.1 5.9

EBITDA 10.3 42.4 23.8 9.3

Operating profit 4.6 58.1 20.3 4.5

PBT -1.7 58.8 22.9 5.3

HSBC Qianhai EPS 0.5 59.5 22.8 5.3

Ratios (%)

Revenue/IC (x) 1.9 1.6 1.4 1.4

ROIC 15.7 17.3 16.2 16.4

ROE 14.4 20.1 21.0 19.1

ROA 6.3 7.3 7.6 7.5

EBITDA margin 14.3 16.0 18.4 19.0



Net debt/equity 20.3 63.5 45.6 30.4

Net debt/EBITDA (x) 0.8 2.0 1.4 1.0

CF from operations/net debt 100.3 51.8 68.4 98.9


EPS Rep (diluted) 0.52 0.83 1.02 1.07


DPS 0.16 0.26 0.31 0.33

Book value 3.80 4.47 5.23 5.99

Valuation data

Year to 12/2018a 12/2019e 12/2020e 12/2021e

EV/sales 2.1 1.9 1.7 1.6

EV/EBITDA 14.8 11.9 9.3 8.2

EV/IC 3.7 2.5 2.3 2.2

PE* 27.3 17.1 14.0 13.3

PB 3.7 3.2 2.7 2.4

FCF yield (%) -5.8 -15.3 4.6 5.8

Dividend yield (%) 1.1 1.8 2.2 2.3


ESG metrics





Social Indicators Board members independence (%) 27.3






Issuer information


Target price (CNY) 19.40 Sector Agricultural Products




Price relative


2.50

4.50

6.50

8.50

10.50

12.50

14.50

16.50

18.50

2.50

4.50

6.50

8.50

10.50

12.50

14.50

16.50

18.50

2015 2016 2017 2018 2019

Tongwei Rel to CSI 300 Index

Financials & valuation: Tongwei Buy

67

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 119. Tongwei: Company structure, April 2019


Public


（002129.SZ)

EVA film

Public

Tongwei (600438 CH)

Feed

Polysilicon and chemicals

48.53%

PV cell and modules Other

Power generation

Tongwei Group

Liu Hanyuan Guan Yamei

80% 20%

Food processing, livestock &

poultry breeding

51.47%


68

Investment summary

Strong growth outlook supported by capacity expansion and low funding costs

The company is the second largest PV mono wafer supplier globally with a 29% market share in

2018. We expect its market share to rise to 35% in 2019 on the production ramp up of the Inner

Mongolia Ph4 project. The company is also a leading semi wafer supplier in China with an 18%

market share in float-zone mono wafer (a wafer used in IGBT) globally. Looking ahead, the

company has ambitious plans to raise its PV mono wafer capacity by 25GW to 55GW (Inner

Mongolia Ph5 project) and semi mono wafer capacity by 1.35m slices/mth to 1.67m slices/mth

by 2022e (Wuxi semi wafer project). Upon completion of these projects, we expect the

company’s revenue to triple from c.RMB14bn in 2018. Compared with its private sector peers

like Longi, Zhonghuan’s borrowing cost is low (averaging 3.7% in the past three years). This

gives the company a competitive advantage in fundraising to support its expansion plans. The

company plans to raise RMB5bn from a private placement (document submitted for CSRC’s

review in January 2019), which could reduce its net gearing from 55% in 2018 to 27% in 2019e.

Exhibit 120. Zhonghuan: Revenue breakdown, 2018

Exhibit 121. Zhonghuan: Earnings to double y-o-y in 2020e on new semi wafer capacities in Wuxi

Source: Wind, company data, HSBC Qianhai Securities Source: Wind, company data, HSBC Qianhai Securities estimates. E = HSBC Qianhai Securities estimates

Renewable energy product

88%

Semiconductor material

7%

Semiconductor device

1%

Power generation

3%

Others1%

-20%

0%

20%

40%

60%

80%

100%

120%

140%

0

500

1,000

1,500

2,000

2,500

2016 2017 2018 2019E 2020E 2021E

Net profit YoY

Rmb mn

Zhonghuan Semiconductor

(002129 CH)

Second largest player in PV mono wafer globally, after Longi

Largest semi float-zone mono-wafer supplier in China, with an 18%

global market share


The company plans to raise

its PV mono wafer capacity

by 25GW to 55GW

69



No. 2 in a near duopoly market

With a 29% market share in 2018, Zhonghuan is the second-largest supplier of PV mono wafer

globally, after Longi. The number three is the much smaller JinkoSolar, which had an 11%

market share in 2018. Unlike Longi and Zhonghuan, JinkoSolar’s mono wafer are mainly used

internally to supply its own solar cell production. This has led to a near duopoly market

dominated by Longi and Zhonghuan. We expect the two largest players to account for about

77% of global PV mono wafer supply in 2019e, up from 65% in 2018. We believe this will give

the two companies significant pricing power over the downstream cell producers.

Exhibit 122. PV mono wafer market share, 2018

Exhibit 123. PV mono wafer market share, 2019e

Source: Company data, HSBC Qianhai Securities Source: Company, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Exhibit 124. PV mono wafer: Capacity schedules of the top 3 producers


Semi wafer revenue to rise 8x in 2018-21e on capacity boost

In 2018, the company produced 1.7m slices of semi wafer, nearly all of which was 8’. As of end-

2018, the company has a semi wafer capacity of 320,000 slices per month, composing of 8’

wafer capacity of 300,000 slices per month and 12’ wafer capacity of 20,000 slices per month.

In November 2017, the company announced that it will partner with the Wuxi government and

Jingsheng to invest USD3bn in a semi wafer project in Wuxi. Zhonghuan will take a 60% stake

in the project, the Wuxi government 30%, and Jingsheng 10%. Upon completion, the project

Longi35%

JinkoSolar11%

Zhonghuan29%

Others25%

Longi42%

JinkoSolar7%

Zhonghuan35%

Others16%

-

5

10

15

20

25

30

35

40

2017 2018 2019E

Longi Zhonghuan JinkoSolar

GWMono-Si Wafer Capacity


70

can supply 750,000 8’ wafer and 600,000 12’ wafer on a monthly basis. We expect the 8’ wafer

production line to be completed by 2020 and the 12’ wafer line to be completed by phases in

2020-22e. Given the capacity boost, we expect Zhonghuan’s semi wafer revenue to rise by 8x

in 2018-21e.

Exhibit 125. Zhonghuan: Wuxi semi wafer project to add monthly capacities of 1.35m slices

Exhibit 126. Zhonghuan: Semi wafer’s margin is well above that of PV wafer



Lower borrowing cost comparing with Longi

Zhonghuan is an SOE with 37% of its shares indirectly held by Tianjin SASAC. We believe the

SOE status allows it access to cheaper borrowing compared with its private sector peers like

Longi. In 2016-18, Zhonghuan’s average borrowing cost was 3.7%, well below Longi’s 5.8%.

The ability to access cheap funding is pivotal for players in the solar industry, which is still in its

expansionary phase.

Exhibit 127. Borrowing cost comparison: Zhonghuan lower than Longi

Exhibit 128. Zhonghuan: We expect PV wafer capacity to rise from 25GW in 2018 to 55GW in 2021e

Source: Company data, HSBC Qianhai Securities Source: Company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

300 300

1050 1050 1050

20 20

170

320

620

0

200

400

600

800

1000

1200

2018 2019E 2020E 2021E 2022E

8' 12'

k slices / mth

0%

5%

10%

15%

20%

25%

30%

35%

2015 2016 2017 2018 2019E 2020E 2021E

PV product Semi wafer

GM

6.5%

5.3% 5.6%

2.1%

3.7% 5.2%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

2016 2017 2018

Longi Zhonghuan

Borrowing cost

10

25 30 30

55

-

10

20

30

40

50

60

2017 2018 2019E 2020E 2021E

Tho

usan

ds

PV Mono-Si Wafer

GW

71


Net gearing to improve in 2019 on the RMB5bn private placement

In January 2019, Zhonghua announced a plan to raise RMB5bn by placing 557m shares, or

20% of the total shares outstanding, to less than 10 strategic investors at RMB8.98/share. Of

the RMB5bn proceeds, RMB4.5bn will be used to fund the 8-12’ semi wafer project in Wuxi and

the remainder to replenish working capital. Upon completion of the private placement, we

expect the company’s net gearing to drop from 55% to 27%.

Investment concerns

Supply chain not as consolidated as Longi’s

Compared with Longi, which has formed a nearly closed loop supply chain, Zhonghuan’s supply

chain is an open-loop. While Longi is able to source the crystal growing equipment used in

mono pulling internally, Zhonghuan relies on third party vendors like Jingsheng M&E. This

leaves Zhonghuan vulnerable to external supply shocks. To mitigate the risk, the company

chooses to invest in or form partnership with its suppliers. For example, in July 2017 the

company took a 30% stake in GCL Poly’s polysilicon project in Xinjiang province to secure its

polysilicon supply. In November 2017, it partnered with Jingsheng M&E to invest in the Wuxi

semiconductor wafer project.

Solar wafer margin lower than Longi’s

In 2015-18, the gross margin of Zhonghuan’s solar wafer business averaged 15%. This is below

Longi’s 25% during the same period (Exhibit 130). We believe this is due to its open-loop supply

chain that leads to higher COGS compared with Longi’s.

Exhibit 129. Zhonghuan: Corporate milestones

Year Corporate Activities Implication

2009 Set up Inner-Mongolia PV material company and invested in mono-Si wafer project Ph1 Entered into PV market 2011 Mono-Si wafer project started production 2012 Invested in mono-Si wafer project Ph2 2012 Supplied CFZ products to overseas customers 2012 Started R&D on diamond wire sawing (DWS) technology Lowered mono-is wafer manufacturing cost 2013 Started mono-Si wafer project Ph3 2015 Set up DZS solar and Tianjin Huanmei Energy Expanded into cell and module business 2016 Started mono-Si wafer project Ph4 2016 Deployment in high-efficiency PERC cell Start developing high-efficiency cell 2017 Invested in high-efficiency shingled-cell module project in Wuxi 2017 Acquired Guodian Solar Improved cell technology 2017 Invested in GCL-Poly Xinjiang Secured upstream polysilicon supply 2017 Partnered with Wuxi government and Jingsheng to invest in semiconductor wafer project in Wuxi Scaling up semi wafer capacity



72

Exhibit 130. Solar wafer gross margin comparison: Zhonghuan lower than Longi


Heavy capex burden in 2019-21 on new capacity build-outs

We expect Zhonghuan’s capex to remain high at around RMB5bn per annum in 2019-21 due to

the construction of its new semi wafer capacities in Wuxi and its new PV wafer capacities in

Inner Mongolia. The Wuxi project has a total capex budget of around RMB20bn, or RMB12bn

for Zhonghuan’s 60% stake. The Inner Mongolia project has a capex budget of around

RMB5bn.

Exhibit 131. Zhonghuan: We expect FCF continue to be negative in 2019-20e

Exhibit 132. Zhonghuan: Net gearing to drop in 2019e post private placement



Financial forecasts

Earnings forecasts


18e. We base our forecasts on the following key assumptions:

Revenue: We forecast a revenue CAGR of 30% in 2018-21e, driven by strong growth of

Zhonghuan’s semi material segment. We expect the semi material segment to register a

113% revenue CAGR in 2018-21e, with the phase-by-phase completion of the company’s

new semi wafer capacities in Wuxi.

14.4%12.5%

18.7%

15.0%

21.5%

28.2%

32.7%

16.3%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

2015 2016 2017 2018

Zhonghuan Longi

GM

-5

-4

-3

-2

-1

-

1

2

2015 2016 2017 2018 2019E 2020E 2021E

Rm

b bn

FCF

0%

10%

20%

30%

40%

50%

60%

2015 2016 2017 2018 2019E 2020E 2021E

Net debt / Equity

73


Gross margin: We forecast gross margin to improve from 17% in 2018 to 23% in 2021e,

driven by a rising contribution from the semi material segment, which has a better margin.

Exhibit 133. Zhonghuan: Segments and full P&L forecasts

RMBm 2018 2019e 2020e 2021e 2018-21e CAGR

Turnover 13,756 18,745 22,105 30,281 30.1% Renewable energy products 12,092 16,160 16,205 19,744 17.8% Semiconductor materials 1,013 1,920 5,202 9,806 113.1% Semiconductor devices 153 153 153 153 0.0% Power generation 335 349 382 416 7.4% Services 48 48 48 48 0.0% Others 114 114 114 114 0.0% Gross Profit 2,387 2,907 4,764 6,938 42.7% Renewable energy products 1,818 1,991 2,744 3,376 22.9% Semiconductor materials 305 634 1,717 3,236 119.8% Semiconductor devices (9) 0 0 0 -100.0% Power generation 214 223 244 265 7.4% Services 34 34 34 34 0.0% Others 26 26 26 26 0.0% Gross Margin 17% 16% 22% 23% Business tax (62) (84) (100) (136) 30.1% Selling expenses (171) (233) (275) (377) 30.1% Admin. expenses (996) (1,312) (1,592) (2,180) 29.8% Asset impairment losses / Fair value changes (190) (85) (127) (201) 2.0% Other gain / (losses) 516 361 180 180 -29.5% Operating profit 1,484 1,553 2,852 4,223 41.7% Net finance charges (618) (628) (538) (559) -3.3% Share of JCE 7 7 7 7 0.0% Profit before taxes 873 931 2,321 3,671 61.4% Tax (84) (139) (347) (550) 87.3% Minorities (157) (157) (590) (1,028) 87.2% Pre-exceptional profit 632 635 1,384 2,094 49.1% Dividend to preferred shareholders and perpetual capital securities (54) (54) (54) (54) 0.0% Exceptionals 0 0 0 0 Net profit 578 581 1,330 2,040 52.2%



We forecast the net debt to equity ratio to fall from 55% in 2018 to 26% in 2021e, supported by

the company’s strong OCF and the RMB5bn private placement in 2019.

Exhibit 134. Zhonghuan: Net debt and cash flow forecasts

RMBm 2018 2019e 2020e 2021e

Net debt/(cash) 8,598 5,689 7,393 6,727 Net debt to equity 55% 27% 32% 26% Cash from Operations 1,708 3,611 3,883 6,348 Cash from Investing (5,759) (4,937) (4,937) (4,937) FCF (4,051) (1,325) (1,054) 1,411



74

Valuation and risks



a 2020e PE of 36x, close to the historical average valuation of 40x and is reasonable in our

view. Key assumptions in our DCF valuation model include:





capital ratio.


changes in working capital) to expand at a CAGR of 11% in 2018-29e, reflecting solid

growth in demand.

Capital expenditure: We assume steady capex of around RMB5bn per annum in 2019-

29e, reflecting high investment in new projects and capacity expansion.


period after 2029.

75


Exhibit 135. Zhonghuan: Discounted cash flow valuation

RMBm 2015 2016 2017 2018 2019e 2020e 2021e 2022e

Profit after tax 212 404 591 789 792 1,974 3,121 3,278 YoY growth 90% 46% 34% 0% 149% 58% 5% Add: Depreciation & amortization 452 672 970 1,482 1,203 1,419 1,660 1,920 Net finance expense 230 109 463 681 628 538 559 632 Operating cash flow before W/C changes

894 1,185 2,024 2,952 2,623 3,930 5,340 5,829

Changes in working capital (72) (373) (893) (1,353) 995 (41) 1,014 Net operating cash flow 822 813 1,131 1,600 3,618 3,889 6,355 5,829 Capex (2,265) (1,746) (4,788) (5,759) (4,937) (4,937) (4,937) (4,986) Free Cash Flow (1,443) (934) (3,657) (4,159) (1,319) (1,047) 1,418 843 Discount Factor 1.00 0.93 0.87 Gross PPE 7,950 10,050 15,147 24,670 28,599 33,334 38,230 43,216 Depreciation Rate 6% 7% 6% 6% 4% 4% 4% 4% PV of FCF (1,047) 1,321 732 RMBm 2023e 2024e 2025e 2026e 2027e 2028e 2029e Terminal

Value Profit after tax 3,441 3,614 3,794 3,984 4,103 4,227 4,353 YoY growth 5% 5% 5% 5% 3% 3% 3% Add: Depreciation & amortization 2,192 2,476 2,773 3,083 3,405 3,741 4,090 Net finance expense 705 780 855 930 1,007 1,084 1,163 Operating cash flow before W/C changes

6,338 6,869 7,422 7,997 8,516 9,052 9,606

Changes in working capital 0 0 0 0 0 0 0 Net operating cash flow 6,338 6,869 7,422 7,997 8,516 9,052 9,606 Capex (5,036) (5,086) (5,137) (5,188) (5,240) (5,293) (5,346) Free Cash Flow 1,302 1,783 2,285 2,808 3,275 3,759 4,260 81,922 Discount Factor 0.81 0.75 0.70 0.66 0.61 0.57 0.53 0.53 Gross PPE 48,252 53,338 58,475 63,664 68,904 74,197 79,543 Depreciation Rate 5% 5% 5% 5% 5% 5% 5% PV of FCF 1,054 1,345 1,606 1,840 2,000 2,139 2,259 43,436 Summary of PV (Enterprise Value) 56,683 Less: Net debt (incl. perpetual) (6,455) Equity value 50,228 Less: Minority interest (2,557) Shareholder Equity Value 47,672 Total shares issued by yr-end 2019 (mn) 3,342 Per Share Value - Rmb 14.3 Assumptions Risk-free rate 2.5% ERPch 6.5% Beta 1.03 Cost of equity = RFR + BETA*ERPch 9.2% Cost of debt 5.0% Income tax 15% After tax cost of debt 4.3% Debt/Capital 38% WACC 7.3% Terminal Growth 2%



76

Exhibit 136. Zhonghuan forward PE: Trading 1SD below the historical average

Exhibit 137. Zhonghuan forward PB: Trading below the historical average


Exhibit 138. Zhonghuan: Earnings sensitivity to gross margin and revenue changes, 2020e


20% -11% 1% 13% 25% 37% 49% 61% 15% -15% -4% 7% 19% 30% 42% 53% 10% -20% -9% 2% 13% 23% 34% 45% 5% -25% -15% -4% 6% 17% 27% 37% 0% -30% -20% -10% 0% 10% 20% 30% -5% -34% -25% -16% -6% 3% 13% 22% -10% -39% -30% -21% -13% -4% 5% 14% -15% -44% -36% -27% -19% -10% -2% 6% -20% -49% -41% -33% -25% -17% -9% -1%


Downside risks:

Weaker-than-expected PV demand: PV wafer accounted for around 65% of Zhonghuan’s

revenue in 2018. Should global PV installations slow, we see downside risks to our

earnings forecasts.

China-US trade tension could hinder equipment procurement from US suppliers:

Localization of semi wafer production equipment is low at just 1-2% and many of the key

equipment comes from US. Hence, heightening China-US trade tensions could harm

Zhonghuan’s semi wafer capacity plans.


Zhonghuan’s financing exercise since its listing in 2007. As the capex burden is likely to

remain high at around RMB5bn per annum in 2019-21e, we see a risk of equity dilution

should the company fail to fulfill its capex needs via debt financing.

Weaker-than-expected semi demand: We expect the semi wafer business to account for

46% of Zhonghuan’s gross profit in 2021e. Should downstream demand be weaker-than-

expected or capacity ramp-ups be slower-than-expected, we see downside risks to our

earnings forecasts.

Greater-than-expected cuts to wafer ASP: We expect mono wafer ASP to decline by

20%/10%/10% in 2019/20/21e. We see risks of larger-than-expected margin erosion if the

ASP cuts are greater than expected.

-

10.0

20.0

30.0

40.0

50.0

60.0

Feb-16 Feb-17 Feb-18 Feb-19

PE Mean +1SD -1SD

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0


PB Mean +1SD -1SD

77



Year to 12/2018a 12/2019e 12/2020e 12/2021e


Revenue 13,756 18,745 22,105 30,281

EBITDA 2,972 2,762 4,277 5,890

Depreciation & amortisation -1,482 -1,203 -1,419 -1,660


Net interest -618 -628 -538 -559

PBT 873 931 2,321 3,671

HSBC Qianhai PBT 873 931 2,321 3,671

Taxation -84 -139 -347 -550

Net profit 632 635 1,384 2,094

HSBC Qianhai net profit 578 581 1,330 2,040



Capex -5,759 -4,937 -4,937 -4,937


Dividends -773 -137 -112 -187

Change in net debt 672 -2,909 1,704 -666

FCF equity -3,488 -2,942 -1,544 -155


Intangible fixed assets 1,754 1,700 1,647 1,594


Current assets 13,893 17,854 17,215 20,959

Cash & others 6,740 8,648 6,945 7,611

Total assets 42,697 50,399 53,284 60,311


Gross debt 11,299 10,299 10,299 10,299

Net debt 4,559 1,650 3,354 2,688




Year to 12/2018a 12/2019e 12/2020e 12/2021e

Y-o-y % change

Revenue 42.6 36.3 17.9 37.0

EBITDA 42.0 -7.1 54.9 37.7


PBT 27.7 6.7 149.2 58.2

HSBC Qianhai EPS -3.2 -11.3 109.7 53.4

Ratios (%)

Revenue/IC (x) 0.9 1.0 1.1 1.3

ROIC 9.7 7.7 11.8 15.2

ROE 4.6 3.6 6.8 9.7

ROA 2.1 1.7 3.8 5.5

EBITDA margin 21.6 14.7 19.4 19.5



Net debt/equity 29.0 7.7 14.4 10.3

Net debt/EBITDA (x) 1.5 0.6 0.8 0.5

CF from operations/net debt 37.5 218.8 115.8 236.1


EPS Rep (diluted) 0.23 0.21 0.41 0.63


DPS 0.03 0.02 0.04 0.06

Book value 4.78 5.63 6.01 6.58

Valuation data

Year to 12/2018a 12/2019e 12/2020e 12/2021e

EV/sales 2.2 1.5 1.3 1.0

EV/EBITDA 10.3 10.1 6.9 4.9

EV/IC 1.9 1.4 1.3 1.1

PE* 44.8 50.6 24.1 15.7

PB 2.0 1.7 1.6 1.5

FCF yield (%) -13.3 -11.2 -5.9 -0.6

Dividend yield (%) 0.3 0.2 0.4 0.6


ESG metrics




CO2 reduction policy [Yes] Female board members (%) 45.5







Issuer information


Target price (CNY) 14.30 Sector Electronic Equipment

RIC (Equity) 002129.SZ Country China



Price relative


1.90

6.90

11.90

16.90

21.90

26.90

31.90

36.90

1.90

6.90

11.90

16.90

21.90

26.90

31.90

36.90

2015 2016 2017 2018 2019

Zhonghuan Semiconductor Rel to CSI 300 Index

Financials & valuation: Zhonghuan Semiconductor Buy

78

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 139. TJ Zhonghuan Semiconductor: Company structure, April 2019


Tianjin SASAC

PublicBohai Industrial Investment Fund


（002129.SZ)

EVA film

Other business

China central electronics group

27.55%

100%

Tianjin SASAC

Tianjin SASAC

PublicBohai Industrial Investment Fund


（002129 CH)

Semiconductor material

Power generation

100%

3.02%59.79%

GuodianTechnology & Environment

Group [1296 HK]

Renewable energy material

Other

Semiconductor device

China Central Electronics Group

27.55%

100%

9.64%

Tianjin SASAC

79


Investment summary

Benefiting from the wafer capacity upcycle

We view Jingsheng as the best name to play the wafer capacity upcycle among our coverage,

given the company’s market leadership in wafer-making equipment in China. The company has

a c50% market share in third-party-made crystal-growing equipment in China. Crystal-growing

equipment is key to the production of PV and semi wafer. We expect Jingsheng’s crystal-

growing equipment to register a 28% revenue CAGR in 2018-21e, driven by capacity expansion

in China’s PV and semiconductor wafer industry, leading to a 25% earnings CAGR in 2018-21e.

Exhibit 140. Jingsheng: Revenue breakdown, 2018

Exhibit 141. Jingsheng: Earnings to double in 2018-21e

Source: Wind, company data, HSBC Qianhai Securities Source: Wind, company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates


Significant order potential from mono-Si wafer capacity expansion

Jingsheng manufactures various type of equipment for the solar supply chain, including

automatic mono-Si growing furnaces, multi-Si casting furnaces, silicon float zone crystal pullers,

and crystal ingot single wire squaring machines (Exhibit 142). Of these, crystal-growing

equipment, including mono-Si growing furnace and multi-Si casting furnace, accounted for 76%

of the company’s revenue in 2018. We expect crystal-growing equipment revenue to rise at a

Crystal growing equipment

76%

Intelligent manufacturing

equipment11%

Sapphire material

5%

Other core business2% Others

6%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

200

400

600

800

1,000

1,200

1,400

2016 2017 2018 2019E 2020E 2021E

Net profit YoY

Rmb mn

Jingsheng M&E (300316 CH)

Top third-party wafer equipment maker in China, with c50% market

share in crystal-growing equipment

Early cycle name benefiting from the wafer capacity upcycle


We forecast a 25% earnings

CAGR in 2018-21e


80

28% CAGR in 2018-21e due to the growing mono wafer market (Exhibit 145). As mentioned

earlier, we expect the world’s mono wafer production to almost double from 52GW in 2018 to

101GW in 2021e. In general, 1GW of mono wafer capacity corresponds with 170 units of mono-

Si growing furnace demand. Hence, we expect the 101GW production in 2021 to translate into

an installation base demand of 17,000 units of mono-Si growing furnaces, up from 9,000 units in

2018. Assuming a 30% market share for Jingsheng, the potential future order size could be

2,400 units (8,000 new units x 30%), or 2x 2018 sales.

Exhibit 143. Mono-Si growing furnace: Installation base to surge on the growing mono wafer market


Exhibit 144. Jingsheng: Mono-Si growing furnace sales to register a 27% CAGR in 2018-21e

Exhibit 145. Jingsheng: Crystal-growing equipment revenue up on a 28% CAGR in 2018-21e



19 33

52

69 88 101

3,280

5,685

8,863

11,735

14,985

17,203

- 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000

-

20

40

60

80

100

120

2016 2017 2018 2019E 2020E 2021E

Tho

usan

ds

Mono-Si Wafer Production Mono-Si growing furnace demand (RHS)

GWUnits

-100%

-50%

0%

50%

100%

150%

200%

0

500

1,000

1,500

2,000

2,500

3,000

2013 2015 2017 2019E 2021E

Mono-Si growing furnace sales YoY

Unit

-100%

-50%

0%

50%

100%

150%

-

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

2013 2015 2017 2019E 2021E

Crystal growing equipment revenue YoY

RMB mn

Exhibit 142. Key equipment that Jingsheng produces

Crystal growing equipment PV intelligent equipment LED intelligent equipment Sapphire material

Automatic Mono-Si Growing Furnace

Mono-Si Ingot Squaring & Grinding Integrated Machine

Automatic Sorting & Testing Equipment for LED Device

Automatic Sapphire Crystal Growing Furnace

Silicon Float Zone Crystal Puller Multi-crystal Rounding & Grinding Integrated Machine

Automatic LED Light Production Line Sapphire Ingot

Multi-Si Casting Furnace Crystal Ingot Single Wire Squaring Machine Sapphire Wafer


81


Capitalising on strong R&D capability

Jingsheng’s R&D expense as percentage of revenue was 7% in 2018, topping that of industry

peers (Exhibit 146). We believe such a high R&D expense ratio supports the company’s leading

market position in high-end PV equipment.

Exhibit 146. Jingsheng has the highest R&D expense ratio among peers


A robust market outlook for semiconductor equipment

The company also supplies crystal-growing furnaces for semiconductor wafer makers. In 2018,

the company signed new semi equipment contracts worth RMB500m, around 16% of the total

value of the new contracts signed during the year. As of end 1Q19, the company has semi

equipment contracts worth RMB558m on its orderbook. We see a robust market outlook for its

semi equipment. Base on the announced capacity expansion plans by major semi wafer

makers, we see an additional monthly semi capacity of 2.5m slices for 8’ and 3.1m slices for 12’

in 2019-22. In general, 1m slices/mth of capacity translates into a semi crystal-growing furnace

demand of 150 units. We therefore see an incremental demand of 840 units of semi crystal-

growing furnace (150 x 5.6) or RMB12bn in contract value. To tap the market upside, Jingsheng

has taken a 10% stake in Zhonghuan’s semi wafer project in Wuxi.

Exhibit 147. Jingsheng: New contract composition, 2017-18

Exhibit 148. Jingsheng: Backlog composition, 2017-18


7.2%

3.7%3.0%

2.2%

0.9%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

Jingsheng FAM Zhonghuan Tongwei Longi

2018

R&D expense as% of revenue

-

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

2017 2018

PV Equipment Semi Equipment

RMB mn New contract

-

500

1,000

1,500

2,000

2,500

3,000

2017 2018

PV equipment Semi equipment

RMB mn Order backlog


82

Exhibit 149. Semi crystal-growing furnace: China’s market size calculation

Exhibit 150. Semi crystal-growing furnace: China’s demand forecasts, 2019-22e

8-inch 12-inch

China's new semi capacity (m slice/mth)

2.5 3.1

Crystal-growing furnace demand (unit)

375 465

ASP (RMBm /unit) 6 20

Crystal-growing furnace demand (RMBm)

2,250 9,300

Source: Company data, HSBC Qianhai Securities Source: Company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Investment concerns

Longi can make its own mono-Si growing furnace

Despite being the largest third-party producer of mono-Si growing furnaces in China, Jingsheng

is unable to penetrate into Longi’s supply chain. This is because Longi, partnered with Naura,

develops mono-Si growing furnaces for its own use. We therefore see a ceiling for Jingsheng’s

market share, given Longi’s dominant position in mono-Si wafer production (Exhibit 151). In our

model, we forecast a 30% market share for Jingsheng’s mono-Si growing furnace, which is

c50% of the addressable market excluding Longi.

Exhibit 151. Mono-Si wafer market share, 2019e

Exhibit 152. Crystal-growing equipment: Jingsheng has a c50% share of the market, excluding Longi



Earnings are highly sensitive to Zhonghuan’s capex plan

Zhonghuan is Jingsheng’s largest customer, accounting for 80% of its 2018 new orders. We

therefore see high earnings sensitivity to Zhonghuan’s investment plan. Also, relying on a single

large customer could constrain the company’s power in terms of negotiation of terms of

-

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

2019E 2020E 2021E 2022E

8-inch 12-inch

RMB mn Crystal-growing funance demand forecasts

Longi42%

JinkoSolar7%

Zhonghuan35%

Others16%

Jingsheng48%NAURA

Technology35%

Jingyuntong Technology

9%

Huasheng Tianlong Photoelectric

8%

3rd party vendor market share, 2012-18

83


payment, we believe. As shown in Exhibit 154, Jingsheng’s trade and notes receivable days

have averaged 350 days in the past three years.

Exhibit 153. Zhonghuan: capex schedule, 2016-21e

Exhibit 154. Jingsheng: Trade and notes receivable days, 2015-18



We expect net gearing to rise in 2019-20e

In 2018, the company took a 10% stake in Zhonghuan’s semi wafer project in Wuxi. The planned

investment of the project is around USD3bn (RMB20bn), or RMB2bn attributable to Jingsheng.

This is a relative heavy burden for Jingsheng in view of the company’s asset size of RMB6.3bn as

of end-2018. We hence expect its net gearing to rise from -10% in 2018 to 4% in 2020.

Exhibit 155. Jingsheng: Net gearing to peak in 2020, subsiding thereafter

Exhibit 156. Jingsheng: We expect FCF to remain negative in 2019e



Financial forecasts

Earnings forecasts


18. We base our forecasts on the following key assumptions:

0

1,000

2,000

3,000

4,000

5,000

6,000

2016 2017 2018 2019E 2020E 2021E

CAPEX

RMB mn

0

50

100

150

200

250

300

350

400

450

500

2015 2016 2017 2018

Trade and notes receivable days

Days

-30%

-25%

-20%

-15%

-10%

-5%

0%

5%

10%

2015 2016 2017 2018 2019E 2020E 2021E

Net debt / Equity

(0.6)

(0.5)

(0.4)

(0.3)

(0.2)

(0.1)

0.0

0.1

0.2

0.3

0.4

2015 2016 2017 2018 2019E 2020E 2021E

Rm

b bn

FCF


84

Revenue: We forecast a revenue CAGR of 22% in 2018-21e, driven by a 28% revenue

CAGR for the crystal growing equipment products.

Gross margin: We forecast gross margin to improve slightly from 40% in 2018 to 44% in

2021e.

Exhibit 157. Jingsheng: Segments and full P&L forecasts

RMBm 2018 2019e 2020e 2021e 2018-21e CAGR

Turnover 2,536 3,271 3,953 4,637 22% Crystal growing equipment 1,940 2,734 3,394 4,054 28% Intelligent manufacturing equipment 277 218 240 264 -2% Sapphire material 125 125 125 125 0% Other core business 49 49 49 49 0% Others 145 145 145 145 0% Gross Profit 1,002 1,397 1,713 2,029 27% Crystal growing equipment 846 1,254 1,561 1,867 30% Intelligent manufacturing equipment 105 92 101 112 2% Sapphire material 16 16 16 16 0% Other core business 3 3 3 3 0% Others 33 33 33 33 0% Gross Margin 40% 43% 43% 44% Business tax (26) (33) (40) (47) 22% Selling expenses (46) (59) (71) (84) 22% Admin. expenses (297) (383) (463) (543) 22% Asset impairment losses / Fair value changes (100) (70) (80) (89) -4% Other gain / (losses) 108 86 86 86 -7% Operating profit 641 939 1,145 1,353 28% Net finance charges 2 -3 -9 -11 -270% Share of JCE 5 5 5 5 0% Profit before taxes 649 941 1,141 1,347 28% Tax (80) (140) (170) (201) 36% Minorities 14 0 0 0 -100% Pre-exceptional profit 582 801 971 1,146 25% Dividend to preferred shareholders and perpetual capital securities

0 0 0 0

Exceptionals 0 0 0 0 Net profit 582 801 971 1,146 25%



We expect net gearing to edge up slightly from -10% in 2018e to 3% in 2021e on investment in

the semi wafer project in Wuxi.

Exhibit 158. Jingsheng: Net debt and cash flow forecasts

RMBm 2018 2019e 2020e 2021e

Net debt/(cash) (434) 124 247 191 Net debt to equity -10% 3% 4% 3% Cash from Operations 166 268 758 976 Cash from Investing (449) (695) (695) (695) FCF (283) (427) 63 281


85


Valuation and risks



a 2020e PE of 19x, 1SD below the historical average valuation. We believe this is reasonable

as Jingsheng’s earnings are tied to wafer capacity expansion, which is highly cyclical. Key

assumptions in our DCF valuation model include:



Cost of debt (COD): We assume the pre-tax cost of debt to be 5% and the after-tax cost of


capital ratio.


changes in working capital) to expand at a CAGR of 13% in 2018-29e, reflecting stable

growth in demand.

Capital expenditure: We expect capex of around RMB700mn per annum in 2019-21e,

reflecting the company’s share of investment in the Wuxi semi wafer project. Thereafter, we

expect a steady capex of around RMB200-300mn per annum in 2022-29e, reflecting

maintenance capex.

Terminal growth rate at 2%, and we assume the company reaches a steady growth period

after 2029e.


86

Exhibit 159. Jingsheng: Discounted cash flow valuation

RMBm 2015 2016 2017 2018 2019e 2020e 2021e 2022e

Profit after tax 113 184 372 568 801 971 1,146 1,283 YoY growth 63% 102% 53% 41% 21% 18% 12% Add: Depreciation & amortization 28 44 56 82 74 97 131 145 Net finance expense 4 6 6 9 3 9 11 12 Operating cash flow before W/C changes 146 234 433 659 877 1,077 1,287 1,439 Changes in working capital (249) (328) (618) (546) (367) (314) (306) Net operating cash flow (103) (94) (185) 113 510 763 981 1,439 Capex (385) (484) 48 (449) (695) (695) (695) (250) Free Cash Flow (488) (578) (137) (335) (185) 68 286 1,189 Discount Factor 1.00 0.90 0.82 Gross PPE 655 903 969 1,436 1,710 2,363 3,054 3,304 Depreciation Rate 4% 5% 6% 6% 4% 4% 4% 4% PV of FCF 68 259 971 RMBm 2023e 2024e 2025e 2026e 2027e 2028e 2029e Terminal

Value Profit after tax 1,412 1,553 1,708 1,845 1,992 2,092 2,155 YoY growth 10% 10% 10% 8% 8% 5% 3% Add: Depreciation & amortization 159 174 190 206 222 239 256 Net finance expense 12 13 14 15 16 17 18 Operating cash flow before W/C changes 1,583 1,740 1,912 2,065 2,230 2,348 2,429 Changes in working capital 0 0 0 0 0 0 0 Net operating cash flow 1,583 1,740 1,912 2,065 2,230 2,348 2,429 Capex (250) (250) (250) (250) (250) (250) (250) Free Cash Flow 1,333 1,490 1,662 1,815 1,980 2,098 2,179 25,674 Discount Factor 0.74 0.67 0.60 0.54 0.49 0.44 0.40 0.40 Gross PPE 3,554 3,804 4,054 4,304 4,554 4,804 5,054 Depreciation Rate 4% 5% 5% 5% 5% 5% 5% PV of FCF 984 994 1,002 989 975 933 876 10,320 Summary of PV (Enterprise Value) 18,370 Less: Net debt (incl. perpetual) (124) Equity value 18,247 Less: Minority interest (173) Shareholder Equity Value 18,074 Total shares issued by year-end 2019 (mn) 1,285 Per Share Value - Rmb 14.1 Assumptions Risk free rate 2.5% ERPch 6.5% Beta 1.28 Cost of equity = RFR + BETA*ERPch 10.8% Cost of debt 5.0% Income tax 15% After tax cost of debt 4.3% Debt/Capital 2% WACC 10.7% Terminal Growth 2%


87


Exhibit 160. Jingsheng forward PE: Trading at low-end of the historical valuation

Exhibit 161. Jingsheng forward PB: Trading in line with the historical average


Exhibit 162. Jingsheng: Earnings sensitivity to gross margin and revenue changes, 2020e


20% 7% 12% 16% 20% 24% 28% 32% 15% 3% 7% 11% 15% 19% 23% 27% 10% -1% 2% 6% 10% 14% 18% 21% 5% -6% -2% 1% 5% 9% 12% 16%

0% -10% -7% -3% 0% 3% 7% 10%

-5% -15% -12% -8% -5% -2% 2% 5% -10% -19% -16% -13% -10% -7% -4% -1% -15% -24% -21% -18% -15% -12% -9% -6% -20% -28% -25% -23% -20% -17% -14% -12%


Downside risks:

Weaker-than-expected PV and semi demand: Jingsheng manufactures crystal-growing

equipment for PV and semi wafer makers. This contributed 76% of revenue in 2018, so a

slowdown in global PV and semi demand could affect earnings negatively.

Weaker-than-expected capex of Zhonghuan: Zhonghuan is Jingsheng’s largest

customer, accounting for 80% of its 2018 new orders. A weaker-than-expected capex plan

by Zhonghuan could affect Jingsheng’s earnings adversely.

Higher-than-expected receivable provision: Given Jingsheng focuses on Zhonghuan as

its major customer, should Zhonghuan delay its payments, Jingsheng could see an uptick

in its receivables provision rate which could in turn affect its earnings adversely.

Weaker-than-expected margin on intensified competition: Rising competition could

undercut prices and adversely affect the company’s gross margins.

-

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

Apr-16 Apr-17 Apr-18 Apr-19

PE Mean +1SD -1SD

0.0

2.0

4.0

6.0

8.0

10.0

May-12 May-14 May-16 May-18

PB Mean +1SD -1SD


88


Year to 12/2018a 12/2019e 12/2020e 12/2021e


Revenue 2,536 3,271 3,953 4,637

EBITDA 728 1,018 1,247 1,488

Depreciation & amortisation -82 -74 -97 -131

Operating profit/EBIT 647 944 1,150 1,358

Net interest 2 -3 -9 -11

PBT 649 941 1,141 1,347

HSBC Qianhai PBT 649 941 1,141 1,347

Taxation -80 -140 -170 -201

Net profit 582 801 971 1,146

HSBC Qianhai net profit 582 801 971 1,146


Cash flow from operations 166 268 758 976

Capex -449 -695 -695 -695

Cash flow from investment -449 -695 -695 -695

Dividends -108 -128 -177 -214

Change in net debt 265 558 123 -56

FCF equity 202 180 373 581




Current assets 4,427 4,623 5,231 6,012

Cash & others 557 -1 -124 -67

Total assets 6,335 7,394 8,606 9,956


Gross debt 162 162 162 162

Net debt -395 162 285 229




Year to 12/2018a 12/2019e 12/2020e 12/2021e

Y-o-y % change

Revenue 30.1 29.0 20.8 17.3

EBITDA 51.2 39.7 22.6 19.3


PBT 51.2 45.1 21.2 18.0

HSBC Qianhai EPS 17.9 35.5 21.2 18.0

Ratios (%)

Revenue/IC (x) 0.8 0.8 0.8 0.8

ROIC 18.4 20.6 20.2 20.1

ROE 15.3 18.2 18.9 19.1

ROA 9.2 11.7 12.1 12.3

EBITDA margin 28.7 31.1 31.6 32.1


EBITDA/net interest (x) 398.9 135.2 139.1

Net debt/equity -9.3 3.3 5.0 3.5

Net debt/EBITDA (x) -0.5 0.2 0.2 0.2

CF from operations/net debt 165.5 265.6 426.4


EPS Rep (diluted) 0.46 0.62 0.76 0.89


DPS 0.10 0.14 0.17 0.20

Book value 3.16 3.68 4.30 5.03

Valuation data

Year to 12/2018a 12/2019e 12/2020e 12/2021e

EV/sales 5.9 4.8 4.0 3.4

EV/EBITDA 20.7 15.3 12.6 10.5

EV/IC 4.4 3.5 3.0 2.5

PE* 26.2 19.3 15.9 13.5

PB 3.8 3.3 2.8 2.4

FCF yield (%) 1.3 1.2 2.4 3.8

Dividend yield (%) 0.8 1.1 1.4 1.6


ESG metrics





Social Indicators Board members independence (%) 30






Issuer information



RIC (Equity) 300316.SZ Country China



Price relative


3.60

8.60

13.60

18.60

23.60

3.60

8.60

13.60

18.60

23.60

2015 2016 2017 2018 2019

Jingsheng M&E Rel to CSI 300 Index

Financials & valuation: Jingsheng M&E Buy

89

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 163. Jingsheng M&E: Company structure, April 2019


Qiu Minxiu

Public

Zhejiang Jingsheng Mechanical & Electrical

（603806 CH)

Intelligent manufacturing equipment

Sapphire material

45.29%

Crystal growing equipment

Other business

Shaoxing ShangyujingshengInvestment Management Consulting

48.31%

Cao Jianwei

15.31%

26.77%

2.77%

He Jun (son of Qiu Minxiu)

8.22%

0.66%

He Jie (daughter of Qiu

Minxiu)

8.76%

2.97%


90

Investment summary

Niche market with limited growth potential

FAM is the largest supplier of EVA film in the world, with around 60% market share in 2018.

EVA film is a key material used to encapsulate and protect solar cells. While the EVA market is

likely to grow with rising global new PV installations, we flag higher cell conversion efficiency

and continuous cuts in post-cell fabrication costs as downside risks. As a result, we expect only

a 3% earnings CAGR for FAM in 2018-21e (Exhibit 165). The company’s diversification into

non-solar downstream (3C and semiconductor) should reduce business concentration risk, but it

will take time for the new business to make a material earnings contribution.

Exhibit 164. FAM: Revenue breakdown, 2018

Exhibit 165. FAM: We expect stable earnings in 2018-21e

Source: Wind, company data, HSBC Qianhai Securities Source: Wind, company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

PV film 86%

Backsheet for PV module

11%

Power generation1%

Electronic material1%

Others1%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

0

100

200

300

400

500

600

700

800

900

2016 2017 2018 2019E 2020E 2021E

Net profit YoY

RMB mn

First Applied Materials

(603806 CH)

Market leader in EVA film with a 60% market share globally

Limited growth potential for the core EVA film business but could

benefit from diversifying into non-solar downstream

Valuation at 25x 2019e PE is rich; initiate with a Reduce rating and a

TP of RMB25.80

We expect only a 3%

earnings CAGR in 2018-21e

91



Market leader in EVA film globally

With around 60% market share in 2018, FAM is the largest supplier of EVA film globally. EVA

films are plastic sheets used to encapsulate solar cells for protection. Because of this

characteristic, sales of EVA film have a high correlation with global new PV installations. In

general, one kW of PV installation translates into 11 sq.m of EVA film demand. Hence, on our

forecast for new PV installations of 117GW globally in 2019, we estimate an EVA film market

size of 1.3bn sq.m or RMB9bn. We expect the market to expand at an 11% CAGR in 2018-21e

on a similar growth rate of new PV installations during the same period.

Exhibit 166. Global EVA film: Market share breakdown, 2018

Exhibit 167. Global EVA film: Sales to surge on rising new PV installations

Source: Company data, HSBC Qianhai Securities Source: Wind, Company data, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Strong balance sheet leads to low interest expenses

FAM has maintained a net cash position since 2014. This has resulted in low interest expenses

of less than RMB8m a year in 2015-18. Given measured expansion plans, we expect the

company to stay in a net cash position in 2019-21e, leading to an annual interest expense of

just RMB1m in 2019-21e.

Exhibit 168. FAM: We expect positive FCF in 2019-21e

Exhibit 169. FAM: To stay in net cash position in 2019-21e



FMA60%Sveck

15%

Hiuv14%

Others11%

1,089 1,144 1,277

1,391 1,547

1,691

0%

2%

4%

6%

8%

10%

12%

14%

-

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2017 2018 2019E 2020E 2021E 2022E

Global EVA film demand YoY

mn sq.m

(0.3)

(0.2)

(0.1)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

2015 2016 2017 2018 2019E 2020E 2021E

Rm

b bn

FCF

-18%

-16%

-14%

-12%

-10%

-8%

-6%

-4%

-2%

0%

2015 2016 2017 2018 2019E 2020E 2021E

Net debt / Equity


92

Diversifying into non-solar downstream should reduce business concentration risk

To reduce its exposure to the solar industry, the company is actively diversifying downstream

into areas like 3C and semiconductor with new products such as light-sensitive film and

aluminum-laminated film. As of end-2018, the company has the capacity to produce 50m sq.m

of light-sensitive film which is mainly used in the production of printed circuit boards (PCB). It

plans to increase its capacity to 200m sq.m by end-2020. The company also has the capacity to

produce 2-5m sq.m of aluminum-laminated film, which is a key material in flexible packaging of

lithium batteries.

Investment concerns

Pressure for grid parity could pass upstream, driving down EVA prices

FAM’s PV film gross margin hinges on EVA film prices (Exhibit 170). EVA film prices have

dropped 54% since 2011 due to price-cutting along the solar industry value chain. EVA film

accounted for 4% of the cost of a typical 290W module in 2018, up from 3% in 2017, on a sharp

decline in module cost (-13% y-o-y) and stable EVA prices (+4% y-o-y). While EVA film prices

seemed stable in 2018, pressure for grid parity by PV operators could pass upstream, forcing

EVA film producers to lower their prices and margins further.

Exhibit 170. FAM: EVA film price has down 54% since 2011

Exhibit 171. EVA accounted for 4% cost of a typical 290W module



Rising conversion efficiency means less usage of EVA film per watt

As mentioned, the demand of EVA film is directly correlated with the area of a PV module. With

rising cell conversion efficiency, the same amount of power could be generated by a smaller

module unit, implying less demand of EVA film on a per watt basis. As shown in Exhibit 172, per

kW EVA usage has declined from 16 sq.m in 2012 to 11 sq.m in 2018, with the rise in cell

conversion efficiency from 17% to 22% during the same period. According to the China PV

Industry Association, the conversion efficiency of PERC P-type mono cell could rise from 21.8%

in 2018 to 22.6% in 2021 (Exhibit 173).

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

0

2

4

6

8

10

12

14

16

18

2011 2013 2015 2017 2019E

EVA Film ASP PV Film GM (RHS)

RMB/sq.m

-

100

200

300

400

500

600

700

800

2017 2018

EVA film Others

RMB / module unit

93


Exhibit 172. Higher conversion efficiency corresponds with a decline in EVA usage per KW, 2012-18

Exhibit 173. PERC mono cell: Conversion efficiency to trend higher in 2018-21e

Source: Company data, HSBC Qianhai Securities Source: CPIA, HSBC Qianhai Securities. E = HSBC Qianhai Securities estimates

Benefits from economies of scale have not been apparent

FAM is the largest EVA producer globally. However, the benefits from economies of scale have

yet to materialize. Despite being 4x as big as Sveck, the second larger producer of EVA film

globally, FAM’s gross margin in 2018 was similar. In fact, the gross margins of the two have

been tracking each other since 2015 (Exhibit 175).

Exhibit 174. FAM’s revenue is about 4x as big as Sveck…

Exhibit 175. …but gross margins of the two are similar


Earnings to decline in 2019 on lack of one-off gains

2018 earnings rose 28% y-o-y to RMB751m, including RMB258mn gains from factory relocation

compensation. Given a lack of one-off gains in 2019, we expect the company’s net profit to

decline 8% y-o-y.

Financial forecasts

Earnings forecasts

We expect a 3% earnings CAGR in 2018-21e, compared with a 5% earnings CAGR in 2015-18.

We base our forecasts on the following key assumptions:

16.0

12.011.017.0%

21.1% 21.8%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

0.0

5.0

10.0

15.0

20.0

2012 2017 2018

EVA usage per kW Cell Conversion Efficiency

sq.m / KW

15%

16%

17%

18%

19%

20%

21%

22%

23%

2018 2019E 2020E 2021E

PERC mono-Si cell

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

2014 2015 2016 2017 2018

Sveck FAM

RMB mn

0%

5%

10%

15%

20%

25%

30%

35%

40%

2014 2015 2016 2017 2018

Sveck FAM

GM


94

Revenue: We forecast a revenue CAGR of 9% in 2018-21e, led by the electronic material

segment. We forecast a 165% revenue CAGR for the electronic material segment in 2018-

21e, driven by new products and capacity expansion.

Gross margin: We forecast gross margin to remain stable at 20-21% in 2018-21e.

Exhibit 176. FAM: Segments and full P&L forecasts

RMBm 2018 2019e 2020e 2021e 2018-21e CAGR

Turnover 4810 5740 6050 6243 9% PV film 4153 5004 5004 5004 6% Backsheet for PV module 511 542 558 575 4% Power generation 38 38 38 38 0% Electronic material 29 78 371 547 165% Hot-melt web 21 21 21 21 0% Other main business 11 11 11 11 0% Others 47 47 47 47 0% Gross Profit 946 1126 1215 1285 11% PV film 795 958 958 958 6% Backsheet for PV module 110 116 120 123 4% Power generation 9 9 9 9 0% Electronic material 2 13 98 164 347% Hot-melt web 5 5 5 5 0% Other main business 5 5 5 5 0% Others 20 20 20 20 0% Gross Margin 20% 20% 20% 21% Business tax (19) (23) (24) (25) 9% Selling expenses (87) (104) (109) (113) 9% Admin. expenses (264) (310) (327) (337) 9% Asset impairment losses / Fair value changes (104) (34) (11) (7) -59% Other gain / (losses) 377 151 151 151 -26% Operating profit 850 806 894 954 4% Net finance charges 5 9 5 10 21% Share of JCE 1 1 1 1 0% Profit before taxes 856 816 899 964 4% Tax (106) (122) (135) (145) 11% Minorities 1 0 0 0 -100% Pre-exceptional profit 751 694 764 820 3% Dividend to preferred shareholders and perpetual capital securities 0 0 0 0 NA Exceptionals 0 0 0 0 NA Net profit 751 694 764 820 3%



We expect FCF to remain positive in 2019-21e, lowering net gearing from -9% in 2018 to -12%

in 2021e.

Exhibit 177. FAM: Net debt and cash flow forecasts

RMBm 2018 2019e 2020e 2021e

Net debt/(cash) (478) (265) (511) (867) Net debt to equity -9% -4% -8% -12% Cash from Operations 170 213 659 786 Cash from Investing 151 (200) (200) (200) FCF 321 13 459 586


95


Valuation and risks



a 2020e PE of 18x, 1SD below the historical average valuation. We believe this is reasonable

as we see limited growth potential for the EVA film industry given the rise in cell efficiency and

the downtrend in post-cell fabrication costs. Key assumptions in our DCF valuation model

include:




debt to be 4.3%. We use our 2020e debt-to-capital ratio of 0% as our long-term debt-to-

capital ratio.


changes in working capital) to expand at a CAGR of 4% in 2018-29e, reflecting steady

growth in demand.

Capital expenditure: We expect a steady capex of around RMB200m per annum in 2019-

29e, reflecting steady investment in maintenance.

Terminal growth rate at 2%, and we assume the company reaches a steady growth period

after 2029e.


96

Exhibit 178. FAM: Discounted cash flow valuation

RMBm 2015 2016 2017 2018 2019e 2020e 2021e 2022e

Profit after tax 647 848 585 750 694 764 820 860 YoY growth 31% -31% 28% -8% 10% 7% 5% Add: Depreciation & amortization 41 57 75 92 73 79 88 99 Net finance expense 30 5 13 (9) (9) (5) (10) (11) Operating cash flow before W/C changes 719 910 673 833 758 838 897 948 Changes in working capital (339) (540) (375) (693) (545) (179) (111) Net operating cash flow 380 370 297 140 213 659 786 948 Capex (199) (517) 241 151 (200) (200) (200) (202) Free Cash Flow 180 (147) 538 291 13 459 586 746 Discount Factor 1.00 0.92 0.84 Gross PPE 641 1,029 1,259 1,617 1,663 1,832 2,026 2,228 Depreciation Rate 6% 6% 6% 6% 4% 4% 4% 4% PV of FCF 459 538 629 RMBm 2023E 2024E 2025E 2026E 2027E 2028E 2029E Terminal

Value Profit after tax 904 949 996 1,046 1,098 1,131 1,165 YoY growth 5% 5% 5% 5% 5% 3% 3% Add: Depreciation & amortization 110 122 134 147 161 175 190 Net finance expense (12) (13) (14) (15) (16) (17) (18) Operating cash flow before W/C changes 1,002 1,058 1,117 1,179 1,243 1,289 1,337 Changes in working capital 0 0 0 0 0 0 0 Net operating cash flow 1,002 1,058 1,117 1,179 1,243 1,289 1,337 Capex (204) (206) (208) (210) (212) (214) (217) Free Cash Flow 798 852 909 968 1,031 1,075 1,120 16,411 Discount Factor 0.77 0.71 0.65 0.60 0.55 0.50 0.46 0.46 Gross PPE 2,432 2,638 2,846 3,056 3,269 3,483 3,700 Depreciation Rate 5% 5% 5% 5% 5% 5% 5% PV of FCF 617 604 592 579 565 541 517 7,579 Summary of PV (Enterprise Value) 13,220 Less: Net debt (incl. perpetual) 265 Equity value 13,485 Less: Minority interest (2) Shareholder Equity Value 13,483 Total shares issued by year-end 2019 (mn) 523 Per Share Value - Rmb 25.8 Assumptions Risk-free rate 2.5% ERPch 6.5% Beta 1.00 Cost of equity = RFR + BETA*ERPch 9.0% Cost of debt 5.0% Income tax 15% After tax cost of debt 4.3% Debt/Capital 0% WACC 9.0% Terminal Growth 2%

Source: Wind, company data, HSBC Qianhai Securities, E= HSBC Qianhai Securities estimates

97


Exhibit 179. FAM forward PE: Trading in-line with the historical average

Exhibit 180. FAM forward PB: Trading around the historical average


Exhibit 181. FAM: Earnings sensitivity to gross margin and revenue changes, 2020e


20% -7% 1% 9% 17% 25% 33% 41% 15% -11% -3% 5% 13% 20% 28% 36% 10% -14% -6% 1% 8% 16% 23% 31% 5% -17% -10% -3% 4% 11% 18% 25%

0% -20% -13% -7% 0% 7% 13% 20%

-5% -23% -17% -11% -4% 2% 9% 15% -10% -27% -20% -14% -8% -2% 4% 10% -15% -30% -24% -18% -13% -7% -1% 5% -20% -33% -28% -22% -17% -11% -6% -1%


Upside risks:

Stronger-than-expected contribution from the electronic material business: FAM is

actively diversifying into downstream areas like 3C and semiconductor. We expect the

electronic material business to contribute 13% earnings to the company in 2021e. Better-

than-expected contribution from the electronic material business could positively impact our

earnings forecasts.

Stronger-than-expected PV demand: Sales of EVA film are highly correlated with global

new PV installations. Hence, stronger-than-expected global PV installations could positively

impact our earnings forecasts.

Better-than-expected margin on lessening competition: Less competition could be

positive to the company’s gross margins.

-

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

Sep-14 Sep-15 Sep-16 Sep-17 Sep-18

PE Mean +1SD -1SD

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Sep-14 Sep-15 Sep-16 Sep-17 Sep-18

PB Mean +1SD -1SD


98


Year to 12/2018a 12/2019e 12/2020e 12/2021e


Revenue 4,810 5,740 6,050 6,243

EBITDA 943 880 973 1,042

Depreciation & amortisation -92 -73 -79 -88

Operating profit/EBIT 850 807 894 954

Net interest 5 9 5 10

PBT 856 816 899 964

HSBC Qianhai PBT 856 816 899 964

Taxation -106 -122 -135 -145

Net profit 751 694 764 820

HSBC Qianhai net profit 751 694 764 820


Cash flow from operations 170 213 659 786

Capex 151 -200 -200 -200

Cash flow from investment 151 -200 -200 -200

Dividends -242 -235 -217 -239

Change in net debt -131 213 -246 -356

FCF equity 993 567 643 707




Current assets 5,014 5,469 5,930 6,418

Cash & others 495 281 528 884

Total assets 6,456 7,038 7,622 8,222

Operating liabilities 845 968 1,005 1,025

Gross debt 55 55 55 55

Net debt -439 -226 -472 -829




Year to 12/2018a 12/2019e 12/2020e 12/2021e

Y-o-y % change

Revenue 4.9 19.3 5.4 3.2

EBITDA 23.4 -6.6 10.5 7.1

Operating profit 23.3 -5.1 10.8 6.7

PBT 26.8 -4.6 10.2 7.2

HSBC Qianhai EPS -1.4 -7.8 10.2 7.2

Ratios (%)

Revenue/IC (x) 1.0 1.1 1.0 1.0

ROIC 15.5 12.8 13.0 13.3

ROE 14.2 12.0 12.2 12.0

ROA 12.3 10.3 10.4 10.3

EBITDA margin 19.6 15.3 16.1 16.7


EBITDA/net interest (x)

Net debt/equity -7.9 -3.8 -7.2 -11.6

Net debt/EBITDA (x) -0.5 -0.3 -0.5 -0.8

CF from operations/net debt


EPS Rep (diluted) 1.44 1.33 1.46 1.57


DPS 0.45 0.42 0.46 0.49

Book value 10.63 11.50 12.55 13.66

Valuation data

Year to 12/2018a 12/2019e 12/2020e 12/2021e

EV/sales 3.5 2.9 2.8 2.6

EV/EBITDA 17.7 19.2 17.1 15.7

EV/IC 3.3 3.0 2.8 2.6

PE* 22.8 24.7 22.4 20.9

PB 3.1 2.9 2.6 2.4

FCF yield (%) 5.8 3.3 3.8 4.1

Dividend yield (%) 1.4 1.3 1.4 1.5


ESG metrics




CO2 reduction policy [Yes] Female board members (%) 14.3







Issuer information






Price relative


15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

55.00

60.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

55.00

60.00

2015 2016 2017 2018 2019

First Applied Material Rel to CSI 300 Index

Financials & valuation: First Applied Material Reduce

99

Eq

uitie

s ●

En

erg

y E

quip

me

nt &

Se

rvic

es

Ju

ne 2

019

Exhibit 182. Hangzhou First Applied Materials: Company structure, April 2019


Zhang Hong

PublicLinan Tongde

Industrial Investment

Hangzhou First Applied Material

（603806 CH)

Backsheet for PV module

Electronic material

100%

5.24%20.13%

Hu Weimin and 36 other individuals

EVA film Power

generation and solar

system EPCOther business

Hangzhou First Technology Group

53.63%

Lin Jianhua

25% 75%

21.3%


100

Glossary

Aluminum-laminated film: A type of multilayer composite material that contains aluminum foil.

The kind of film that FAM produces is specifically used in flexible packaging of lithium batteries.

The composite incorporates the good properties of both aluminum (as a metal) and plastics,

possessing high physical strength, acid and alkali resistance, good conductivity to heat and

ductility.

Balance of system (BOS): The balance of system encompasses all components of a

photovoltaic system other than the photovoltaic panels. This includes wiring, switches, a

mounting system, one or many solar inverters, a battery bank and battery charger.

Crystal-growing furnace: A piece of equipment used to produce monocrystalline ingot or multi-

crystalline bricks. It typically includes a chamber and hot zone, vacuum and gas system, rotate

system, power supply and controls, and water cooling system.

c-Si: Crystalline silicon (c-Si) is the crystalline forms of silicon, either multicrystalline silicon

(multi-Si) consisting of small crystals, or monocrystalline silicon (mono-Si), a continuous crystal.

Czochralski (CZ) method: A seed crystal silicon rod is placed on the surface of the molten

silicon in the crucible. It is pulled up while rotating it to form a monocrystalline ingot having the

same orientation of atoms as the seed crystal.

Diamond wire sawing (DWS): It is the process of using wire of various diameters and lengths,

impregnated with diamond dust of various sizes to cut through materials. Because of the

hardness of diamonds, this cutting technique can cut through almost any material that is softer

than the diamond abrasive. DWC is also more practical and less expensive than some other

cutting techniques.

Effective operating hours: Calculated as total electricity generated divided by total installed

capacity. It measures how effective the power generation asset is utilised. Normally, effective

operating hours for PV solar ranges from 700-1,500 hours per year.

EVA film: Film made of ethylene vinyl acetate (EVA), a chemical copolymer. The film is

essentially plastic sheets used to encapsulate the solar cell for protection.

Float-zone: Float-zone silicon is a high-purity alternative to crystals grown by the Czochralski

process. In forming a single crystal, a molten silicon is slowly passed along a rod or bar of

silicon, which acts as a filter. As a result, impurities in the molten region tend to stay in the

molten region rather than be incorporated into the solidified region. Therefore, the

concentrations of light impurities, such as carbon and oxygen, are extremely low.

Grid parity: Grid parity occurs when an alternative energy source can generate power at a

levelized cost of electricity (LCOE) that is less than or equal to the price of power from the

electricity grid. The term is most commonly used when discussing renewable energy sources,

notably solar power and wind power.

Levelized cost of electricity (LCOE): It is an economic assessment of the average total cost to

build and operate a power-generating asset over its lifetime divided by the total energy output of

the asset over that lifetime. It is calculated as the net present value of the total costs over life

time divided by total electrical energy produced over lifetime.

Light-sensitive film: A type of thin film material used to create copper pattern on the printed

circuit board (PCB). How it works: after exposure to a high intensity ultraviolet light, light-

sensitive film (together with the copper underneath) not protected by a piece of mask can be

washed away by chemicals. Part of the light-sensitive film (together with the copper underneath)

that was covered by the masks, however, can persist through chemical erosion. Therefore, by

101


washing away light-sensitive film that is not protected by the mask from the light, the process

creates a copper pattern that matches the mask pattern on the PCB.

Mono-Si: The entire sample of monocrystalline silicon is one single, continuous and unbroken

crystal as its structure contains no grain boundaries.

Multi-Si: Polycrystalline silicon is a material consisting of multiple small silicon crystals.

N-type cell: An n-type cell has the boron-doped Si wafer on the surface (positively charged),

and the phosphorus-doped wafer on the back (negatively charged).

N-type silicon: An n-type silicon is doped with phosphorus, which has one more electron than

silicon (making part of the cell negatively charged).

Passivation: Passivation involves the creation of an outer layer of shield material against

corrosion, created by chemical reaction with the base material, or allowed to build from

spontaneous oxidation in the air.

Photovoltaic (PV): The conversion of light into electricity using semiconducting materials that

exhibit the photovoltaic effect.

P-type cell: A p-type cell has the phosphorus-doped Si wafer on the surface (negatively

charged), and the boron-doped wafer on the back (positively charged).

P-type silicon: A p-type silicon usually dopes its silicon wafer with boron, which has one less

electron than silicon (making part of the cell positively charged).


102

This page has been left blank intentionally

103




104


105


Disclosure appendix

Analyst Certification

The following analyst(s), economist(s), or strategist(s) who is(are) primarily responsible for this report, including any analyst(s)

whose name(s) appear(s) as author of an individual section or sections of the report and any analyst(s) named as the covering

analyst(s) of a subsidiary company in a sum-of-the-parts valuation certifies(y) that the opinion(s) on the subject security(ies) or

issuer(s), any views or forecasts expressed in the section(s) of which such individual(s) is(are) named as author(s), and any other

views or forecasts expressed herein, including any views expressed on the back page of the research report, accurately reflect

their personal view(s) and that no part of their compensation was, is or will be directly or indirectly related to the specific

recommendation(s) or views contained in this research report: Corey Chan and Dun Wang

Important disclosures

Equities: Stock ratings and basis for financial analysis

HSBC and its affiliates, including the issuer of this report (“HSBC”) believes an investor's decision to buy or sell a stock should

depend on individual circumstances such as the investor's existing holdings, risk tolerance and other considerations and that

investors utilise various disciplines and investment horizons when making investment decisions. Ratings should not be used or

relied on in isolation as investment advice. Different securities firms use a variety of ratings terms as well as different rating

systems to describe their recommendations and therefore investors should carefully read the definitions of the ratings used in

each research report. Further, investors should carefully read the entire research report and not infer its contents from the rating

because research reports contain more complete information concerning the analysts' views and the basis for the rating.

From 23rd March 2015 HSBC has assigned ratings on the following basis:

The target price is based on the analyst’s assessment of the stock’s actual current value, although we expect it to take six to 12

months for the market price to reflect this. When the target price is more than 20% above the current share price, the stock will

be classified as a Buy; when it is between 5% and 20% above the current share price, the stock may be classified as a Buy or a

Hold; when it is between 5% below and 5% above the current share price, the stock will be classified as a Hold; when it is between

5% and 20% below the current share price, the stock may be classified as a Hold or a Reduce; and when it is more than 20%

below the current share price, the stock will be classified as a Reduce.

Our ratings are re-calibrated against these bands at the time of any 'material change' (initiation or resumption of coverage, change

in target price or estimates).

Upside/Downside is the percentage difference between the target price and the share price.

Prior to this date, HSBC’s rating structure was applied on the following basis:

For each stock we set a required rate of return calculated from the cost of equity for that stock’s domestic or, as appropriate,

regional market established by our strategy team. The target price for a stock represented the value the analyst expected the

stock to reach over our performance horizon. The performance horizon was 12 months. For a stock to be classified as Overweight,

the potential return, which equals the percentage difference between the current share price and the target price, including the

forecast dividend yield when indicated, had to exceed the required return by at least 5 percentage points over the succeeding 12

months (or 10 percentage points for a stock classified as Volatile*). For a stock to be classified as Underweight, the stock was

expected to underperform its required return by at least 5 percentage points over the succeeding 12 months (or 10 percentage

points for a stock classified as Volatile*). Stocks between these bands were classified as Neutral.

*A stock was classified as volatile if its historical volatility had exceeded 40%, if the stock had been listed for less than 12 months

(unless it was in an industry or sector where volatility is low) or if the analyst expected significant volatility. However, stocks which

we did not consider volatile may in fact also have behaved in such a way. Historical volatility was defined as the past month's

average of the daily 365-day moving average volatilities. In order to avoid misleadingly frequent changes in rating, however,

volatility had to move 2.5 percentage points past the 40% benchmark in either direction for a stock's status to change.


106

Rating distribution for long-term investment opportunities

As of 19 June 2019, the distribution of all independent ratings published by HSBC is as follows:

For the purposes of the distribution above the following mapping structure is used during the transition from the previous to current

rating models: under our previous model, Overweight = Buy, Neutral = Hold and Underweight = Sell; under our current model Buy

= Buy, Hold = Hold and Reduce = Sell. For rating definitions under both models, please see “Stock ratings and basis for financial

analysis” above.

For the distribution of non-independent ratings published by HSBC, please see the disclosure page available at

http://www.hsbcnet.com/gbm/financial-regulation/investment-recommendations-disclosures.

To view a list of all the independent fundamental ratings disseminated by HSBC during the preceding 12-month period, please

use the following links to access the disclosure page:

Clients of Global Research and Global Banking and Markets: www.research.hsbc.com/A/Disclosures

Clients of HSBC Private Banking: www.research.privatebank.hsbc.com/Disclosures

HSBC & Analyst disclosures

Disclosure checklist

Company Ticker Recent price Price date Disclosure

LONGI GREEN 601012.SS 21.77 18 Jun 2019 6, 7 TONGWEI 600438.SS 13.89 18 Jun 2019 4, 7

Source: HSBC

1 HSBC has managed or co-managed a public offering of securities for this company within the past 12 months.

2 HSBC expects to receive or intends to seek compensation for investment banking services from this company in the next 3

months.

3 At the time of publication of this report, HSBC Securities (USA) Inc. is a Market Maker in securities issued by this

company.

4 As of 31 May 2019, HSBC beneficially owned 1% or more of a class of common equity securities of this company.

5 As of 30 April 2019, this company was a client of HSBC or had during the preceding 12 month period been a client of

and/or paid compensation to HSBC in respect of investment banking services.


and/or paid compensation to HSBC in respect of non-investment banking securities-related services.


and/or paid compensation to HSBC in respect of non-securities services.

8 A covering analyst/s has received compensation from this company in the past 12 months.

9 A covering analyst/s or a member of his/her household has a financial interest in the securities of this company, as

detailed below.

10 A covering analyst/s or a member of his/her household is an officer, director or supervisory board member of this

company, as detailed below.

11 At the time of publication of this report, HSBC is a non-US Market Maker in securities issued by this company and/or in

securities in respect of this company

12 As of 13 Jun 2019, HSBC beneficially held a net long position of more than 0.5% of this company’s total issued share

capital, calculated according to the SSR methodology.

13 As of 13 Jun 2019, HSBC beneficially held a net short position of more than 0.5% of this company’s total issued share

capital, calculated according to the SSR methodology.

Buy 52% ( 30% of these provided with Investment Banking Services )

Hold 38% ( 27% of these provided with Investment Banking Services )

Sell 10% ( 21% of these provided with Investment Banking Services )

107


HSBC and its affiliates will from time to time sell to and buy from customers the securities/instruments, both equity and debt

(including derivatives) of companies covered in HSBC Research on a principal or agency basis or act as a market maker or

liquidity provider in the securities/instruments mentioned in this report.

Analysts, economists, and strategists are paid in part by reference to the profitability of HSBC which includes investment banking,

sales & trading, and principal trading revenues.

Whether, or in what time frame, an update of this analysis will be published is not determined in advance.

Non-U.S. analysts may not be associated persons of HSBC Securities (USA) Inc, and therefore may not be subject to FINRA

Rule 2241 or FINRA Rule 2242 restrictions on communications with the subject company, public appearances and trading

securities held by the analysts.

Economic sanctions imposed by the EU and OFAC prohibit transacting or dealing in new debt or equity of Russian SSI entities.

This report does not constitute advice in relation to any securities issued by Russian SSI entities on or after July 16 2014 and as

such, this report should not be construed as an inducement to transact in any sanctioned securities.

For disclosures in respect of any company mentioned in this report, please see the most recently published report on that company

available at www.hsbcnet.com/research. HSBC Private Banking clients should contact their Relationship Manager for queries

regarding other research reports. In order to find out more about the proprietary models used to produce this report, please contact

the authoring analyst.

Additional disclosures

1 This report is dated as at 20 June 2019.

2 All market data included in this report are dated as at close 14 June 2019, unless a different date and/or a specific time of

day is indicated in the report.

3 HSBC has procedures in place to identify and manage any potential conflicts of interest that arise in connection with its

Research business. HSBC's analysts and its other staff who are involved in the preparation and dissemination of

Research operate and have a management reporting line independent of HSBC's Investment Banking business.

Information Barrier procedures are in place between the Investment Banking, Principal Trading, and Research businesses

to ensure that any confidential and/or price sensitive information is handled in an appropriate manner.

4 You are not permitted to use, for reference, any data in this document for the purpose of (i) determining the interest

payable, or other sums due, under loan agreements or under other financial contracts or instruments, (ii) determining the

price at which a financial instrument may be bought or sold or traded or redeemed, or the value of a financial instrument,

and/or (iii) measuring the performance of a financial instrument.

5 This report may be a translation of a report authored in another language. If so, and if there is any discrepancy between

versions, the original-language version shall prevail.

6 At the time of publication of this report, HSBC Qianhai Securities Limited does not hold 1% or more of a class of common

equity securities of this company.

Production & distribution disclosures

1. This report was produced and signed off by the author on 20 Jun 2019 06:43 GMT.

2. In order to see when this report was first disseminated please see the disclosure page available at

https://research.hsbcqh.com.cn/R/34/2qVPWRP


108

Disclaimer *Legal entities as at 30 November 2017

‘UAE’ HSBC Bank Middle East Limited, Dubai; ‘HK’ The Hongkong and Shanghai Banking Corporation Limited, Hong Kong;

‘TW’ HSBC Securities (Taiwan) Corporation Limited; 'CA' HSBC Securities (Canada) Inc.; HSBC Bank, Paris Branch; HSBC

France; ‘DE’ HSBC Trinkaus & Burkhardt AG, Düsseldorf; 000 HSBC Bank (RR), Moscow; ‘IN’ HSBC Securities and Capital

Markets (India) Private Limited, Mumbai; ‘JP’ HSBC Securities (Japan) Limited, Tokyo; ‘EG’ HSBC Securities Egypt SAE,

Cairo; ‘CN’ HSBC Investment Bank Asia Limited, Beijing Representative Office; The Hongkong and Shanghai Banking

Corporation Limited, Singapore Branch; The Hongkong and Shanghai Banking Corporation Limited, Seoul Securities

Branch; The Hongkong and Shanghai Banking Corporation Limited, Seoul Branch; HSBC Securities (South Africa) (Pty)

Ltd, Johannesburg; HSBC Bank plc, London, Madrid, Milan, Stockholm, Tel Aviv; ‘US’ HSBC Securities (USA) Inc, New

York; HSBC Yatirim Menkul Degerler AS, Istanbul; HSBC México, SA, Institución de Banca Múltiple, Grupo Financiero

HSBC; HSBC Bank Australia Limited; HSBC Bank Argentina SA; HSBC Saudi Arabia Limited; The Hongkong and Shanghai

Banking Corporation Limited, New Zealand Branch incorporated in Hong Kong SAR; The Hongkong and Shanghai Banking

Corporation Limited, Bangkok Branch; PT Bank HSBC Indonesia; HSBC Qianhai Securities Limited

Issuer of report


Block 27 A&B, Qianhai Enterprise Dream Park, 63

Qianwan Yi Road, Shenzhen-Hong Kong Cooperation

Zone, Shenzhen, China

Phone number: +86 755 8898 3288

Website: www.hsbcqh.com.cn

In the People’s Republic of China (“PRC”) (Excluding special administrative regions of Hong Kong and Macao), this document is issued and approved by HSBC Qianhai Securities Limited for

the information of its clients only; it is not intended for and should not be distributed to retail customers in the PRC. HSBC Qianhai Securities Limited is regulated by the China Securities

Regulatory Commission (“CSRC”) and is qualified to engage in Securities Investment Advisory Business in the PRC [91440300MA5EPLHG1B]. All inquiries by recipients must be directed to

the HSBC Qianhai Securities Limited contact in the PRC. If it is received by a customer of an affiliate of HSBC Qianhai Securities Limited, its provision to the recipient is subject to the terms of

business in place between the recipient and such affiliate. This document is being supplied to you solely for your information and may not be redistributed or passed on, directly or indirectly, to

any other person, in whole or in part, for any purpose. This document is not and should not be construed as an offer to sell or solicitation of an offer to purchase or subscribe for any investment.

HSBC Qianhai Securities Limited has based this document on information obtained from sources it believes to be reliable but which it has not independently verified; HSBC Qianhai Securities

Limited makes no guarantee, representation or warranty and accepts no responsibility or liability as to its accuracy or completeness. The decision and responsibility on whether or not to purchase,

subscribe or sell (as applicable) must be taken by the investor. No consideration has been given to the particular investment objectives, financial situation or particular needs of any recipient.

Any form of written or verbal commitment on the sharing of gains or losses from the securities investment between the investors and the HSBC Qianhai Securities Limited arising from the

research services provided are not permitted and shall be invalid. Expressions of opinion are those of the Research Division of HSBC Qianhai Securities Limited only and are subject to change

without notice. From time to time research analysts conduct site visits of covered issuers. HSBC Qianhai Securities Limited and its affiliates and/or their officers, directors and employees may

have positions in any securities mentioned in this document (or in any related investment) and may from time to time add to or dispose of any such securities (or investment) to the extent

permitted by law. HSBC Qianhai Securities Limited and its affiliates may act as market maker or have assumed an underwriting commitment in the securities of companies discussed in this

document (or in related investments), may sell them to or buy them from customers on a principal basis and may also perform or seek to perform investment banking or underwriting services

for or relating to those companies and may also be represented on the supervisory board or any other committee of those companies. This document may contain hyperlinks to external websites

for convenience of its recipients. HSBC Qianhai Securities Limited are not responsible for any content therein. HSBC policies prohibit research analysts from accepting payment or reimbursement

for travel expenses from the issuer for such visits.

The Hongkong and Shanghai Banking Corporation Limited owns 51% and Qianhai Financial Holdings Co., Ltd. (“QFH”) owns 49% of shares in HSBC Qianhai Securities Limited, which prepared

and/or contributed to this research report. HSBC Qianhai Securities Limited has established policies and procedures reasonably designed to prevent QFH from exercising direct or indirect

influence over the content of HSBC Qianhai Securities Limited research reports and the choice of companies that will be the subject of research reports. Furthermore, HSBC Qianhai Securities

Limited has established additional policies and procedures reasonably designed to prevent any person or entity, whether from within HSBC Qianhai Securities Limited, QFH or otherwise, from

influencing the activities of the HSBC Qianhai Securities Limited’s research analysts or the content of research reports.

The information and opinions in this research report were distributed by The Hongkong and Shanghai Banking Corporation Limited in Hong Kong, which accepts legal responsibility for its

contents. This research report is intended to provide information to The Hongkong and Shanghai Banking Corporation Limited's institutional and professional investor (as defined in Schedule 1

to the Securities and Future Ordinance (Cap. 571)) ("SFO") customers only; it is not intended for and should not be distributed to customers in Hong Kong who are not "professional investors".

The Hongkong and Shanghai Banking Corporation Limited is regulated by the Hong Kong Monetary Authority. The Hongkong and Shanghai Banking Corporation Limited makes no

representations that the products or services mentioned in this document are available to persons in Hong Kong. Any enquiries with respect to any matters arising from, or in connection with,

this research report should be directed to the recipient’s contact person at The Hongkong and Shanghai Banking Corporation Limited. The analyst(s) named in this research report who is(are)

not employed by and/or accredited to The Hongkong and Shanghai Banking Corporation Limited is(are) not licensed to carry on any regulated activities in Hong Kong under the SFO. The

analyst(s) is(are) only named in this research report as being a source of the information contained herein and does(do) not purport to carry on any regulated activities in Hong Kong under the

SFO or hold himself/herself(themselves) out as being able to do so.

HSBC Securities (USA) Inc., a US-registered broker-dealer and member of FINRA, accepts responsibility for the content of this research report prepared by its non-US foreign affiliate. All U.S.

persons receiving and/or accessing this report and wishing to effect transactions in any security discussed herein should do so with HSBC Securities (USA) Inc. in the United States and not with

its non-US foreign affiliate, the issuer of this report.In the UK this report may only be distributed to persons of a kind described in Article 19(5) of the Financial Services and Markets Act 2000

(Financial Promotion) Order 2005. The protections afforded by the UK regulatory regime are available only to those dealing with a representative of HSBC Bank plc in the UK. In Singapore, this

publication is distributed by The Hongkong and Shanghai Banking Corporation Limited, Singapore Branch for the general information of institutional investors or other persons specified in

Sections 274 and 304 of the Securities and Futures Act (Chapter 289) (“SFA”) and accredited investors and other persons in accordance with the conditions specified in Sections 275 and 305

of the SFA. Only Economics or Currencies reports are intended for distribution to a person who is not an Accredited Investor, Expert Investor or Institutional Investor as defined in SFA. The

Hongkong and Shanghai Banking Corporation Limited, Singapore Branch accepts legal responsibility for the contents of reports pursuant to Regulation 32C(1)(d) of the Financial Advisers

Regulations. This publication is not a prospectus as defined in the SFA. This publication is not a prospectus as defined in the SFA. It may not be further distributed in whole or in part for any

purpose. The Hongkong and Shanghai Banking Corporation Limited Singapore Branch is regulated by the Monetary Authority of Singapore. Recipients in Singapore should contact a "Hongkong

and Shanghai Banking Corporation Limited, Singapore Branch" representative in respect of any matters arising from, or in connection with this report. Please refer to The Hongkong and Shanghai

Banking Corporation Limited Singapore Branch’s website at www.business.hsbc.com.sg for contact details. In Australia, this publication has been distributed by The Hongkong and Shanghai

Banking Corporation Limited (ABN 65 117 925 970, AFSL 301737) for the general information of its “wholesale” customers (as defined in the Corporations Act 2001). Where distributed to retail

customers, this research is distributed by HSBC Bank Australia Limited (ABN 48 006 434 162, AFSL No. 232595). These respective entities make no representations that the products or services

mentioned in this document are available to persons in Australia or are necessarily suitable for any particular person or appropriate in accordance with local law. No consideration has been

given to the particular investment objectives, financial situation or particular needs of any recipient. This publication is distributed in New Zealand by The Hongkong and Shanghai Banking

Corporation Limited, New Zealand Branch incorporated in Hong Kong SAR.

In Japan, this publication has been distributed by HSBC Securities (Japan) Limited. It may not be further distributed in whole or in part for any purpose. In Korea, this publication is distributed

by The Hongkong and Shanghai Banking Corporation Limited, Seoul Securities Branch ("HBAP SLS") for the general information of professional investors specified in Article 9 of the Financial

Investment Services and Capital Markets Act (“FSCMA”). This publication is not a prospectus as defined in the FSCMA. It may not be further distributed in whole or in part for any purpose.

HBAP SLS is regulated by the Financial Services Commission and the Financial Supervisory Service of Korea.

In Canada, this document has been distributed by HSBC Securities (Canada) Inc. (member IIROC), and/or its affiliates. The information contained herein is under no circumstances to be

construed as investment advice in any province or territory of Canada and is not tailored to the needs of the recipient. No securities commission or similar regulatory authority in Canada has

reviewed or in any way passed judgment upon these materials, the information contained herein or the merits of the securities described herein, and any representation to the contrary is an

offense.

If you are an HSBC Private Banking (“PB”) customer with approval for receipt of relevant research publications by an applicable HSBC legal entity, you are eligible to receive this publication. To

be eligible to receive such publications, you must have agreed to the applicable HSBC entity’s terms and conditions (“KRC Terms”) for access to the KRC, and the terms and conditions of any

other internet banking service offered by that HSBC entity through which you will access research publications using the KRC. Distribution of this publication is the sole responsibility of the

HSBC entity with whom you have agreed the KRC Terms.

If you do not meet the aforementioned eligibility requirements please disregard this publication and, if you are a customer of PB, please notify your Relationship Manager. Receipt of research

publications is strictly subject to the KRC Terms, which can be found at https://research.privatebank.hsbc.com/ – we draw your attention also to the provisions contained in the Important Notes

section therein.

© Copyright 2019, HSBC Qianhai Securities Limited, ALL RIGHTS RESERVED. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any

means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of HSBC Qianhai Securities Limited. MCI (P) 065/01/2019 and MCI (P) 008/02/2019

[1121859]

Head of Research, HSBC Qianhai Securities Steven Sun +86 755 8898 3158 [email protected]

Deputy Head of Research, Head of Research Product, HSBC Qianhai Securities John Chung

China Equity Strategy

Analyst, Head of China Equity Strategy Research Steven Sun +86 755 8898 3158 [email protected]

Associate Kate Zhang

Consumer

Analyst, Head of A-share Food & Beverage and Pulp & Paper Research Katharine Song +86 755 8898 3142 [email protected]

Analyst, A-share Food & Beverage and Pulp & Paper Research Darron Xue +86 755 8898 3407 [email protected]

Associate Joseph Zhou

Associate Li Quan

Healthcare

Analyst, Head of Greater China Healthcare Research Zhijie Zhao +86 755 8898 3144 [email protected]

Analyst, A-share Healthcare Research Esther Wen +86 755 8898 3492 [email protected]

Industrials and Environmental Services

Analyst, Head of A-share Industrials and Environmental Research Bonan Li +86 755 8898 3139 [email protected]

Infrastructure

Analyst, Head of A-share Infrastructure Research Corey Chan +86 755 8898 3404 [email protected]

Analyst, A-share Infrastructure Research Dun Wang +86 755 8898 3460 [email protected]

Petrochemical & New Materials

Analyst, Head of A-share Petrochemical and New Materials Eric Shen +86 755 8898 3403 [email protected]

Associate Yi Ru

Telecoms, Media & Technology

Analyst, Head of A-share Technology Hardware Research Frank He +86 755 8898 3136 [email protected]

Analyst, Head of A-share Media & Internet Research Yi Guo +86 755 8898 3137 [email protected]

Analyst, A-share Media & Internet Jing Han +86 755 8898 3147 [email protected]

Analyst, Head of A-share IT Software Research Sijie Ma +86 755 8898 3140 [email protected]

Associate Chase Ding

Transportation and Logistics

Analyst, Head of A-share Transportation & Logistics Research David Wu +86 755 8898 3436 [email protected]

Associate Sonia Luo

HSBC Qianhai Research Team

mailto:[email protected]




Sp

otligh

t: Ch

ina S

olar Equ

ipm

ent

Jun

e 2019Eq

uities // En

ergy Eq

uip

men

t & S

ervices

Issuer of report: HSBC Qianhai Securities Limited

Block 27 A&B, Qianhai Enterprise Dream Park 63 Qianwan Yi Road, Shenzhen-Hong Kong Cooperation Zone,

Shenzhen, ChinaTelephone: +86 755 8898 3288

Website: https://research.hsbcqh.com.cn

* Employed by a non-US affiliate of HSBC Securities (USA) Inc, and is not registered / qualified pursuant to FINRA regulations

Our multi-channel researchWebsite

® All our Research, including multi-asset, thematic and daily reports, is available on research.hsbc.com

® Subscribe by asset class, analyst, region, country, sector, ticker, and periodical

Mobile app ® Available on iOS and Android phones and tablets

® Download the app from the App Store or Google Play, search Global Research

Podcasts ® Subscribe to all our audio and video reports on your personal phone or tablet

® Special weekly podcasts include The Macro Brief and The Equity Brief

Other platforms ® Bloomberg ® Bluematrix ® Factset ® Markit Hub

® Red Deer ® Refinitiv ® Research Exchange ® ResearchPool

® S&P Global Market Intelligence ® Visible Alpha/ONEaccess

Main contributorsCorey Chan* (S1700518100001)Head of A-share Infrastructure ResearchHSBC Qianhai Securities [email protected]+86 755 8898 3404

Corey Chan joined HSBC Qianhai Securities Limited in 2018 as Head of A-share Infrastructure Research. Previously, Corey worked as an equity analyst for a US investment bank in Hong Kong, where he focused on the China capital goods sector. He holds a bachelor of finance degree from the University of Hong Kong.

Dun Wang* (S1700519060002)Analyst, A-share InfrastructureHSBC Qianhai Securities [email protected]

+86 755 8898 3460

Dun Wang joined HSBC Qianhai Securities Limited in 2018 as an equity research analyst, covering A-share infrastructure, electrical equipment and renewable energy sectors. Previously, Dun worked as an offshore research assistant for a US investment bank and an analyst for a private equity firm. He holds a master’s degree in accounting from the Foster School of Business at the University of Washington and a bachelor’s degree in commerce (honours) from the University of British Columbia.

https://itunes.apple.com/app/id1248271935

http://play.google.com/store/apps/details?id=com.hsbc.hsbcglobalresearch.rapp

www.research.hsbc.com

东西智库，专注知识分享

Documents

China Solar Equipment...world’s largest solar market in 2010. China soon took over and solar capacity grew from 19GW in 2013 to 175GW in 2018. But this led to overcapacity and a