Taiwan Compound Semiconductor Sector

DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

CREDIT SUISSE SECURITIES RESEARCH & ANALYTICS BEYOND INFORMATION®

Client-Driven Solutions, Insights, and Access

11 November 2015

Asia Pacific/Taiwan

Equity Research

Technology (Technology - Foundries TW (Asia))

Taiwan Compound Semiconductor Sector

INITIATION

Riding on wireless spec upgrades

Figure 1: RF component revenues to see 13% CAGR over 2015-17

0%

10%

20%

30%

0

4,500

9,000

13,500

18,000

2013 2014 2015E 2016E 2017E

US$

mn

RF component revenues Growth (YoY) Source: Qorvo, Skyworks, Richwave, Credit Suisse estimates

■ Initiating coverage on Win Semiconductors Corp (Win Semi) (OUTPERFORM, TP NT$62), Visual Photonics Epitaxy Co., Ltd (VPEC) (OUTPERFORM, TP NT$47) and Advanced Wireless Semiconductor Company (AWSC) (NEUTRAL, TP NT$88). We initiate coverage on the Taiwan Compound Semiconductor Sector, as we expect the sector to benefit from the 13% RF component revenue CAGR over 2015-17, driven by (1) the LTE migration leading to higher RF content and more infrastructure build, and (2) the WiFi spec upgrade for higher data rates and IoT expanding the WiFi TAM. We prefer Win Semi due to its strong customer relationship, technology independence and product mix migration, and VPEC on its diversified customer base, expansion into the optical space and a strong

balance sheet.

■ Duopoly in compound semiconductor foundry. We believe compound semiconductor (e.g., GaAs) should stay as the choice for RF components (especially power amplifiers) in most applications, given its superior properties against silicon in handling high power and high frequency signals, as well as the still evolving performance and increasing levels of integration. We are of the view that the duopolistic structure, increasing outsourcing trend and high switching costs should point to a relatively stable competitive

landscape, with slim chances of new entrants in the near to medium term.

■ LTE a two-fold driver for RF component growth. We expect the 26%

CAGR in LTE mobile device shipments and the much higher RF content

(US$6-15 vs ~US$3 for 3G) to support 12% RF revenue CAGR in mobile

applications over 2015-17. Moreover, LTE subscriber net additions

expanding at a 25% CAGR in 2015-17 could lead to more infrastructure

expansions, driving a 5% RF revenue CAGR in this space over 2015-17.

■ WiFi spec upgrade and IoT enlarging the total addressable market. We see the growth opportunities for RF components in the WiFi market coming from (1) the upgrade to the more advanced 802.11ac standard for higher speed, and (2) IoT further expanding the total addressable market. We

forecast a 22% RF revenue CAGR in the WiFi application over 2015-17.

■ Risks. Key risks to our view include weaker LTE device shipment, slower WiFi spec upgrade and weaker IoT device proliferation.

Research Analysts

Derrick Yang

886 2 2715 6367

[email protected]

Jerry Su

886 2 2715 6361

[email protected]

11 November 2015

Taiwan Compound Semiconductor Sector 2

Focus charts and table Figure 2: RF components seeing 13% CAGR in 2015-17 Figure 3: Duopoly in compound semiconductor foundry

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

20,000

2013 2014 2015E 2016E 2017E

US

$mn

Mobile WiFi Infrastructure Others

2015-17 RF revenue CAGR

- 13% for overall RF component

- 12% for mobile

- 5% for infrastructure

- 22% for WiFi

Win Semi 63%

AWSC 34%

GCS 2% Others 1%

Source: Qorvo, Skyworks, Richwave, Credit Suisse estimates Source: Gartner, Credit Suisse estimates

Figure 4: Much higher RF content for LTE smartphones Figure 5: Deployment of smaller LTE cell sites to increase

0

2

4

6

8

10

12

14

16

18

2G 3G Regional LTE Global LTE

US

$

Filters Switches Powr amplifiers Others

0

100

200

300

400

500

600

700

800

2013 2014 2015E 2016E 2017E

Tho

usan

d un

its

Micro cell Pico cell Small cell

Source: Qorvo, Credit Suisse Source: IDC Credit Suisse

Figure 6: WiFi spec upgrade playing out Figure 7: IoT device growth to expand the TAM for WiFi

18%27%

35%43%

97%82%

73%65%

57%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015E 2016E 2017E

802.11ac 802.11a/b/g/n

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

2013 2014 2015E 2016E 2017E 2018E 2019E 2020E

mn

units

Automotive Consumer Generic Business Vertical Specific

Source: Richwave, Skyworks, Credit Suisse estimates Source: Gartner, Credit Suisse

Figure 8: Valuation comparison—RF component supply chain

Reporting Price Marketcap CS 12mth

Company Currency 11/10/2015 US$ mn Rating Target 2015 2016 2015 2016 2015 2016 2015 2016 2015 2016

Compound semiconductor foundry

3105.TWO Win Semi TWD 48.50 903 O 62.0 4.53 5.03 10.7 9.6 1.7 1.6 16.4 17.0 5.6 6.2

8086.TWO AWSC TWD 78.60 329 N 88.0 7.83 8.37 10.0 9.4 3.5 3.1 40.0 34.7 6.8 7.5

Compound semiconductor epiwafer

2455.TW VPEC TWD 36.50 275 O 47.0 2.39 2.57 15.2 14.2 2.5 2.5 16.6 17.6 6.9 7.4

IQE.L IQE * GBP 24.25 244 NC NA 2.53 2.80 9.6 8.7 1.2 1.0 12.9 12.6 - -

3081.TWO Landmark Opto TWD 425.00 909 NC NA 13.75 18.15 30.9 23.4 13.4 9.9 56.9 48.6 1.7 2.4

4005.T Sumitomo Chemical JPY 684.0 9,075 N 650.0 31.94 55.68 21.4 12.3 1.4 1.3 7.3 11.0 1.3 2.0

RF component IDM

SWKS.OQ Skyworks * USD 84.89 16,192 NC NA 5.27 6.25 16.1 13.6 5.0 4.1 33.3 32.3 0.7 1.2

QRVO.OQ Qorvo * USD 54.09 8,088 NC NA 4.75 4.59 11.4 11.8 NA 1.4 NA 8.6 NA -

AVGO.OQ Avago USD 126.50 34,833 NC NA 7.81 9.10 16.2 13.9 8.2 5.4 58.5 47.4 - -

6981.T Murata JPY 19,225.0 33,039 O 22,000.0 792.19 1,036.45 24.3 18.5 3.6 3.3 16.1 18.5 0.9 1.5

RF component fabless

MCHP.OQ Microchip USD 47.82 9,712 O 52.0 2.66 2.61 18.0 18.3 4.7 4.8 28.4 24.1 - -

Average 16.7 14.0 4.5 3.5 28.6 24.8 2.4 2.6

Dividend yield (%)ROAE (%)EPS P/E (X) P/B (X)

Source: Company data, the BLOOMBERG PROFESSIONAL™ service, Credit Suisse estimates; "*" estimates represent IBES consensus

11 November 2015


Riding on wireless spec upgrades We initiate coverage on the Compound Semiconductor Sector in Taiwan. We expect the

sector to benefit from the 13% RF component revenue CAGR over 2015-17 driven by (1)

the LTE migration leading to higher RF content and more infrastructure build, and (2) the

WiFi spec upgrade for higher data rates and IoT expanding the WiFi total addressable

market (TAM). We expect the competitive landscape to be relatively stable for the

compound semiconductor foundry sector given the duopolistic structure, increasing

outsourcing trend and high switching costs, with slim likelihood of new entrants in the near

to medium term. We prefer Win Semi and VPEC this space.

Duopoly in compound semiconductor foundry sector

Though silicon-based RF components have been promoted for years, we believe

compound semiconductors (e.g., GaAs) should stay as the choice for

RF components (especially power amplifiers) in most applications, given its superior

characteristics in handling high power and high frequency signals, as well as the still

evolving performance and increasing levels of integration. We are of the view that the

duopolistic structure, increasing outsourcing trend and high switching costs should point to

a relatively stable competitive landscape for the compound semiconductor foundry market,

with slim likelihood of new entrants in the near to medium term.

LTE a two-fold driver for RF component growth

LTE migration is the major driver for RF component demand on both device and

infrastructure fronts. We expect the 26% CAGR in LTE device shipments (penetration into

overall cellular devices to 70% in 2017 from 33% in 2014) and the much higher RF content

for complex frequency bands in LTE models (US$6-15 vs ~US$3 for 3G) to support a 12%

revenue CAGR in mobile applications over 2015-17. In addition, LTE subscriber net

additions expanding at a 25% CAGR should lead to more infrastructure build, especially

complementary smaller cell sites, driving a 5% CAGR in this space over 2015-17.

WiFi spec upgrade and IoT enlarging the TAM

We see the growth opportunities for RF components in the WiFi market coming from (1)

the upgrade into the more advanced 802.11ac standard, which tends to support dual

bands (2.4 GHz and 5.0 GHz) and more spatial streams; and (2) IoT further expanding the

total addressable market by driving the annual connected device shipments at a CAGR of

39% to 8.3 bn units in 2020. We forecast a 22% RF revenues CAGR from WiFi in 2015-17.

Prefer Win Semi (3105.TWO) and VPEC (2455.TW)

We prefer Win Semi and VPEC over AWSC. We think Win Semi is well positioned with its

strong customer relationship, technology independence and product mix shift towards

higher margin WiFi and infrastructure applications. Our TP of NT$62 is based on 2016E

P/E of 12x, which is the average of its historical trading range of 6-20x.

We also prefer VPEC (2455.TW) as the second largest epiwafer supplier in the world, on

its diversified customer base, the expansion into optical communication and the high

dividend payout supported by its solid balance sheet/free cash flows. Our TP of NT$47 is

based on 2016E P/E of 18x (17x excluding the net cash on hand), which is similar to the

average of its historical trading range of 4-30x.

We rate ASWC (8086.TWO) NEUTRAL. Though we believe it is a quality name, its heavy

reliance on Skyworks could be a mixed blessing, as the benefit of the close relationship

could be coupled with higher business volatility. Our TP of NT$88 is based on 2016E P/E

of 11x, which is at a discount vs Win Semi, as we factor in the potential single client risk.

Despite AWSC having higher ROEs, we believe the stock is fairly valued at the current

levels.

13% RF component

revenue CAGR over 2015-

17

Relatively stable competitive

landscape with slim chances

of new entrants in the near

to medium term

LTE migration to drive RF

revenue CAGR of 12% for

devices and 5% for

infrastructure over 2015-17

WiFi spec upgrade and IoT

to drive 22% RF revenue

CAGR over 2015-17

Initiating coverage on Win

Semi and VPEC with

OUTPERFORM ratings

Initiating coverage on

AWSC with NEUTRAL

11 November 2015


Valuation comparison Figure 9: Valuation comparison – RF component supply chain

Reporting Price Marketcap CS 12mth

Company Currency 11/10/2015 US$ mn Rating Target 2015 2016 2015 2016 2015 2016 2015 2016 2015 2016

Compound semiconductor foundry

3105.TWO Win Semi TWD 48.50 903 O 62.0 4.53 5.03 10.7 9.6 1.7 1.6 16.4 17.0 5.6 6.2

8086.TWO AWSC TWD 78.60 329 N 88.0 7.83 8.37 10.0 9.4 3.5 3.1 40.0 34.7 6.8 7.5

Compound semiconductor epiwafer

2455.TW VPEC TWD 36.50 275 O 47.0 2.39 2.57 15.2 14.2 2.5 2.5 16.6 17.6 6.9 7.4

IQE.L IQE * GBP 24.25 244 NC NA 2.53 2.80 9.6 8.7 1.2 1.0 12.9 12.6 - -

3081.TWO Landmark Opto TWD 425.00 909 NC NA 13.75 18.15 30.9 23.4 13.4 9.9 56.9 48.6 1.7 2.4

4005.T Sumitomo Chemical JPY 684.0 9,075 N 650.0 31.94 55.68 21.4 12.3 1.4 1.3 7.3 11.0 1.3 2.0

RF component IDM

SWKS.OQ Skyworks * USD 84.89 16,192 NC NA 5.27 6.25 16.1 13.6 5.0 4.1 33.3 32.3 0.7 1.2

QRVO.OQ Qorvo * USD 54.09 8,088 NC NA 4.75 4.59 11.4 11.8 NA 1.4 NA 8.6 NA -

AVGO.OQ Avago USD 126.50 34,833 NC NA 7.81 9.10 16.2 13.9 8.2 5.4 58.5 47.4 - -

6981.T Murata JPY 19,225.0 33,039 O 22,000.0 792.19 1,036.45 24.3 18.5 3.6 3.3 16.1 18.5 0.9 1.5

RF component fabless

MCHP.OQ Microchip USD 47.82 9,712 O 52.0 2.66 2.61 18.0 18.3 4.7 4.8 28.4 24.1 - -

Average 16.7 14.0 4.5 3.5 28.6 24.8 2.4 2.6

Dividend yield (%)ROAE (%)EPS P/E (X) P/B (X)

Source: Company data, the BLOOMBERG PROFESSIONAL™ service, Credit Suisse estimates; "*" estimates represent IBES consensus

Figure 10: Win Semi—12-month forward P/E Figure 11: Win Semi—12-month forward P/B

5.0

7.0

9.0

11.0

13.0

15.0

17.0

19.0

21.0

Dec-1

1

Apr-

12

Aug-1

2

Dec-1

2

Apr-

13

Aug-1

3

Dec-1

3

Apr-

14

Aug-1

4

Dec-1

4

Apr-

15

Aug-1

5

Average = 12

+1 Std dev = 14

-1 Std dev = 9

+2 Std dev = 17

-2 Std dev = 6

Win Semi PE (x)

0.5

0.7

0.9

1.1

1.3

1.5

1.7

1.9

2.1

2.3

2.5

Dec-1

1

Apr-

12

Aug-1

2

Dec-1

2

Apr-

13

Aug-1

3

Dec-1

3

Apr-

14

Aug-1

4

Dec-1

4

Apr-

15

Aug-1

5

Average = 1.52

+1 Std dev = 1.8

-1 Std dev = 1.3

+2 Std dev = 2.0

-2 Std dev = 1.0

Win Semi PB (x)

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

Figure 12: VPEC—12-month forward P/E Figure 13: VPEC—12-month forward P/B

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

Jan-0

9

May-0

9

Sep-0

9

Jan-1

0

May-1

0

Sep-1

0

Jan-1

1

May-1

1

Sep-1

1

Jan-1

2

May-1

2

Sep-1

2

Jan-1

3

May-1

3

Sep-1

3

Jan-1

4

May-1

4

Sep-1

4

Jan-1

5

May-1

5

Sep-1

5

Average = 19

+1 Std dev = 24

-1 Std dev = 13

+2 Std dev = 29

-2 Std dev = 8

VPEC PE (x)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Jan-0

9

May-0

9

Sep-0

9

Jan-1

0

May-1

0

Sep-1

0

Jan-1

1

May-1

1

Sep-1

1

Jan-1

2

May-1

2

Sep-1

2

Jan-1

3

May-1

3

Sep-1

3

Jan-1

4

May-1

4

Sep-1

4

Jan-1

5

May-1

5

Sep-1

5

Average = 2.6

+1 Std dev = 3.3

-1 Std dev = 1.9

+2 Std dev = 4.0

-2 Std dev = 1.2

VPEC PB (x)


11 November 2015


Duopoly in the compound semiconductor foundry sector Though silicon-based power amplifiers (PAs) have been promoted for years, we believe

compound semiconductor (e.g., GaAs) should stay as the choice for RF components

(especially PA) in most applications, given its superior characteristics in handling high

power and high frequency signals with efficient power consumption, higher data rates and

better signal quality, as well as the still evolving performance and increasing levels of

integration. We are of the view that the duopolistic structure (Win Semi and AWSC),

increasing outsourcing trend and high switching costs should point to a relatively stable

competitive landscape for the compound semiconductor foundry market, with slim chances

of new entrants in the near to medium term.

RF front end playing a key part in wireless

communication

What is the RF front end?

The radio frequency (RF) front end is a key module in the wireless communication

mechanism. In the typical handset configuration, it refers to the components between the

antenna and the baseband processor, including switches, power amplifiers, low-noise

amplifiers (LNA), filters and mixers. The function of the RF front end is to take the signal

received by the antenna, amplify it, filter out noises, and down-convert its frequency in the

receiver mode. Similarly, in the transmitter mode, it takes the signal sent out by the

baseband processor, up-converts its frequency, filters out the noises and amplifies the

voltage before the signal goes to the antenna for delivery. The RF front end is important in

wireless communication because it has direct influence on the quality of signals both

received and transmitted.

Handset being the biggest consumer of RF front end components

RF front ends could be commonly seen in various wireless communication devices and

equipment, such as handsets, WiFi-enabled devices, GPS systems, gateway routers,

cellular base stations, satellite TVs, etc. With the annual shipment of around two billion

units and the increasing cellular frequency bands, handsets have been the biggest

consumer of RF components in the past few years.

Figure 14: Simplified RF front end block diagram

Low noise amplifier

PA

Mixer

Mixer

Antenna switch

Antenna

Filter

Filter

VCO/PLL

Baseband

RF front end

Source: Credit Suisse

RF component market dominated by four players

Skyworks, Qorvo, Avago and Murata taking 79% of the market share

The RF component market has been dominated by a few players, as the top four players

Skyworks, Qorvo, Avago and Murata, collectively account for 79% of the global share. We

believe that the competitive landscape should remain similar in the near future as the trend

of modularisation should favour bigger players, who have more comprehensive product

portfolios to put into modules, instead of supplying discrete components.

RF front end refers to

components including

switches, PAs, LNAs, filters

and mixers

Handset being the biggest

consumer for RF

components

RF component market

dominated by Skyworks,

Qorvo, Avago and Murata

11 November 2015


Other players in the market include MACOM, Analog Devices, NXP, Freescale, Sumitomo

Electric, Mitsubishi Electric, RDA, HiSilicon, Richwave, etc.

While Chinese companies have been aggressive expanding into many tech sectors in the

past few years such as semiconductor, hardware components, etc., we haven't seen

Chinese player gaining too much traction in the RF component space yet. Our checks

suggest that even in the China smartphone market, ~90% of the RF components are

supplied by international vendors and only ~10% are supplied by their domestic vendors

(such as RDA, HiSilicon, Vanchip, etc.). We think this reflects that fact that performance is

usually prioritised over costs in RF components, as they have a significant influence over

signal quality.

Figure 15: Global RF component end applications (2014) Figure 16: Global RF component market share (2014)

Mobile 60%WiFi 21%

Infrastructure 12%

Others 7%

Skyworks 24%

Qorvo 22%

Avago 18%

Murata 15%

Others 21%

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

Foundry business model to gain more traction

Similar to the silicon semiconductor sector, fabless companies in the compound

semiconductor sector could leverage the manufacturing capability of the foundry partners

to avoid the equipment investment. In addition, integrated device manufacturers (IDMs)

have also been outsourcing their orders to foundry players to alleviate the capex burden

and focus their resources on technology innovation.

Figure 17: RF component supply chain

Backend

VPEC (2455 TT)

InteliEpi (4971 TT)

IQE (IQE LN)

Lingsen (2369 TT)

Tong Hsing (6271 TT)

Shunsin (6451 TT)

Win Semi (3105 TT)

AWSC (8086 TT)

GCS (4991 TT)

Epi Wafer Foundry

IDMFabless

Hitachi Cable (5812 JP)

Sumitomo Chem (4005 JP)

Furukawa Electric (5801 JP)

Murata (6981 JP)

Ampak (not listed)

Transcom (5222 TT)

Hexawave (not listed)

Microchip (MCHP US)

Richwave (4968 TT)

Maxtek (3315 TT)

RFIC (not listed)

RDA (not listed)

Avago (AVGO US)

Skyworks (SWKS US)

Qorvo (QRVO US)

Anadigics (ANAD US)

Qorvo (QRVO US)

Wavetek (not listed)

Airoha (not listed)

Hi-Silicon (not listed)

Vanchip (not listed)

Source: Company data, Credit Suisse

Foundry business model

helps alleviate the capex

burden of RF component

vendors

11 November 2015


Compound semiconductor still the preferred choice

for RF components

Silicon-based RF component solutions focusing on lower cost and higher integration

Silicon-based technologies such as bulk CMOS, silicon-on insulator (SOI), silicon-on

sapphire (SOS) have been tapping into the RF component market in the past few years.

Qualcomm and Peregrine are the most aggressive among players promoting the silicon-

based RF solutions. The manufacturing of these silicon-based RF components is usually

done by IDMs or silicon foundry players such as TSMC. The major value proposition from

the silicon-based RF component include:

■ Lower cost: by leveraging the lower silicon wafer cost and the 8" manufacturing

capacity (vs 6" for GaAs), silicon-based RF components could be offered at a 10-20%

lower cost compared to GaAs-based solutions.

■ Higher level of integration: as most semiconductors are silicon-based, such as

baseband processors, controllers, etc., manufacturing RF components with silicon-

based technologies could lead to higher level of integration with other components.

Compound semiconductor offering better performance

While SOI is now becoming the mainstream technology for switches in the RF front end

module, we believe that compound semiconductor (especially GaAs) will remain as the

mainstream technology for RF components (especially PA) due to its better performance

in handling higher power, high frequency signals. Advantages for GaAs power amplifiers

include:

■ Better power efficiency: with 5-10% power-added efficiency vs CMOS PAs, GaAs

PAs are more efficient in power consumption, which will lead to longer battery life for

mobile devices.

■ Higher breakdown voltage: with the higher gain and higher breakdown voltage,

GaAs PA can offer better reliability, especially in handling the high power signals.

■ Better linearity: GaAs PA offers better linearity, which means it could better preserve

the integrity of the complex modulation formats and thus the signal quality amid high

data rates.

■ High electron mobility: GaAs has the electron mobility six times higher than that for

silicon, making it capable of operating at high frequency of up to 300 GHz.

■ Lower noise: GaAs PA is less susceptible to noise, which could lead to better signal

quality.

■ Shorter development leadtime: As silicon-based PAs are less mature in technology

and need more circuit designs to compensate the inferior properties, it usually takes

higher development cost and longer leadtime.

■ Shorter production cycle: It usually takes 4-6 weeks of production cycle time for

GaAs-based PA vs 8-12 weeks for silicon-based PA.

GaAs PA performance still improving and seeing more integration

While silicon-based PA suppliers are improving the performance of their solutions, through

different technologies such as envelop tracking modulation, GaAs PAs are also marching

ahead to better address the increasingly stringent requirement in the wireless

communication standards. For example, Win Semi claims that it had come up with a new

generation of GaAs Heterojunction Bipolar Transistor (HBT) technology for PA every two

years in the past, with better overall efficiency.

Silicon-based PA solutions

offer lower cost and higher

levels of integration

GaAs PAs offer better

power consumption, linearity

and signal quality at higher

power, high frequency

environments

GaAs PA performance still

improving

11 November 2015


In addition, the trend of component integration is also ongoing in the GaAs space. For

example, vendors are adopting the Enhancement/Depletion Mode Pseudomorphic High

Electron Mobility Transistor (E/D pHEMT) technology to integrate the low-noise amplifier

(LNA), switch and logic control onto one die. Further integration of the PA, LNA, switch

and logic control could be achieved through the Bipolar High Electron Mobility Transistor

(BiHEMT) technology. We believe that higher levels of integration could reinforce the

competitiveness in offering lower costs and shorter design leadtime for its customers.

Figure 18: GaAs PA performance is still improving Figure 19: The integration trend in the RF front end

HBT pHEMT pHEMT CMOS

Discrete PA + LNA + Switch + Logic

HBT

PA + LNA + Switch + Logic

Integrated PA + LNA + Switch + Logic

BiHEMT

E/D pHEMT

Source: Win Semi, Credit Suisse Source: Win Semi, Credit Suisse

Silicon-based PA to see better opportunities in mature technologies

With superior characteristics, ever-evolving performance and more value added from

higher levels of integration, we expect GaAs to stay as the choice for PA in mid- to high-

end applications, while silicon-based PA might gain some traction in mature segments. As

illustrated in the chart below, we believe that GaAs-based solutions should take the

majority of the PA market for handsets, while silicon-based solutions should start to pick

up a bit more share in the 3G space as the technology is mature and more vendors are

prioritising costs over performance, similar to what we saw in the 2G handset market

several years ago. Despite the dominance of GaAs in the PA space, major players such

as Skyworks, Qorvo, Avago, Murata, etc., also have their internal silicon-based solutions

for the low-end segment.

Figure 20: GaAs PA to stay as the major solution for new technologies

Source: Win Semi, Credit Suisse

Integration also happening

in the GaAs space

GaAs PA to stay as the

mainstream, while silicon

PA to see more

opportunities in mature

technologies

11 November 2015


Win Semi and AWSC top two foundry players

34% of GaAs capacity owned by foundry players

We estimate that foundry players own around 34% of the compound semiconductor

capacity globally, while the remaining 66% of the capacity stays within IDM players in

2015. While some IDM players also leverage some of its capacity providing the foundry

service to maintain its utilisation, we think the business model is unsustainable given the

conflicts of interests and should be gradually phased out, leading to more outsourcing

orders shifted towards independent foundry players. Currently, Win Semi and AWSC are

the top two GaAs foundry players, collectively taking 97% of the foundry capacity share.

Due to the duopolistic nature of the compound semiconductor foundry space, we believe

that the competition should be relatively benign. In addition, RF component suppliers have

their own processes and it usually takes a few years to qualify a new foundry partner, so

we think the high switching costs should make RF component vendors less incentivised in

changing foundry partners frequently just to bargain for better prices.

Though there are some LED players, especially from China, planning to tap into the

compound semiconductor foundry sector by leveraging their existing knowledge in the

LED space, we think the high entry barriers will prevent them from being a competitive

player in the market at least in the next 3-5 years.

Figure 21: Global compound semiconductor capacity by

foundry/IDM

Figure 22: Global compound semiconductor foundry

capacity share (2015E)

25% 27% 31% 33% 34%

75% 73% 69% 67% 66%

0%

20%

40%

60%

80%

100%

2011 2012 2013 2014 2015E

Foundry IDM

Win Semi 63%

AWSC 34%

GCS 2% Others 1%

Source: Gartner, Credit Suisse estimates Source: Gartner, Credit Suisse estimates

Foundry players own 34% of

the compound

semiconductor capacity

Duopolistic structure,

increasing outsourcing and

high switching costs should

point to a relatively stable

competitive landscape

11 November 2015


LTE a two-fold driver for RF component growth The LTE migration is a driver for RF component demand in both device and infrastructure

fronts. We expect the 26% CAGR in LTE mobile device shipments (penetration into overall

cellular device up to 70% in 2017 from 33% in 2014) and the much higher RF content for

complex frequency bands in LTE models (US$6-15 vs ~US$3 for 3G) should support a

12% revenue CAGR in mobile applications over 2015-17. In addition, LTE subscriber net

additions expanding at a CAGR of 25% over 2015-17 could lead to more infrastructure

build, especially complementary smaller cell sites, driving a 5% CAGR in this space.

LTE devices with higher RF content to support growth

Smartphone shipment growth slowing down, though the LTE migration undergoing

Solid smartphone shipments growth and the higher RF content for smartphones have

been supporting the demand for RF components in the past few years. Credit Suisse

forecasts the global smartphone shipments growth to slow down to a 10% CAGR over

2015-17 vs 38% over 2012-14, as the smartphone penetration into the overall handset

market is expected to climb to 71% in 2015 and 78% in 2017 from 42% in 2012.

Despite the slower growth in smartphone shipments in the next few years, we believe the

migration into LTE smartphones should continue to play out for higher data transmission

rates, as the demand for more high-quality multimedia content and cloud storage is on a

rising trend. According to IDC, the global LTE subscriber net additions should see a 25%

CAGR over 2015-17 to reach a total subscriber base of 708 mn.

Looking at the global telecom service subscriber base, there could be only 7% of the users

migrating to the LTE service by the end of 2015, but the ratio is expected to increase to

18% by the end of 2018, as telecom service operators are aggressively promoting the LTE

service, with more comprehensive LTE infrastructure coverage, hoping to generate more

revenues from higher mobile data consumption.

Figure 23: Global smartphone shipments growth slowing

down

Figure 24: LTE subscriber new additions to see a 25%

CAGR over 2015-17

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2010 2011 2012 2013 2014 2015E 2016E 2017E

mn

units

Global smartphone shipment Growth (YoY) Penetration

0

50

100

150

200

250

2013 2014 2015E 2016E 2017E

subs

crib

ers

(mn)

Asia/Pacific CEMA Latin America North America Western Europe

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

Smartphone shipments

growth slowing down to 10%

over 2015-17 on high

penetration

LTE subscriber new

additions seeing a 25%

CAGR over 2015-17

18% LTE penetration into

the telecom services

subscriber base in 2018

11 November 2015


70% of the cellular device shipments to be LTE-enabled in 2017

As more subscribers are migrating into the LTE service, we believe that more LTE-

enabled mobile devices will be available in the market. According to Navian, around 50%

of the mobile device shipments (mostly handsets) are estimated to be LTE-enabled in

2015 and it forecasts the portion to further increase to 70% in 2017, representing a 29%

CAGR in the LTE-enabled device shipments over 2015-17 vs a 9% CAGR for overall

mobile devices.

Figure 25: Global mobile device shipments forecast Figure 26: Mobile device shipments breakdown by cellular

standard

1,600

1,700

1,800

1,900

2,000

2,100

2,200

2013 2014 2015E 2016E 2017E

mn

units

30%17% 12% 10% 8%

51%

50%

38%29%

22%

19%33%

50%61%

70%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015E 2016E 2017E

2G 3G 4G

Source: Navian, Credit Suisse Source: Navian, Credit Suisse

Much higher RF content for LTE models

We see higher RF content in LTE-enabled devices (US$6-15 vs ~US$3 for 3G and <US$1

for 2G), as they need more RF components to handle more complicated LTE frequency

bands. In addition, the fact that LTE frequency bands are much more fragmented and thus

more different bands in operations means that the incremental RF component demand per

device could be much more significant compared to the 2G to 3G transition.

Usually 2G/3G (GSM/CDMA/UMTS/TD-SCDMA) smartphones support frequency bands

include 850Hz/900Hz/1800Hz/1900Hz for GSM/EDGE, 800Hz/1700Hz/1900Hz/2100Hz

for CDMA EV-DO, 850Hz/900Hz/1700Hz/1900Hz/2100Hz for UMTS, and 1900Hz/2000Hz

for TD-SCDMA, while there are 44 different frequency bands for LTE (band 1-44). Other

things being equal, smartphones supporting more bands get better chances of

uninterrupted signal during global roaming, given that telecom operators are licensed to

provide services at different frequency bands. Moreover, there might be more networks of

different frequency bands in operation in the same region. In both scenarios, more RF

component will need to be incorporated into the device to avoid signal interruptions.

On average, a typical 2G handset will need two units of power amplifiers and a typical 3G

smartphone will need 3-4 units of power amplifiers. When it comes to an LTE smartphone,

it will take 4-6 units of power amplifiers, even under the multiband-multimode power

amplifier (MMPA) architecture, where one single power amplifier is responsible for

handling multiple bands.

Under the MMPA architecture, the quantity of power amplifiers might distort the magnitude

of the incremental demand for RF components from LTE smartphones and RF dollar

content should be a better indicator. According to Qorvo's estimate, a typical GSM/CDMA

smartphone only needs <US$1 worth of RF components, a typical UMTS/TD-SCDMA

smartphone needs ~US$3 worth of RF components, and a typical LTE smartphone needs

US$6 worth of RF components. Furthermore, it will take US$15+ worth of RF components

to make the LTE smartphone work under different frequency bands around the world.

70% of the mobile device

shipments will be LTE-

enabled in 2017 vs 50% in

2015

More frequency bands in

LTE devices represent more

RF opportunities

LTE has more bands than

the total of other existing

cellular air interfaces

LTE models need 4-6 PAs

vs 3-4 for 3G and two for 2G

US$6-15 RF dollar content

in LTE smartphones vs

~US$3 for UMTS and

<US$1 for GSM/CDMA

11 November 2015


Figure 27: Higher RF content for LTE models Figure 28: More power amplifiers needed in LTE models

0

2

4

6

8

10

12

14

16

18

2G 3G Typical LTE Global LTE

US

$

Filters Switches Powr amplifiers Others

2

3-4

4-6

2G 3G LTE

# of PA per handset

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

Smartphones increasingly supporting more LTE bands

Though currently there is no single smartphone model supporting all available LTE bands,

we are seeing the trend of more frequency bands being supported by one single

smartphone model. Looking at iPhone models over generations in the past few years, we

found that the number of LTE bands supported by the iPhone model has been increasing

from 2-5 bands for iPhone 5, 7-13 bands for iPhone 5S, 16-20 bands for iPhone 6 to 22-23

bands for the latest iPhone 6S.

While packing more RF content into smartphone models to fit into different markets around

the world could increase the bill of materials (BOM) costs, the savings from more efficient

supply chain management, inventory control and economies of scale in manufacturing

might provide incentive for smartphone brands to move in this direction.

Figure 29: more LTE bands supported by iPhone models over generations

iPhone 5

Launched

Model A1428

(GSM)A1429 (CDMA) A1429 (CDMA)

A1533

(GSM)A1533 (CDMA) A1453 A1457 A1530

A1549

(GSM)A1549 (CDMA) A1586 A1633 A1688

Total bands 11 16 11 20 24 26 15 18 25 29 35 38 37

2G/3G bands 9 11 8 9 13 13 8 8 9 13 15 15 15

LTE bands 2 5 3 11 11 13 7 10 16 16 20 23 22

1

2

3

4

5

7

8

12

13

17

18

19

20

25

26

27

28

29

30

38

39

40

41

iPhone 6s

Sep-15

iPhone 5S

Sep-12 Sep-13

iPhone 6

Sep-14


Carrier aggregation potentially leading to more bands in commercial operations

Carrier aggregation (CA) is the technology first defined in 3GPP Release 10 (1Q11) to

increase the bandwidth through aggregating multiple component carriers (i.e., smaller

frequency bands) for the LTE-Advanced standard. According to the definition, each

component carrier could have a bandwidth of 1.4, 3, 5, 10, 15 or 20 MHz and a maximum

of five component carriers can be aggregated to create the maximum bandwidth of 100

MHz. In addition to the higher data rates, carrier aggregation could be done through

component carriers in the same frequency band (both continuous carriers or non-

More and more LTE bands

being supported by

smartphone models

Carrier aggregation to help

increase the bandwidth and

thus the transmission speed

11 November 2015


continuous carriers) or in different frequency bands to help telecom operators better

leverage all the spectrum assets.

Figure 30: Carrier aggregation combining smaller carriers

to enhance peak data rates

Figure 31: Multiple ways of aggregating component

carriers

Source: Qualcomm, Credit Suisse Source: 3GPP, Credit Suisse

The implications to RF components from the carrier aggregation include: 1) potentially

more frequency bands in commercial operation. Before the carrier aggregation

technology, some telecom operators might have fragmented bandwidths that are too small

to support the data rate for the LTE service. Those smaller bands could now possibly work

coherently to be in commercial operation, potentially leading to more frequency bands for

mobile devices to support; 2) RF components of better quality needed. With more

smaller bands joining the commercial operations, RF components of better quality

(especially filters) might need to be adopted, as there are more signals of adjacent

frequencies that could cause interference.

According to Navian, LTE cellular devices that can support the carrier aggregation

technology will see significant growth in the next few years, with a 60% CAGR in

shipments in 2015-17 vs a 29% CAGR for overall LTE cellular devices. This will lead to

52% CA technology penetration in overall LTE cellular shipments in 2017 vs 34% in 2015.

Figure 32: 60% CAGR in CA-capable LTE cellular device

shipments in 2015-17

Figure 33: 52% of the LTE mobile device shipments will

be CA-capable in 2017

345517

669 740 814

346

600

885

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2013 2014 2015E 2016E 2017E

mn

units

Non-CA LTE devices CA-enabled LTE devices

100%

83%

66%55%

48%

17%

34%45%

52%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015E 2016E 2017E

Non-CA LTE devices CA-enabled LTE devices

Source: Navian, Credit Suisse Source: Navian, Credit Suisse

12% CAGR in RF revenues from mobile devices

Based on the assumption of mid single-digit annual global mobile device shipments,

increasing LTE penetration, higher RF content in LTE devices, we forecast the RF

component revenues to see a 12% CAGR over 2015-17 in the mobile application to reach

US$9.7 bn in 2017.

Implications from CA: (1)

potentially more bands in

commercial operations; (2)

RF components of better

quality needed

CA-capable LTE device

shipments to witness a 60%

CAGR in 2015-17 vs 29%

for overall LTE devices

12% CAGR for mobile RF

revenues over 2015-17

11 November 2015


Figure 34: 12% revenue CAGR from the mobile application over 2015-17

0

2,000

4,000

6,000

8,000

10,000

12,000

2013 2014 2015E 2016E 2017E

US

$mn

Source: Qorvo, Skyworks, Credit Suisse estimates

Increasing subscribers leading to more

infrastructure build

LTE macro cell site construction sustaining at similar pace in the next few years

As telecom operators around the world hope to expand the average revenues per user

(ARPU) by stimulating the mobile data consumption from subscribers through the LTE

service, the base station construction has been booming in the past few years amid the

increasing data traffic. According to 4G Americas' research, there were 360 commercially

launched LTE networks globally at the end of 4Q14, compared to only 200 networks at the

end of 2013 and 139 at the end of 2012. 4G Americas forecasts that the number of LTE

networks in commercial operation should further increase to 535 by end of 2017.

According to other research by IDC, there were 400+ thousand LTE macro cell sites

worldwide (each cell site could host multiple base stations) at the end of 2014 and the

base should increase to 700+ thousand sites by the end of 2017, at the pace of about 100

thousand sites annually in the next few years.

Figure 35: More LTE networks in commercial operations

globally

Figure 36: LTE macro cell site installed base to increase

steadily

2 1651

139

200

360

460500

535575

600

0

100

200

300

400

500

600

700

2009 2010 2011 2012 2013 2014 2015E 2016E 2017E 2018E 2019E

0

100

200

300

400

500

600

700

800

2013 2014 2015E 2016E 2017E

Tho

usan

d un

its

Source: GSA, Credit Suisse Source: IDC, Credit Suisse

Complementary smaller cell site deployment to follow up

Looking forward, while the incremental macro cell site construction might remain at a

similar pace over 2015-17, the deployment of smaller cell sites such as micro cells, pico

460 LTE networks in

commercial operations by

end of 2015 vs 331 by end

of 2014

LTE macro sites to increase

at an annual pace of 100

thousand sites over the next

few years

11 November 2015


cells and femto cells will likely catch up in the next few years to complement the macro cell

sites. These smaller cells operate at lower power voltages and narrower ranges (usually

less than 2 kilometres vs up to 35 kilometres for macro cell). They can be built at a lower

cost and are used to bridge the coverage gaps between macro cells or increase the

network capacity in high-density areas like coffee shops, stadiums, office buildings, etc.

IDC forecasts that smaller LTE cell site installed base could expand from ~600 thousand

units at the end of 2014 to ~1,700 thousand units by the end of 2017, with a 39% CAGR

for the installed base growth and a 35% CAGR in the annual new additions in 2015-17,

which will lead to more demand for RF components from the telecom infrastructure space.

Figure 37: Smaller cell site installed base to catch up Figure 38: Smaller cells to complement macro cells

0

100

200

300

400

500

600

700

800

2013 2014 2015E 2016E 2017E

Tho

usan

d un

its

Micro cell Pico cell Small cell

Source: IDC, Credit Suisse Source: 3GPP, Credit Suisse

5G to kick another round of infrastructure demand from 2020

While the LTE migration is still undergoing for the telecom service sector, standard-setting

organisations and related equipment players have started to define the next generation of

the air interface, or the 5G standard. The improvement is along the similar trajectory as in

the upgrades in the past few generations, including higher data rate, better traffic handling

capability, etc. Though the standard has not been finalised yet, some believe that it could

support 10-20Gbps downlink data rate vs up to 1.2Gbps for 4G LTE, handle 10,000x more

traffic simultaneously vs 4G LTE, and could operate at the >5GHz frequency spectrum vs

<5GHz spectrum currently used by exiting air interfaces.

While networking equipment suppliers including Nokia, Alcatel-Lucent, ZTE, Huawei, etc.,

have showcased their preliminary 5G technology, it is expected that 5G commercial

operation should be seen in 2020. Though this is still five years out, we believe that higher

rates, better traffic capacity and the new frequency bands are all positive development for

the RF component demand in the longer term.

Figure 39: Timeline of cellular generations

Source: RYSAVY Research

New additions in smaller

LTE cell sites to see a 35%

CAGR over 2015-17

5G to be the next cellular

standard beyond 2020,

positive to RF component

demand in the longer term

11 November 2015


5% CAGR for RF revenues from the infrastructure in 2015-17

We believe that LTE infrastructure growth in the next few years should be more

significantly from the complementary smaller cell sites, while the macro cell site installed

base should expand at a similar pace. As these complementary cell sites are usually of

smaller scales and thus consume less RF components each unit, we forecast a 5% RF

revenue CAGR from the infrastructure in 2015-17, lower than the high single-digit CAGR

seen in 2013-15.

Figure 40: RF revenues from infrastructure to see 5% CAGR in 2015-17

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2013 2014 2015E 2016E 2017E

US

$mn

Source: Qorvo, Credit Suisse estimates

11 November 2015


WiFi spec upgrade and IoT enlarging the TAM We see the growth opportunities for RF components in the WiFi market coming from (1)

the upgrade into the more advanced 802.11ac standard, which tends to adopt more RF

components for dual bands (2.4 GHz and 5.0 GHz) and more spatial streams; and (2) IoT

further expanding the total addressable market, with a 39% CAGR in connected device

shipments in 2015-20. We forecast a 22% RF revenue CAGR in the WiFi application.

WiFi spec upgrade positive to RF consumption

802.11ac for higher data rate and better signal quality

Similar to the cellular air interface, the transmission data rate has been one of the focuses

in the WiFi standard evolution. Built upon its predecessors including 802.11a/b/g/n,

802.11ac is the latest WiFi standard defined by IEEE LAN/MAN Standards Committee in

December 2013 and is now penetrating into more and more WiFi application, including

smartphones, tablets, NBs, routers, etc. It has several advantages over 802.11n.

■ Up to 160MHz bandwidth: As 802.11ac operates at 5.0GHz, where the overall

bandwidth is wider, it can offer a single channel bandwidth of up to 160MHz,

compared to only up to 40MHz channel bandwidth from 802.11n, given that the latter

operates at a much more crowded 2.4GHz frequency band. Wider bandwidth means

more data can be transmitted simultaneously.

■ Higher data load: 802.11ac adopts the more advanced 256 quadrature Amplitude

Modulation (QAM) vs 64 QAM for 802.11n or earlier version. Simply put, the 256 QAM

technology allows the system to load more data into each packet, which could lead to

33% higher data rates vs 64 QAM, with all other things being equal.

■ Up to eight spatial streams: 802.11ac supports up to 8 spatial streams vs up to 4

spatial streams for 802.11n. Theoretically, the number of spatial stream is in a linear

correlation with the total data transmission rate.

■ Multi-user support: 802.11ac can simultaneously communicate with different devices

through different spatial streams, while 802.11n can only communicate with one

device at any given time. This could avoid a long queue caused by big data traffic from

one single device.

■ Beam forming: 802.11ac supports the function to detect the position of the connected

device and strengthen the signal towards that direction to improve signal quality.

■ Less signal interference: As 5.0 GHz is a less crowded frequency band compared to

the 2.4 GHz band (signal from Bluetooth, remote control, etc., also work at similar

frequency bands), the signal interference is less of an issue for 802.11ac.

Figure 41: Evolution of the WiFi spec

802.11 802.11a 802.11b 802.11g 802.11n 802.11ac

Release Jun-97 Sep-99 Sep-99 Jun-03 Oct-09 Dec-13

Frequency band 2.4GHz 5.0GHz 2.4GHz 2.4GHz 2.4/5.0GHz 5.0GHz

Bandwidth (MHz) 22 20 22 20 20/40 20/40/80/160

Modulation 64 QAM 64 QAM 64 QAM 64 QAM 64 QAM 256 QAM

Stream data rate (Mbps) up to 2 up to 54 up to 11 up to 54 up to 150 up to 433.3/866.7

MIMO streams NA NA NA NA 4 8

Multi-user MIMO NA NA NA NA NA Yes

Beam forming NA NA NA NA NA Yes

Source: WiFi Alliance, Credit Suisse

802.11ac can offer higher

data rate and better signal

quality

11 November 2015


Both residential and commercial users are adopting 802.11ac

According to IDC, only 27% of the shipments in home routers, gateways and access

points (AP) supported the 802.11ac standard in 2014, but the portion is estimated to

increase to 34% in 2017, representing a 41% CAGR from 29mn units in 2014 to 81mn

units in 2017. Due to higher ASPs, 802.11ac-enabled home routers, gateways and APs

are forecast to take 65% of the revenue share in 2017, up from 53% in 2014.

A similar trend is expected to play out in the commercial application. IDC estimates that

61% of the shipment in enterprise wireless APs will support the 802.11ac standards in

2017 vs only 12% in 2014. This represents a 93% shipments CAGR and 65% revenue

CAGR over 2015-17.

Figure 42: Home router/gateway/AP migrating to 802.11ac Figure 43: Enterprise wireless AP also adopting 802.11ac

70% 70% 68% 66%

65%22%27%

31%

34%

34%

0

50

100

150

200

250

2013 2014 2015E 2016E 2017E

mn

units

802.11g 802.11n 802.11ac

41% shipment CAGR for 802.11ac in 2015-17

99%88%

70% 54%39%

1% 12% 46% 61%

0

5

10

15

20

25

30

2013 2014 2015E 2016E 2017Em

n un

its

802.11a/b/g/n 802.11ac

93% shipment CAGR for 802.11ac in 2015-17

Source: IDC, Credit Suisse Source: IDC, Credit Suisse

Smartphone WiFi also migrating into 802.11ac

For the smartphone function, HTC M7 was the first one to adopt the 802.11ac standard in

2013, followed by most of the flagship models from the Android camp in the same year

and Apple iPhone one year later. Starting from 2015, we are seeing more mid to low-end

smartphones adopting the 802.11ac standard, which we think is due to more

router/gateway/access points supporting the 802.11ac standard. We expect the

penetration of 802.11ac in mid- to low-end smartphones to continue rising in the next few

years.

Figure 44: More mid- to low-end smartphones will support 802.11ac WiFi

2013 2014 2015

High end models (>US$400)

802.11ac HTC M7, Samsung GS4,

Sony Xperia Z1, LG G2

Apple iPhone 6, HTC M8, Samsung

GS 5, Sony Xperia Z2/3, LG G3,

Xiaomi Mi4, OPPO Find 7,

ZTE Nubia Z7

Apple iPhone 6S, HTC M9, Samsung

GS6, Sony Xperia Z4/5, LG G4, LG

Nexus 5X, Huawei Nexus 6P, Huawei

Honor 7, ZTE Grand S3, LeTV One

802.11a/b/g/n Apple iPhone 5S, Huawei Ascend P6 Huawei Honor 6, Huawei Ascend P7 Huawei Mate S

Mid to low end models (<US$400)

802.11ac Xiaomi Mi3 Xiaomi Mi4 Xiaomi Redmi Note 2, Xiaomi Mi 4c,

ZTE Blade S6

802.11a/b/g/n HTC Desire 600,

Samsung Galaxy Grand 2

HTC Desire 820,

Samsung Galaxy A3, Sony Xperia C3

HTC Desire 816G, Samsung Galaxy

J7, Xiaomi Redmi 2, Sony Xperia C4,

LG G4c, TCL Idol 3


802.11ac consumes more RF components

802.11ac-enabled devices tend to adopt more RF components because of more frequency

band support and the more spatial streams. Despite the fact that 802.11ac only operates

at the 5.0 GHz frequency band, most devices also support the 2.4 GHz band to be

backward compatible with older generation 802.11a/b/g/n. These dual-band WiFi devices

usually need to double the RF content. In addition, as 802.11ac supports up to 8 spatial

802.11ac home

router/gateway/AP to see a

41% shipments CAGR over

2015-17

802.11ac enterprise

wireless AP to see a 93%

shipment CAGR over 2015-

17

Mid- to low-end

smartphones are also

migrating into 802.11ac

802.11ac devices need

more RF component to

support dual bands and

more spatial streams

11 November 2015


streams simultaneously, proportionally more RF components will be needed in these

devices.

Internet of Things (IoT) expanding the addressable

market for WiFi

39% CAGR in IoT device shipments CAGR over 2015-20

According to Gartner, global IoT unit shipments will be growing at a 39% CAGR over

2015-20 across various sectors including automotive, consumer, automation, energy,

manufacturing, government, etc. This will drive the IoT installed base to be boosted from

4.9 bn units in 2015 to over 25 bn units in 2020. As the concept of the IoT has been to

share information between devices, wireless (and wired, to a lesser extent) communication

has been the key part in connecting all these individual devices.

Figure 45: Worldwide IoT device installed base expanding

at 37% CAGR in 2015-20

Figure 46: Worldwide IoT device shipment growing at 39%

CAGR in 2015-20

0

5,000

10,000

15,000

20,000

25,000

30,000

2013 2014 2015E 2016E 2017E 2018E 2019E 2020E

mn

units


0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

2013 2014 2015E 2016E 2017E 2018E 2019E 2020E

mn

units


Source: Gartner, Credit Suisse estimates Source: Gartner, Credit Suisse estimates

WiFi to be the major wireless standard for IoT

While WiFi is not the only wireless connectivity solutions for IoT, we believe it is in a better

position for those devices that need higher data rate and over a longer distance. Other

wireless interfaces including Bluetooth and ZigBee are more suitable for smaller data

volume within a short distance of usually less than 30m at a lower power consumption.

Figure 47: Comparison of different wireless connectivity standards

Wireless standard WiFi Bluetooth ZigBee

Developing institution WiFi Alliance Bluetooth SIG ZigBee Alliance

International standard 802.11 802.15.1 802.15.4

Frequency band 2.4 GHz / 5.0 GHz 2.4 GHz 2.4 GHz

Data rate 866.7 Mbps 24 Mbps 250 kbps

Power consumption High Low Lowest

Signal range 250m 10m 10-30m


According to Gartner, global semiconductor revenues from IoT is forecast to grow at a

30% CAGR from US$12.4 bn in 2014 to US$43.5 bn in 2020. Among all the

semiconductor revenues, around 15-20% will be associated with the wireless connectivity

solutions and WiFi should stay as the major standard in the next few years, followed by

Bluetooth and cellular. We thus see the IoT as another secular opportunity for the RF

component demand.

IoT device unit shipments to

grow at a 39% CAGR over

2015-20

WiFi to be the major

wireless connectivity for IoT

15-20% of the IoT

semiconductor revenues are

associated with wireless

connectivity and around half

of that is WiFi

11 November 2015


Figure 48: 30% CAGR for IoT semi revenues in 2015-20 Figure 49: WiFi to be the major wireless standard for IoT

0

0

0

0

0

0

0

0

0

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

50,000

2013 2014 2015E 2016E 2017E 2018E 2019E 2020E

US

$mn

AP/FPGA Microcontroller Sensor

Bluetooth Cellular Wi-Fi

Other Wireless Wireline Growth (YoY)

32% 28% 25% 24% 26% 27% 28% 29%

10%13% 17% 19%

21% 22% 19% 19%

48% 49% 48% 46% 40% 37% 36% 35%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015E 2016E 2017E 2018E 2019E 2020E

Bluetooth Cellular Wi-Fi ZigBee Other wireless

Source: Gartner, Credit Suisse Source: Gartner, Credit Suisse

22% CAGR for WiFi RF revenues in 2015-17; 13% CAGR for overall RF revenues in

2015-17

Based on the assumption of a 10% CAGR in WiFi device shipments and the 802.11ac

penetration rising to 43% in 2017 from 27% in 2015, we expect the RF component

revenues from the WiFi application to see a 22% CAGR in 2015-17. Along with our

forecast in other major applications, we expect to see 13% overall RF component revenue

CAGR in 2015-17.

Figure 50: 10% CAGR for WiFi device shipments in 2015-17 Figure 51: 43% of WiFi devices will be 802.11ac in 2017

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

2013 2014 2015E 2016E 2017E

mn

units

18%27%

35%43%

97%82%

73%65%

57%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2013 2014 2015E 2016E 2017E

802.11ac 802.11a/b/g/n

Source: Richwave, Credit Suisse estimates Source: Skyworks, Credit Suisse estimates

Figure 52: RF revenues from WiFi to see 22% CAGR in

2015-17

Figure 53: Overall RF component revenues to see 13%

CAGR in 2015-17

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

2013 2014 2015E 2016E 2017E

US

$mn

0%

5%

10%

15%

20%

25%

30%

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

2013 2014 2015E 2016E 2017E

US

$mn

Mobile WiFi Infrastructure Others Growth (YoY)

Source: Richwave, Skyworks, Credit Suisse estimates Source: Qorvo, Skyworks, Richwave, Credit Suisse estimates

11 November 2015


Initiating coverage on the Taiwan Compound Semiconductor Sector; prefer Win Semi and VPEC We initiate coverage on the compound semiconductor sector in Taiwan, as we expect the

sector to benefit from the increasing RF component demand driven by (1) the LTE

migration leading to higher RF content and more infrastructure deployment, and (2) the

WiFi spec upgrade for higher data rates and IoT expanding the total addressable market.

We expect the competitive landscape to be relatively stable for the compound

semiconductor foundry sector given the duopolistic structure, increasing outsourcing trend

and high switching costs. We prefer Win Semi and VPEC in the Taiwan compound

semiconductor space.

Win Semi (3105.TWO, OUTPERFORM, TP NT$62)

We initiate coverage on Win Semi with an OUTPERFORM rating and a TP of NT$62,

based on 2016E P/E of 12x, which is the average of its historical trading range of 6-20x.

As we expect the RF component revenues to continue growing, driven by smartphone RF

content growth and infrastructure build amid the LTE migration, the WiFi spec upgrade

and the connectivity demand from IoT devices, we think Win Semi is well positioned to

benefit from the trend with its strong customer relationship, technology independence and

margin expansions on favourable product mix.

VPEC (2455.TW, OUTPERFORM, TP NT$47)

We initiate coverage on VPEC with an OUTPERFORM rating and a TP of NT$47, based

on 2016E P/E of 18x, which is the average of its historical trading range of 4-30x. We see

VPEC as a good proxy for the RF growth trend with its diversified customer base, as it

supplies epiwafers to both IDM and foundry players. In addition, the expansion into the

optical space could lead to better margins and the superior profitability/strong balance

sheet should sustain the high dividend yield of 7-8% over 2015-17.

AWSC (8086.TWO, NEUTRAL, TP NT$88)

We initiate coverage on AWSC with a NEUTRAL rating and a target price of NT$88, which

is based on 2016E P/E of 11x, which is at a discount vs its peer Win Semi, as we factor in

the potential single client risk, due to the lack of a stable P/E history. We believe AWSC is

a quality name operating in a duopolistic competitive landscape, though the high reliance

on Skyworks could be a mixed blessing, as the benefit of the close relationship could be

coupled with higher business volatility. Although the new business and new customer

addition could help the outlook into 2016, we think shares are fairly valued at this level.

LTE migration, WiFi spec

upgrade and IoT should be

the major drivers for RF

component demand

Strong customer

relationship, technology

independence and margin

expansions on a favourable

product mix

Diversified customer base,

expansion into the optical

communication and the

sustaining high dividend

yield

A quality name in a

duopolistic competitive

landscape, though fairly

valued

11 November 2015


Asia Pacific / Taiwan

Win Semiconductors Corp

(3105.TWO / 3105 TT) INITIATION

Well positioned for the RF demand growth

■ Initiating coverage with OUTPERFORM. We initiate coverage on Win

Semi with an OUTPERFORM rating and a TP of NT$62, based on 12x

2016E EPS. As we expect the RF component revenues to continue growing,

driven by smartphone RF content growth and infrastructure build amid the

LTE migration, the WiFi spec upgrade and the connectivity demand from IoT

devices, we think Win Semi is well positioned to benefit from the trend with

its strong customer relationship, technology independence and margin

expansions on favourable product mix.

■ Strong customer relationship with major players. Win Semi has been

working leading RF component vendors including Avago, Skyworks, Murata,

MACOM, Analog Device, RDA, etc., to provide the foundry service. We

expect the relationship to sustain due to limited capacity expansions among

IDMs, long qualification time for new foundry partners and the duopolistic

structure in the GaAs foundry space.

■ Technology independence adding more value. Win Semi has in-house

technologies for products operating at 100MHz-100GHz in a wide range of

applications including cellular, WiFi, base station, satellite TV, cable TV, auto

radar, etc. Though it takes close cooperation between RF vendor and the

foundry partner to work on the products, we believe the technology

independence could add more value and further strengthen the relationship.

■ Margin expansion on operating leverage and product mix shift. We

forecast the operating margin to expand from 28.4% in 2015 to 29.1% in

2017, driven by its more favourable product mix and operating leverage.

While mobile should remain as the biggest application for Win Semi, we

expect the contribution from higher-margin WiFi and infrastructure to

gradually increase from 53% in 2015 to 56% in 2017 for a better product mix.

■ Valuation. Our TP of NT$62 is based on 12x 2016E EPS. Historically, its

shares have been trading at 6-20x 12M forward P/E, with the mid-cycle

average at 12x. As we forecast a 13% OP CAGR over 2015-17, similar to

2013-15, we think the mid cycle average is a reasonable target for valuation. Share price performance

80

130

180

20

30

40

50

60

Nov-13 Mar-14 Jul-14 Nov-14 Mar-15 Jul-15

Price (LHS) Rebased Rel (RHS)

The price relative chart measures performance against the

TAIWAN SE WEIGHTED INDEX which closed at 8536.9 on

10/11/15

On 10/11/15 the spot exchange rate was NT$32.7/US$1

Performance over 1M 3M 12M Absolute (%) 21.7 86.1 85.5 — Relative (%) 20.6 83.1 91.1 —

Financial and valuation metrics

Year 12/14A 12/15E 12/16E 12/17E Revenue (NT$ mn) 9,910.0 11,907.4 13,186.8 14,839.5 EBITDA (NT$ mn) 4,196.5 5,323.9 6,115.8 6,948.1 EBIT (NT$ mn) 2,314.7 3,375.3 3,789.0 4,321.9 Net profit (NT$ mn) 1,963.5 2,693.2 2,998.0 3,425.8 EPS (CS adj.) (NT$) 3.31 4.53 5.03 5.75 Change from previous EPS (%) n.a. Consensus EPS (NT$) n.a. 4.55 5.03 5.28 EPS growth (%) 9.7 36.8 11.1 14.3 P/E (x) 14.6 10.7 9.6 8.4 Dividend yield (%) 3.1 0.4 5.6 6.2 EV/EBITDA (x) 6.8 5.4 4.5 3.8 P/B (x) 1.8 1.7 1.6 1.5 ROE (%) 12.7 16.4 17.0 17.9 Net debt/equity (%) Net cash Net cash Net cash Net cash

Source: Company data, Thomson Reuters, Credit Suisse estimates.

Rating OUTPERFORM Price (10 Nov 15, NT$) 48.50 Target price (NT$) 62.00¹ Upside/downside (%) 27.8 Mkt cap (NT$ mn) 29,537.7 (US$ 903.4) Enterprise value (NT$ mn) 29,001 Number of shares (mn) 609.02 Free float (%) 82.9 52-week price range 48.5 - 23.9 ADTO - 6M (US$ mn) 16.2

*Stock ratings are relative to the coverage universe in each

analyst's or each team's respective sector.

¹Target price is for 12 months.

Research Analysts

Derrick Yang

886 2 2715 6367

[email protected]

Jerry Su

886 2 2715 6361

[email protected]

11 November 2015


Win Semiconductors Corp 3105.TWO / 3105 TT Price (09 Nov 15): NT$47.00, Rating:: OUTPERFORM, Target Price: NT$62.00, Analyst: Derrick Yang

Target price scenario

Scenario TP %Up/Dwn Assumptions

Upside 81.00 72.34 16X P/E Central Case 62.00 31.91 12X P/E Downside 35.00 (25.53) 7X P/E

Key earnings drivers 12/14A 12/15E 12/16E 12/17E

Mobile 5,087 5,653 6,021 6,539 WiFi — — — — Infrastructure 2,948 4,203 4,812 5,442 1,875 2,051 2,353 2,858 — — — —

Income statement (NT$ mn) 12/14A 12/15E 12/16E 12/17E

Sales revenue 9,910 11,907 13,187 14,839 Cost of goods sold 6,400 7,227 7,983 8,968 SG&A 633.1 699.2 765.3 872.4 Other operating exp./(inc.) (1,320) (1,342) (1,677) (1,949) EBITDA 4,196 5,324 6,116 6,948 Depreciation & amortisation 1,882 1,949 2,327 2,626 EBIT 2,315 3,375 3,789 4,322 Net interest expense/(inc.) 31.1 5.8 15.0 7.9 Non-operating inc./(exp.) 143.8 45.2 (1.6) (1.6) Associates/JV 1.6 1.6 1.6 1.6 Recurring PBT 2,429 3,416 3,774 4,314 Exceptionals/extraordinaries — — — — Taxes 465.5 723.1 776.0 888.2 Profit after tax 1,963 2,693 2,998 3,426 Other after tax income — — — — Minority interests — — — — Preferred dividends — — — — Reported net profit 1,963 2,693 2,998 3,426 Analyst adjustments — — — — Net profit (Credit Suisse) 1,963 2,693 2,998 3,426

Cash flow (NT$ mn) 12/14A 12/15E 12/16E 12/17E

EBIT 2,315 3,375 3,789 4,322 Net interest — — — — Tax paid — — — — Working capital (52.5) 189.9 216.3 (231.3) Other cash & non-cash items 1,536 1,299 1,536 1,730 Operating cash flow 3,798 4,864 5,541 5,821 Capex (738) (3,279) (2,500) (2,500) Free cash flow to the firm 3,060 1,585 3,041 3,321 Disposals of fixed assets — — — — Acquisitions — — — — Divestments — — — — Associate investments — — — — Other investment/(outflows) (418.1) (507.6) 5.5 (12.6) Investing cash flow (1,156) (3,787) (2,495) (2,513) Equity raised 194 (1,618) — — Dividends paid (1,110) (119) (1,616) (1,799) Net borrowings (764.8) (2.5) (10.2) 24.7 Other financing cash flow — — — — Financing cash flow (1,680) (1,739) (1,626) (1,774) Total cash flow 962 (662) 1,421 1,534 Adjustments — — — — Net change in cash 962 (662) 1,421 1,534

Balance sheet (NT$ mn) 12/14A 12/15E 12/16E 12/17E

Cash & cash equivalents 4,676 4,014 5,434 6,968 Current receivables 690 991 1,069 1,229 Inventories 1,500 1,850 1,776 2,298 Other current assets 259.0 235.1 253.6 291.6 Current assets 7,125 7,090 8,533 10,788 Property, plant & equip. 11,653 12,534 12,732 12,632 Investments 2,694 3,001 3,001 3,001 Intangibles 54.4 51.0 51.0 51.0 Other non-current assets 290.2 910.6 879.6 867.0 Total assets 21,816 23,587 25,198 27,339 Accounts payable 930 1,089 1,177 1,353 Short-term debt 545.4 697.2 695.2 700.2 Current provisions — — — — Other current liabilities 1,273 1,931 2,082 2,395 Current liabilities 2,749 3,718 3,954 4,449 Long-term debt 2,938 2,780 2,771 2,791 Non-current provisions — — — — Other non-current liab. 189.3 193.7 193.7 193.7 Total liabilities 5,876 6,691 6,920 7,433 Shareholders' equity 15,940 16,896 18,278 19,905 Minority interests — — — — Total liabilities & equity 21,816 23,587 25,198 27,339

Per share data 12/14A 12/15E 12/16E 12/17E

Shares (wtd avg.) (mn) 592.7 594.4 595.5 595.5 EPS (Credit Suisse) (NT$)

3.31 4.53 5.03 5.75 DPS (NT$) 1.50 0.20 2.71 3.02 BVPS (NT$) 26.9 28.4 30.7 33.4 Operating CFPS (NT$) 6.4 8.2 9.3 9.8

Key ratios and valuation

12/14A 12/15E 12/16E 12/17E

Growth(%) Sales revenue (5.5) 20.2 10.7 12.5 EBIT 9.7 45.8 12.3 14.1 Net profit 8.4 37.2 11.3 14.3 EPS 9.7 36.8 11.1 14.3 Margins (%)

EBITDA 42.3 44.7 46.4 46.8 EBIT 23.4 28.3 28.7 29.1 Pre-tax profit 24.5 28.7 28.6 29.1 Net profit 19.8 22.6 22.7 23.1 Valuation metrics (x) P/E 14.2 10.4 9.3 8.2 P/B 1.75 1.65 1.53 1.41 Dividend yield (%) 3.18 0.43 5.77 6.43 P/CF 7.33 5.74 5.05 4.81 EV/sales 2.77 2.36 2.02 1.69 EV/EBITDA 6.54 5.28 4.36 3.62 EV/EBIT 11.9 8.3 7.0 5.8 ROE analysis (%) ROE 12.7 16.4 17.0 17.9 ROIC 12.4 17.1 18.4 21.0 Asset turnover (x) 0.45 0.50 0.52 0.54 Interest burden (x) 1.05 1.01 1.00 1.00 Tax burden (x) 0.81 0.79 0.79 0.79 Financial leverage (x) 1.37 1.40 1.38 1.37 Credit ratios Net debt/equity (%) (7.5) (3.2) (10.8) (17.5) Net debt/EBITDA (x) (0.28) (0.10) (0.32) (0.50) Interest cover (x) 74 582 253 550


11 November 2015


Well positioned for the RF demand growth We initiate coverage on Win Semi with an OUTPERFORM rating and a TP of NT$62,

based on 12x 2016E EPS. As we expect the RF component revenues to continue growing,

driven by smartphone RF content growth and infrastructure build amid the LTE migration,

the WiFi spec upgrade and the connectivity demand from IoT devices, we think Win Semi

is well positioned to benefit from the trend with its strong customer relationship, technology

independence and margin expansions on a favourable product mix.

Strong customer relationship with major players

Win Semi has been working leading RF component vendors including Avago, Skyworks,

Murata, MACOM, Analog Device, RDA, etc., to provide the foundry service. The top three

customers (Avago, Skyworks and Murata) each account for around 20% of Win Semi's

total revenues, while fabless players in Taiwan and China collectively account for around

15% of its total revenues. Other smaller customers collectively account for around 25% of

its total revenues.

As RF components are mostly customised products with different designs and processes

from different vendors, foundry players need to co-work with their customers in the product

development, which leads to a long qualification of a few years if a RF vendor would like to

add a new foundry partner. This, along with the duopolistic structure of the GaAs foundry

space, makes us believe that Win Semi's strong customer relationship should sustain into

the future.

Figure 54: Win Semi—63% share in the GaAs foundry

sector (2015E)

Figure 55: Win Semi—top three customers accounting for

~60% of total revenues (2015E)

Win Semi 63%

AWSC 34%

GCS 2% Others 1%

Avago~20%

Skyworks~20%

Murata~20%

CH/TW fabless~15%

Others~25%


Technology independence adding more value

Win Semi has in-house technologies for products operating at 100MHz-100GHz in a wide

range of application including cellular, WiFi, base station, satellite TV, cable TV, auto radar,

etc. Though it takes close cooperation between RF vendor and the foundry partner to work

on the products, the technology independence at Win Semi could add more value to its

customers and further strengthen the relationship.

On top of the discrete RF components, Win Semi could also integrate the manufacturing

process of different components onto one chip. For example, by adopting the E/D pHEMT

technology, Win Semi could integrate the LNA, switch and controller logic into one chip.

Furthermore, the BiHEMT technology could be used to integrated PA, LNA, switch and

controller logic all onto one chip. We think high levels of integration could help on the cost

structure and also fit into the trend of RF component modularisation. In addition, this could

Well positioned to benefit

from the trend with its strong

customer relationship,

technology independence

and margin expansions on

favourable product mix

Top 3 customers Avago,

Skyworks and Murata each

has ~20% of Win Semi's

revenues

In-house technology for

various application ranging

from 100MHZ to 100GHz

More integration through

new technology

11 November 2015


help them regain some shares in the components where silicon solution have higher

penetration (e.g., switch).

Figure 56: Win Semi—broad portfolio of technologies Figure 57: Higher level of integration for GaAs-based RF

components

HBT pHEMT pHEMT CMOS

Discrete PA + LNA + Switch + Logic

HBT

PA + LNA + Switch + Logic

Integrated PA + LNA + Switch + Logic

E/D pHEMT

BiHEMT

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

Margin expansions on operating leverage and

product mix shift

We forecast Win Semi's operating margin to expand from 28.3% in 2015 to 29.1% in 2017,

driven by its more favourable product mix and operating leverage. While mobile

application should remain as the biggest application for Win Semi in the next few years,

we expect the contribution from WiFi and infrastructure applications to gradually increase

from 53% in 2015 to 56% in 2017 on 802.11ac spec upgrade, IoT device proliferation and

share gains in the infrastructure space. As the WiFi and infrastructure products carry

higher margins, we expect the product mix shift to be positive to its overall margin.

Figure 58: Win Semi—mobile accounting for 45-50% of

total revenues

Figure 59: Win Semi—product mix turning slightly more

favourable

55% 54% 51% 47% 46% 44%

30% 31%30% 35% 36% 37%

15% 15% 19% 17% 18% 19%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2012 2013 2014 2015E 2016E 2017E

Mobile WiFi Infrastructure

45% 46%49%

53% 54% 56%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0%

10%

20%

30%

40%

50%

60%

70%

2012 2013 2014 2015E 2016E 2017E

Revenue contribution from infrastructure and WiFI (LHS) Gross margin (RHS)


Financial analysis

3Q15 GM expanded on favourable product mix; high single-digit QoQ for 4Q15 sales

Win Semi's 3Q15 revenues were down 6% QoQ and down 5% YoY, on channel inventory

digestion and end customer's model transition. 3Q15 gross margin expanded by 1.7 pp to

40.4%, reaching an all-time high due to the favourable product mix and NTD depreciation.

Despite the revenue decline, its operating expenses grew 17% QoQ to NT$350 mn on

higher R&D expenses and more employee bonus.

56% of revenues from

higher-margin WiFi and

infrastructure applications in

2017 vs 53% in 2015

11 November 2015


For 4Q15, management guides for the revenues to be up by high single digit QoQ on

restocking demand and the continuous WiFi spec upgrade trend. It also guides for the

gross margin to be similar to 3Q15. We model for 4Q15 revenues to be up 6% QoQ and

14% YoY, and the gross margin to be 40.1%.

WiFi and infrastructure to outperform mobile in 2016

We model for 11% top line growth to NT$13,187 mn, driven by the LTE migration, WiFi

spec upgrade, IoT device proliferation and share gains in the infrastructure space. We

expect the gross margin to improve slightly to 39.5% in 2016 vs 39.3% in 2015, reflecting

the shift to higher margin business. With the operating leverage, we expect the operating

expense ratio to trend down to 10.7% in 2016 vs 11.0% in 2015, putting the operating

margin at 28.7% in 2016 vs 28.3% in 2015.

We expect the mobile application revenue to see high single-digit YoY growth in 2016, as

its major customer focuses on the high-end smartphone segment and its fabless

customers could see a slower growth in China on its higher-than-average LTE device

penetration. We expect the WiFi and infrastructure revenues to see double-digit YoY

growth as the WiFi spec upgrade to the 802.11ac standard will play out, IoT devices will

enlarge the total addressable market and it will continue to gain shares in the infrastructure

space on Qorvo phasing out the foundry business.

Figure 60: Win Semi—quarterly revenues and margins Figure 61: Win Semi—margin moving along with

utilisation

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

1Q12 3Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16E 3Q16E

NT

$mn

Revenues Gross margin Operating margin

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0%

20%

40%

60%

80%

100%

120%

1Q12 3Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16E 3Q16E

Utilization (LHS) Gross margin


Figure 62: Win Semi—CS estimate vs consensus

(NT$ mn) 4Q15E 1Q16E 2015E 2016E

CS estimate Consensus CS estimate Consensus CS estimate Consensus CS estimate Consensus

Revenue 3,089 3,117 2,678 2,920 11,907 11,946 13,187 12,901

Sequential growth 6% 7% -13% -6% 20% 21% 11% 8%

Gross profits 1,239 1,255 1,017 1,123 4,681 4,698 5,204 5,064

Operating profits 872 675 691 637 3,375 3,416 3,789 3,621

Net profits 733 753 566 666 2,693 2,710 2,998 2,998

EPS (NT$) 1.23 1.27 0.95 1.12 4.53 4.56 5.03 5.03

GM 40.1% 40.3% 38.0% 38.5% 39.3% 39.3% 39.5% 39.3%

OPM 28.2% 21.7% 25.8% 21.8% 28.3% 28.6% 28.7% 28.1%

NM 23.7% 24.2% 21.1% 22.8% 22.6% 22.7% 22.7% 23.2%

Source: Company data, the BLOOMBERG PROFESSIONAL™ service, Credit Suisse estimates

Cash flow and balance sheet analysis

We estimate that Win Semi will generate NT$2.2-4.2 bn free cash flows, or NT$3.8-7.1

free cash flow per share over 2015-17, reflecting a free cash yield of 7.7-14.6%. The

improving free cash flow yield is reflecting the growth in earnings, partially offset by the

increasing capex. Capex is expected to be over NT$3 bn due to the construction of a new

fab for additional capacity of 3,000 wafers/months ramped in 2H16, and should go down to

11% top line growth in 2016,

with WiFi and infrastructure

to outperform

NT$3.8-7.1 free cash flows

per share in 2015-17

11 November 2015


NT$2.5 bn in 2016-17, implying 17-19% capex-to-sales ratio. We estimate the

depreciation-to-sales ratio to be around 15-20% in 2015-17, similar to the level in 2013-14,

but higher than the low teens level seen in 2011-12, due to the capex increase in the past

few years.

Given that Win Semi has been able to generate free cash flow despite the increasing

capex, we assume the dividend payout ratio to be 60% in 2015-17, similar to the level in

2012-14 (if the 20% capital reduction and thus NT$2 cash returned to shareholders was

included in the 2014 payout). Based on our earnings forecast and dividend payout

assumption, we expect to see 5.6-7.1% dividend yield in 2015-17.

Figure 63: Win Semi—cash flow summary (2011-17E)

(NT$ mn) 2011 2012 2013 2014 2015E 2016E 2017E

Revenues 8,901 11,238 10,481 9,910 11,907 13,187 14,839

Capex 3,336 3,317 2,815 738 3,279 2,500 2,500

Capex / Revenues (%) 37.5 29.5 26.9 7.4 27.5 19.0 16.8

Dep and Amort 1,034 1,307 1,822 1,882 1,949 2,327 2,626

Dep / Revenues (%) 11.6 11.6 17.4 19.0 16.4 17.6 17.7

Operating cash flow 1,736 2,233 5,490 3,798 4,864 5,541 5,821

Free Cash flow -1,136 -335 2,813 3,406 2,234 3,832 4,217

FCF per share -2.3 -0.6 4.7 5.7 3.8 6.4 7.1

FCF yield (%) -7.5 -1.9 15.4 16.8 7.7 13.3 14.6

Dividend per share 0.8 1.5 1.5 0.2 2.7 3.0 3.5

Dividend yield (%) 2.7 4.4 4.9 0.6 5.6 6.2 7.1

Source: Company data, Credit Suisse estimates

Initiating coverage with OUTPERFORM and NT$62 TP

We initiate coverage on Win Semi with an OUTPERFORM rating and a target price of

NT$62, based on 12x 2016E EPS. Historically, its shares have been trading at 6-20x 12M

forward P/E, with the mid-cycle average at 12x. As we forecast a 13% operating profit

CAGR in 2015-17, similar to 2012-15, we think the mid-cycle average is a reasonable

target for valuation.

Downside risks

Downside risks to our positive view on Win Semi would include:

■ Slower shipments growth in the LTE-enabled devices: As mobile devices are the

largest application for Win Semi for 45-50% of its total revenues and LTE smartphones

have a much higher RF content (US$6-15 vs ~US$3 for 3G), the weaker shipments

growth in LTE smartphone will impact Win Semi's revenue growth.

■ Lower capex among telecom operators for LTE infrastructure due to macro

uncertainties. Infrastructure accounted for around 15-20% of revenues for Win Semi

and orders in this application usually carry higher margins, so the slower base station

and complementary cell site construction could pose some risks to Win Semi's

business outlook.

■ Weaker WiFi demand on delayed spec upgrade to 802.1ac and softer IoT device

proliferation. WiFi accounted for 30-35% of Win Semi's total revenues and the portion

should further increase on the spec upgrade and IoT demand, so slower WiFi demand

will be a negative to Win Semi.

11 November 2015


Figure 64: Win Semi—12-month forward P/E Figure 65: Win Semi—12-month forward P/B

5.0

7.0

9.0

11.0

13.0

15.0

17.0

19.0

21.0

Dec-1

1

Apr-

12

Aug-1

2

Dec-1

2

Apr-

13

Aug-1

3

Dec-1

3

Apr-

14

Aug-1

4

Dec-1

4

Apr-

15

Aug-1

5

Average = 12

+1 Std dev = 14

-1 Std dev = 9

+2 Std dev = 17

-2 Std dev = 6

Win Semi PE (x)

0.5

0.7

0.9

1.1

1.3

1.5

1.7

1.9

2.1

2.3

2.5

Dec-1

1

Apr-

12

Aug-1

2

Dec-1

2

Apr-

13

Aug-1

3

Dec-1

3

Apr-

14

Aug-1

4

Dec-1

4

Apr-

15

Aug-1

5

Average = 1.52

+1 Std dev = 1.8

-1 Std dev = 1.3

+2 Std dev = 2.0

-2 Std dev = 1.0

Win Semi PB (x)


CS HOLT® indicates a similar valuation

We use CS HOLT, a CS valuation tool that derives the stock price based on a company’s

cash flow return on investment (CFROI®) and estimated asset growth rates. Based on our

modelled assumptions, CS HOLT would value Win Semi at NT$63.55, similar to our target

price.

Figure 66: Relative wealth chart and HOLT-derived valuation

Source: Credit Suisse HOLT®

11 November 2015


Company profile

Company background

Founded in October 1999, Win Semi has been focusing on manufacturing compound

semiconductor wafers and was the first independent 6” GaAs foundry in the world. Win

Semi provides the manufacturing service of the RF components, mostly PA, switch, and

LNA, to its customers for various end applications, including handset, tablet, NB,

consumer electronics, automobile, telecom base station, satellite, cable TV, etc. Win Semi

currently has two fabs in operation in Taiwan with a total capacity of 24,000 GaAs wafers

per month and should have another fab enter into mass production in 2016. Win Semi was

listed on the Emerging Board in October 2009 and on the Taipei Exchange (former OTC)

in December 2011.

QFII holdings and major shareholders

Win Semi has a fairly diversified shareholder structure, as top ten shareholders only own

25.2% of the company. Management owns 7-8% of the company, while Inventec owns 2%

and Inventec Chairman Yeh’s family owns ~9% in total. The QFII holdings at Win Semi

has been rising from ~4% in April 2014 to 31.7% currently.

Figure 67: Win Semi—QFII holdings vs share prices Figure 68: Win Semi—major shareholders

0

10

20

30

40

50

60

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

Dec-11 Apr-12 Aug-12 Dec-12 Apr-13 Aug-13 Dec-13 Apr-14 Aug-14 Dec-14 Apr-15 Aug-15

NT

$

QFII Holdings Share price (NT$, RHS)

Shareholder Stake

1 Fubon Life Insurance Co., Ltd. 4.4%

2 Yeh Kuo I 4.0%

3 Ch'en Chin Ts'ai 3.0%

4 New Labor Pension Fund 2.6%

5 Nan Shan Life Insurance Co., Ltd. 2.4%

6 HSBC & Rebeco Capital Growth Fund Invest 2.2%

7 Tian Ho Enterprise Co., Ltd. 2.0%

8 Inventec Corporation 2.0%

9 Yeh Li Ch'eng 1.9%

10 Hsieh Shih Ch'uan-b 0.9%

Top 10 25.2% Source: Company data, TEJ, Credit Suisse Source: Company data, TEJ, Credit Suisse

Figure 69: Win Semi—management profile

Title Name Experience Education

Chairman Dennis Chen General Manager as Namchow Chemical

Master of Public Administration from the University of San

Francisco

Master of Accounting from Tamkang University

CEO Yuchi Wang Research staff at Bell Lab Assistant Professor at Yuan Ze University

Ph. D. of Material Engineering from Rutgers University

Business Unit GM William Chang GM at Huga Opto Ph. D. of Chemical Engineering from Clemson University

Senior VP Changhwang Hua Engineer Director at Skyworks Ph. D. of Material Science from Stanford University

VP Joseph Liu Senior Principal Staff of Lockheed Martin Ph. D. from Pennsylvania State University

VP Brian Lee Operation Manager at UMC Masters from the University Southern California

Senior VP Steve Chen VP of Operation Service Center at Win Semi MBA from Rutgers University

Senior VP Kyle Chen Fab Director at Macronix President of Yokogawa Taiwan

EMBA from National Taiwan University

AVP Annie Yu Supervisor at Huga Opto Master in Law from Cornell University

AVP SY Wang Manager at Macronix Master of Industrial Engineering from National Tsing Hua

University

Director of Finance Joe Tsen Business Manager at CIBC (Canada) Master of Finance from Baruch College of City University

of New York


24,000/month GaAs

capacity

11 November 2015


Figure 70: Win Semi—key milestones

Time Milestone

Oct-1999 Founded in Taiwan

Dec-1999 Ground Breaking Ceremony for Corporate Offices & Fab A

Jul-2000 Facility construction complete

Sep-2000 Process equipment installation complete

Sep-2000 Process qualification initiated for HBT

Sep-2000 Preliminary Design Kit Available

Nov-2000 Completes production of Asia's first 6" GaAs HBT MMIC wafer

Nov-2000 Completes production of Asia's first 6",GaAs pHEMT wafer

Mar-2001 Presents the world's first 6",0.15μm GaAs pHEMT MMIC wafer

Apr-2001 Fab A begins production-level Foundry Operations

May-2001 Produced the worlds’ first 6" 50μm thick pHEMT MMIC wafer

Jun-2001 Presents 6"GaAs 1μm HBT MMIC wafer

Nov-2001 Began foundry production for 0.15μm pHEMT

May-2002 World’s first 0.15um mHEMT wafer on 6” GaAs

May-2002 Signed Strategic Alliance and Wafer Supply Agreements with Raytheon RF Components

Jun-2002 Production release of 2um HBT

Jun-2002 Demonstrated world’s first 25-um thick, 6-inch pHEMT wafer

Jun-2002 First 0.5um switch pHEMT wafer for customer

Oct-2002 Foundry production of 0.5-um power pHEMT started

Nov-2002 Production release of 1-um HBT announced

Nov-2002 Pilot run release 50-um thick 0.15 um power pHEMT announced

Jan-2003 Foundry production of customer-specific 3um HBT started.

Feb-2003 First 0.15um mHEMT wafer for customer

Dec-2003 Shipped more than 1.5 million WLAN PAs in a single month

Feb-2004 Baseline release of 2um High Ruggedness HBT

Sep-2004 Pilot run of 0.5 um 35V pHEMT for CATV

May-2005 Successfully developed HBT process for high efficiency, high performance 2G/3G cellphone application

Apr-2007 Acquired additional Huaya land and factory in Science Park

Apr-2007 Established MMIC and Optoelectronic business units

Apr-2008 Acquired additional land in Huaya Science Park

May-2008 Successfully developed 0.25um pHEMT MMIC wafer process

Jun-2008 Skyworks Solution announced WIN as their official GaAs foundry

Oct-2008 First pilot run of HBT GaAs wafer in Fab B

Apr-2009 Fab B formally announced for mass production

Jul-2009 Successfully integrates HBT and pHEMT processes named as BiHEMT process

Oct-2009 Listed on the Emerging Board

Oct-2009 Announcing strategic partnership with Anadigics

Apr-2010 Acquired Fab B second phase land and factory

May-2010 Start renovating Fab B second phase.

Dec-2010 Developed high performance application for multiphase, multisystem cell phone switch process

Dec-2010 Successfully developed world's first 6” 0.1um MMIC wafer

Dec-2011 Listed on the Taipei Stock Exchange (formerly OTC)

Dec-2013 Became the only GaAs foundry in the world's Top 12 foundries and took 62.4% of pure-play foundry service market share

Dec-2013 Won 3rd National Industrial Innovation Award, Outstanding Enterprise Innovation Award category

Feb-2014 Won 2nd Taiwan Mittelstand Award

Apr-2015 Ranked among the top 5% in the corporate governance evaluation among all listed companies on TWSE and TSE


11 N

ov

em

ber 2

015

Taiw

an

Co

mp

ou

nd

Sem

ico

nd

ucto

r Secto

r 3

1

Figure 71: Win Semi—quarterly income statement Year-end 31 Dec (NT$mn) 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15E 1Q16E 2Q16E 3Q16E 4Q16E 2013 2014 2015E 2016E

Sales 1,667 2,461 3,069 2,713 2,807 3,099 2,913 3,089 2,678 3,470 3,708 3,331 10,481 9,910 11,907 13,187

COGS 1,167 1,599 1,932 1,702 1,744 1,898 1,735 1,850 1,660 2,099 2,225 1,999 7,249 6,400 7,227 7,983

Gross profits 500 862 1,137 1,011 1,063 1,201 1,178 1,239 1,017 1,371 1,483 1,332 3,232 3,510 4,681 5,204

Total operating exp. 297 293 271 334 288 300 351 366 327 359 369 360 1,123 1,195 1,305 1,415

Operating exp. promotion 34 29 38 34 34 34 35 36 35 35 35 33 128 136 139 138

Operating exp. administrative 94 106 133 164 128 126 150 158 139 154 167 167 499 497 561 627

Operating expense R&D 169 157 100 136 127 140 166 173 153 170 167 160 495 562 606 650

EBIT 204 569 865 677 774 902 827 872 691 1,011 1,114 973 2,110 2,315 3,375 3,789

Non-oper. income (Loss) 46 6 24 38 (32) (59) 137 (5) (5) (4) (3) (4) 103 114 41 (15)

Interest expenses (14) (12) (12) (10) (7) (5) (10) (10) (10) (10) (10) (10) (81) (48) (32) (42)

Interest income 3 4 5 5 7 7 7 5 6 7 8 7 17 17 26 27

Other non-op. income/(loss) 57 14 31 43 (32) (61) 140 - - - - - 167 145 47 -

Pre-tax Income 250 575 889 715 742 843 964 868 686 1,008 1,112 969 2,212 2,429 3,416 3,774

Income taxes exp./(gains) 37 135 136 157 124 306 159 134 120 302 189 165 401 465 723 776

Net income before extraord. 213 440 753 557 618 537 805 733 566 705 923 804 1,812 1,963 2,693 2,998

Net income 213 440 753 557 618 537 805 733 566 705 923 804 1,812 1,963 2,693 2,998

EPS (NT$) 0.36 0.74 1.27 0.94 1.04 0.90 1.35 1.23 0.95 1.18 1.55 1.35 3.02 3.31 4.53 5.03

Aveage shares 592 592 594 594 594 595 595 595 595 595 595 595 600 593 594 595

EBITDA 671 1,039 1,338 1,149 1,247 1,376 1,310 1,390 1,245 1,584 1,705 1,582 3,932 4,196 5,324 6,116

Margins (%)

Gross margin 30.0 35.0 37.0 37.3 37.9 38.8 40.4 40.1 38.0 39.5 40.0 40.0 30.8 35.4 39.3 39.5

Operating margin 12.2 23.1 28.2 24.9 27.6 29.1 28.4 28.2 25.8 29.1 30.1 29.2 20.1 23.4 28.3 28.7

Tax rate 14.6 23.5 15.3 22.0 16.7 36.3 16.5 15.5 17.5 30.0 17.0 17.0 18.1 19.2 21.2 20.6

Net margin 12.8 17.9 24.5 20.5 22.0 17.3 27.6 23.7 21.1 20.3 24.9 24.1 17.3 19.8 22.6 22.7

QoQ (%)

Sales (13) 48 25 (12) 3 10 (6) 6 (13) 30 7 (10)

COGS (17) 37 21 (12) 2 9 (9) 7 (10) 26 6 (10)

Gross profit (2) 72 32 (11) 5 13 (2) 5 (18) 35 8 (10)

Operating profit (17) 180 52 (22) 14 16 (8) 5 (21) 46 10 (13)

Net profit 69 106 71 (26) 11 (13) 50 (9) (23) 25 31 (13)

YoY (%)

Sales (41) (23) 22 41 68 26 (5) 14 (5) 12 27 8 (7) (5) 20 11

COGS (40) (25) 9 20 49 19 (10) 9 (5) 11 28 8 (5) (12) 13 10

Gross profit (45) (18) 50 97 113 39 4 23 (4) 14 26 8 (11) 9 33 11

Operating profit (67) (26) 78 177 281 58 (4) 29 (11) 12 35 11 (14) 10 46 12

Net profit (73) (20) 116 342 190 22 7 32 (8) 31 15 10 10 8 37 11


11 November 2015



Visual Photonics Epitaxy Co., Ltd (2455.TW / 2455 TT)

INITIATION

A good proxy for the RF growth trend

■ Initiating coverage with OUTPERFORM. We initiate coverage on VPEC with an OUTPERFORM rating and a TP of NT$47, based on 18x 2016E EPS. We see VPEC as a good proxy for the RF growth trend with its diversified customer base. In addition, the expansion into the optical space could lead to better margins and the superior profitability/strong balance sheet should sustain the high dividend yield of 7-8% over 2015-17.

■ A good proxy for RF growth with diversified customer base. With ~20%

global share in the GaAs epiwafer market, VPEC has a solid group of

customers including Win Semi, Skyworks, Qorvo, Avago, AWSC, etc., in

both IDM and foundry segments. We think the diversified customer base

makes it a good proxy of the RF demand growth. ■ Increasing contribution from optical positive to margins. On top of the

RF component application, VPEC is seeing increasing contribution from the optical communication sector, where VPEC leverages its expertise to provide epiwafers for photo diodes with better margins. It should have ~15% revenues from this in 2015 and should further increase with the expansion into laser diodes in 2016, leading to a more favourable product mix.

■ Superior profitability and strong balance sheet to support high

dividend payout. Though VPEC has a much smaller scale compared to its

major competitor IQE (55-60% global market share), VPEC has been able to

consistently generate higher operating margins by 15-20 pp in the past few

years, showcasing its superior execution, process control and competitive

cost structure. In addition, VPEC has a strong balance sheet, without any

debt and NT$5.7 cash per share as of 3Q15.

■ Valuation. Our TP of NT$47 is based on 18x 2016E EPS (16x excluding net

cash). Historically, its shares have been trading at 4-30x 12M forward P/E,

with the mid-cycle average at 18-19x. We think the market has been giving

VPEC some valuation premium because (1) it supplies to both IDMs and

foundry players for better order stability and (2) it’s migrating into the optical

comm space, where peers are trading at over 20x P/E for higher margins.

Share price performance

80

130

180

20

30

40

50

60





10/11/15


Performance over 1M 3M 12M Absolute (%) -7.0 -7.8 33.0 — Relative (%) -8.1 -10.9 38.6 —


Year 12/14A 12/15E 12/16E 12/17E Revenue (NT$ mn) 2,073.4 2,302.5 2,590.3 2,920.7 EBITDA (NT$ mn) 739.9 855.1 956.4 1,061.1 EBIT (NT$ mn) 524.5 646.5 740.1 839.0 Net profit (NT$ mn) 468.4 590.4 633.2 717.6 EPS (CS adj.) (NT$) 1.90 2.39 2.57 2.91 Change from previous EPS (%) n.a. Consensus EPS (NT$) n.a. 2.45 2.65 2.88 EPS growth (%) 1.7 26.0 7.3 13.3 P/E (x) 19.2 15.2 14.2 12.5 Dividend yield (%) 5.5 5.5 6.9 7.4 EV/EBITDA (x) 10.4 8.7 7.7 6.8 P/B (x) 2.6 2.5 2.5 2.5 ROE (%) 13.3 16.6 17.6 19.7 Net debt/equity (%) Net cash Net cash Net cash Net cash


Rating OUTPERFORM Price (10 Nov 15, NT$) 36.50 Target price (NT$) 47.00¹ Upside/downside (%) 28.8 Mkt cap (NT$ mn) 8,998.8 (US$ 275.2) Enterprise value (NT$ mn) 7,420 Number of shares (mn) 246.54 Free float (%) 89.9 52-week price range 56.8 - 27.1 ADTO - 6M (US$ mn) 12.7




Research Analysts

Derrick Yang

886 2 2715 6367

[email protected]

Jerry Su

886 2 2715 6361

[email protected]

11 November 2015


Visual Photonics Epitaxy Co. 2455.TW / 2455 TT Price (09 Nov 15): NT$37.15, Rating:: OUTPERFORM, Target Price: NT$47.00, Analyst: Derrick Yang


Scenario TP %Up/Dwn Assumptions Upside 64.00 72.27 25X P/E Central Case 47.00 26.51 18X P/E Downside 31.00 (16.55) 12X P/E


RF 1,814 1,968 2,177 2,436 Optical Comm — — — — 259.0 334.0 413.2 484.6 — — — — — — — —


Sales revenue 2,073 2,302 2,590 2,921 Cost of goods sold 1,352 1,445 1,624 1,828 SG&A 93.0 100.5 112.6 127.9 Other operating exp./(inc.) (112.0) (98.2) (102.2) (96.5) EBITDA 740 855 956 1,061 Depreciation & amortisation 215.5 208.6 216.4 222.1 EBIT 524.5 646.5 740.1 839.0 Net interest expense/(inc.) — (4.0) (4.9) (5.2) Non-operating inc./(exp.) 38.6 40.5 (1.6) (1.6) Associates/JV 1.6 1.6 1.6 1.6 Recurring PBT 564.6 692.5 744.9 844.2 Exceptionals/extraordinaries — — — — Taxes 96.2 102.2 111.7 126.6 Profit after tax 468.4 590.4 633.2 717.6 Other after tax income — — — — Minority interests — — — — Preferred dividends — — — — Reported net profit 468.4 590.4 633.2 717.6 Analyst adjustments — — — — Net profit (Credit Suisse) 468.4 590.4 633.2 717.6


EBIT 524.5 646.5 740.1 839.0 Net interest — — — — Tax paid — — — — Working capital (120.6) 58.7 (78.9) (49.4) Other cash & non-cash items 159.4 152.4 109.5 100.7 Operating cash flow 563.2 857.6 770.6 890.3 Capex (86.1) (100.3) (100.0) (100.0) Free cash flow to the firm 477.1 757.3 670.6 790.3 Disposals of fixed assets — — — — Acquisitions — — — — Divestments — 0.40 — — Associate investments — — — — Other investment/(outflows) 29.7 9.6 2.5 0.8 Investing cash flow (56.4) (90.3) (97.5) (99.2) Equity raised 5.7 (7.2) — — Dividends paid (493.1) (493.1) (619.9) (664.9) Net borrowings (0.31) — — — Other financing cash flow — — — — Financing cash flow (487.7) (500.3) (619.9) (664.9) Total cash flow 19.1 267.0 53.3 126.2 Adjustments — — — — Net change in cash 19.1 267.0 53.3 126.2


Cash & cash equivalents 1,312 1,579 1,632 1,758 Current receivables 374.4 324.1 448.2 491.8 Inventories 355.6 289.0 402.7 424.8 Other current assets 35.3 35.1 41.1 48.8 Current assets 2,077 2,227 2,524 2,723 Property, plant & equip. 1,774 1,685 1,569 1,447 Investments 66.4 36.9 36.9 36.9 Intangibles 1.1 1.8 1.8 1.8 Other non-current assets 28.6 28.0 25.1 23.9 Total assets 3,947 3,978 4,157 4,233 Accounts payable 269.3 197.6 331.0 314.2 Short-term debt — — — — Current provisions — — — — Other current liabilities 171.4 184.8 216.2 257.0 Current liabilities 440.8 382.4 547.2 571.3 Long-term debt — — — — Non-current provisions — — — — Other non-current liab. 0.17 0.17 0.17 0.17 Total liabilities 440.9 382.6 547.3 571.4 Shareholders' equity 3,506 3,596 3,609 3,662 Minority interests — — — — Total liabilities & equity 3,947 3,978 4,157 4,233



1.90 2.39 2.57 2.91 DPS (NT$) 2.00 2.00 2.51 2.70 BVPS (NT$) 14.2 14.6 14.6 14.9 Operating CFPS (NT$) 2.28 3.48 3.13 3.61


12/14A 12/15E 12/16E 12/17E

Growth(%) Sales revenue (3.6) 11.0 12.5 12.8 EBIT 0.7 23.3 14.5 13.4 Net profit 1.7 26.0 7.3 13.3 EPS 1.7 26.0 7.3 13.3 Margins (%)

EBITDA 35.7 37.1 36.9 36.3 EBIT 25.3 28.1 28.6 28.7 Pre-tax profit 27.2 30.1 28.8 28.9 Net profit 22.6 25.6 24.4 24.6 Valuation metrics (x) P/E 19.6 15.5 14.5 12.8 P/B 2.61 2.55 2.54 2.50 Dividend yield (%) 5.38 5.38 6.77 7.26 P/CF 16.3 10.7 11.9 10.3 EV/sales 3.78 3.29 2.91 2.53 EV/EBITDA 10.6 8.9 7.9 7.0 EV/EBIT 15.0 11.7 10.2 8.8 ROE analysis (%) ROE 13.3 16.6 17.6 19.7 ROIC 19.7 26.2 31.5 36.7 Asset turnover (x) 0.53 0.58 0.62 0.69 Interest burden (x) 1.08 1.07 1.01 1.01 Tax burden (x) 0.83 0.85 0.85 0.85 Financial leverage (x) 1.13 1.11 1.15 1.16 Credit ratios Net debt/equity (%) (37.4) (43.9) (45.2) (48.0) Net debt/EBITDA (x) (1.77) (1.85) (1.71) (1.66) Interest cover (x) — (163) (152) (161)


11 November 2015


A good proxy for the RF growth trend We initiate coverage on VPEC with an OUTPERFORM rating and a TP of NT$47, based

on 18x 2016E EPS. We see VPEC as a good proxy for the RF growth trend with its

diversified customer base for both IDM and foundry players. In addition, the expansion into

the optical space could lead to better margins and the superior profitability/strong balance

sheet should sustain the high dividend yield of 7-8% in 2015-17.

A good proxy for RF growth with the diversified customer base

As a compound semiconductor epiwafer supplier, VPEC is the upstream supplier to

foundry and IDM players. It purchases the substrates (from AXT, Freiberger, Sumitomo

Chemical, etc.) and chemicals (from Dow Chemical, SAFC, etc.) and grows different

layers of chemical materials onto the substrate in the Metal Organic Chemical Vapour

Deposition (MOCVD) chamber. It competes with IQE, Hitachi Cable, Sumitomo Chemical,

Skyworks, etc., in the space.

According to the data from Strategy Analytics, VPEC has around 21% global market share,

only second to IQE's 58% in 2013. Though the research house Strategy Analytics has

ceased the survey in the GaAs epiwafer market, we estimate that the market share gap

between VPEC and IQE has been narrowed as of now, given that VPEC's top line has

outgrown IQE's in 2014-1H15.

VPEC supplies to most of the major players in the IDM and foundry sectors, including Win

Semi, Skyworks, Qorvo, Avago, AWSC, etc. We think the diversified customer base

makes it a good proxy for the RF component demand growth trend, as it is relatively

immune from the share shift in the end market among its customers.

Figure 72: VPEC—diversified customer base (2014) Figure 73: VPEC—~20% share in the global GaAs

epiwafer market (2013)

Win Semi 28%

Skyworks 19%

Qorvo 14%

Avago 11%

AWSC 9%

Others 19%

IQE 58%

VPEC 21%

Hitachi Cable 5%

Sumitomo Chem 2%

Skyworks 5%

Others 10%

Source: Company data, Credit Suisse Source: Strategy Analytics, Credit Suisse

Increasing contribution from optical positive to

margins

While VPEC has the majority of revenues from the RF component in the wireless

communication application, it has been also tapping into other opportunities that could also

leverage its expertise in the compound semiconductor materials. Photo diodes used in the

transceiver module for the optical communication application is one of these areas. The

epiwafer wafer used to manufacture photo diodes shares some common know-how with

the epiwafers for the RF components, as they are both associated with growing epi layers

Diversified customer base

and diversification into the

optical communication

space

GaAs epiwafer supplier

competing with IQE, Hitachi

Cable, Sumitomo Chemical,

Skyworks, etc.

21% market share in 2013

vs 58% for IQE

Leveraging the expertise

into the optical

communication space

11 November 2015


on a substrate in the MOCVD chamber, just that the materials and epi layer structures are

different.

VPEC ships most of its epiwafers to the optical transceiver module suppliers in the greater

China area and is competing with IQE, Landmark, etc., in this segment. VPEC's epiwafers

are mostly used in the fibre to the home (FTTH) market, where the demand has been

driven by increasing broadband coverage and speed upgrade. China has been one of the

most aggressive countries promoting the FTTH infrastructure, as it had articulated the

blueprint in the "Broadband China" initiative that it will expand the FTTH coverage in China

to 200 mn and 300 mn households in 2015 and 2020, compared to 130 mn in 2013.

Figure 74: Optical transceiver architecture Figure 75: FTTH coverage target in the Broadband China

initiatives

Laser diode inside

Photo diode inside

0

50

100

150

200

250

300

350

2013 2015E 2020Em

n ho

useh

olds

FTTH coverage

Source: Finisar, Credit Suisse Source: The State Council of PRC, Credit Suisse

We estimate that VPEC will have ~15% of its revenues generated from the optical

communication application in 2015 vs 12% in 2013 and the contribution should continue to

rise to 17% in 2017 as it plans to introduce another product line for laser diodes in 2016 to

expand its total addressable market in the optical communication space.

Given that the margin from the optical communication is higher than its corporate average,

we believe that the product mix shift should be accretive to its overall margin in the next

few years.

Figure 76: VPEC—~15% of revenues from optical

communication in 2015

Figure 77: VPEC—more contribution from optical

communication positive to margins

97%93%

88%85% 84% 83%

3%7%

12%15% 16% 17%

75%

80%

85%

90%

95%

100%

2012 2013 2014 2015E 2016E 2017E

RF Optical Communication

3%

7%

12%

15%16%

17%

32%

33%

34%

35%

36%

37%

38%

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

2012 2013 2014 2015E 2016E 2017E

Revenue contribution from optical communication (LHS) Gross margin (RHS)


Superior profitability and strong balance sheet to

support high dividend payout

Though VPEC operates with a much smaller scale compared to its major competitor IQE,

with only 45% of IQE's revenue size in 1H15, VPEC has been able to consistently

generate higher operating margin by 15-20 pp in the past few years. We think this is

reflective of its superior execution, process control and competitive cost structure.

Riding on the FTTH demand

for optical transceivers

15% revenues from the

optical communication

space in 2015

Consistent 15-20 pp OPM

premium over top

competitor IQE

11 November 2015


Figure 78: VPEC—~45% as big as IQE in revenues Figure 79: VPEC—consistent 15-20 pp premium over IQE

in operating margins

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

2010 2011 2012 2013 2014 1H15

NT

$mn

VPEC IQE

0%

5%

10%

15%

20%

25%

30%

35%

2010 2011 2012 2013 2014 1H15

VPEC IQE (Wireless segment)

VPEC consistently has 15-20pp premium over IQE


VPEC has a strong balance sheet. It currently has no interest-bearing debt on the balance

sheet and the net debt to equity ratio was at -40% as of 3Q15. This is still improving given

that we forecast the company to generate NT$800-900 mn free cash flow per year over

2015-17. We expect the net cash on hand to reach NT$6-7 per share over 2015-17. With

the solid financial position and decent capability in generating free cash flows, we expect

its dividend payout ratio to be 105% in 2015-17, similar to the level seen in 2013-14,

implying 7-8% dividend yield in 2015-17.

Figure 80: VPEC—sustaining free cash flows Figure 81: VPEC—gearing continuing to improve

0

100

200

300

400

500

600

700

800

900

1,000

2011 2012 2013 2014 2015E 2016E 2017E

NT

$mn

Free cash flow

-60%

-50%

-40%

-30%

-20%

-10%

0%

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2011 2012 2013 2014 2015E 2016E 2017E

NT

$mn

Cash and equivalent Net debt to equity


Figure 82: VPEC—NT$6-7 net cash per share in 2015-17 Figure 83: VPEC—7-8% dividend on 100%+ payout ratio

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

2011 2012 2013 2014 2015E 2016E 2017E

NT

$

Net cash per share

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

0%

20%

40%

60%

80%

100%

120%

140%

2011 2012 2013 2014 2015E 2016E 2017E

Dividend payout Dividend yield


Strong balance sheet and

free cash flows supporting

higher dividend payout

11 November 2015


Financial analysis

Inventory adjustment in 3Q15; flat to be down single digit QoQ for 4Q15 revenues

VPEC 3Q15 revenues were down 9% QoQ and down 3% YoY, reflecting the inventory

adjustment in the China smartphone market and the weaker purchasing power on

currency depreciation in emerging markets. 3Q15 gross margin came down to 35.8% vs

38.6% in 2Q15, as the impact from the lower utilisation outweighed the benefit of a more

favourable product mix and NTD depreciation.

Looking into 4Q15, management expects the revenues to be down single digit QoQ, as

the seasonality will be partially offset by some restocking demand on lean inventories.

4Q15 gross margin is guided to be similar to 3Q15. We forecast 4Q15 revenues to be

down 9% QoQ and the gross margin to be 35.5%.

Riding on the RF component demand and more contribution from optical in 2016

Looking into 2016, we forecast 13% YoY revenue growth to NT$2,590 mn, riding on the

RF component sector growth and more traction in the optical communication space. We

expect the gross margin to expand slightly to 37.3% in 2016 from 37.2% in 2015, on the

modest product mix shift to higher contribution from the optical communication space. We

model for some operating leverage and expect the operating expense ratio to decline to

8.8% in 2016 from 9.2% in 2015, leading to the operating margin of 28.6% in 2016 vs

28.1% in 2015.

By different applications, we expect the revenues from the RF application to grow 10%

YoY, similar to the overall sector growth and the revenues from the optical communication

application to outgrow the corporate average (20% CAGR over 2015-17 vs corporate

average of 12%) as it will introduce the new product line for laser diodes in the FTTH

market.

Figure 84: VPEC—quarterly revenues and margins Figure 85: VPEC—revenue breakdown

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0

100

200

300

400

500

600

700

800

1Q12 3Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16E 3Q16E

NT

$mn


97%93%

88%85% 84% 83%

3%7%

12%15% 16% 17%

75%

80%

85%

90%

95%

100%

2012 2013 2014 2015E 2016E 2017E

RF Optical Communication


Figure 86: VPEC—CS estimate vs consensus

(NT$ mn) 4Q15E 1Q16E 2015E 2016E


Revenue 538 568 501 563 2,302 2,332 2,590 2,565

Sequential growth -9% -4% -7% -1% 11% 12% 13% 10%

Gross profits 191 203 175 212 857 870 967 977

Operating profits 140 148 124 159 647 655 740 755

Net profits 114 128 107 139 590 604 633 653

EPS (NT$) 0.46 0.52 0.43 0.56 2.39 2.45 2.57 2.65

GM 35.5% 35.8% 35.0% 37.6% 37.2% 37.3% 37.3% 38.1%

OPM 26.0% 26.1% 24.8% 28.2% 28.1% 28.1% 28.6% 29.4%

NM 21.2% 22.5% 21.3% 24.7% 25.6% 25.9% 24.4% 25.5%


13% top line growth for

2016; margin expansion on

efficiency improvement and

product mix

11 November 2015



We estimate that VPEC will generate NT$0.8-0.9 bn free cash flows, or NT$3.2-3.7 free

cash flow per share in 2015-17, reflecting a free cash flow yield of 8.6-10.1%, vs 7.5% in

2014. The improving free cash flow yield is resulting from better earnings and stable

capex. Capex is expected to be ~NT$100 mn in 2015-17, implying 3.4-4.3% capex to

sales ratio. We estimate the depreciation-to-sales ratio to be around 7.6-9.1% in 2015-17,

slightly down from ~10% in 2013-14, as capex has been lower than its depreciation

expenses in the past few years.

Given that VPEC has been able to generate free cash and the strong cash position, we

assume VPEC to maintain the payout ratio of 105% in 2015-17, similar to the level in

2013-14. Based on our earnings forecast and dividend payout, we expect to see 6.9-8.4%

dividend yield in 2015-17.

Figure 87: VPEC—cash flow summary (2011-17E)

(NT$ mn) 2011 2012 2013 2014 2015E 2016E 2017E

Revenues 2,153 2,249 2,150 2,073 2,302 2,590 2,921

Capex 489 280 125 86 100 100 100

Capex / Revenues (%) 22.7 12.5 5.8 4.2 4.4 3.9 3.4

Dep and Amort 196 226 215 215 209 216 222

Dep / Revenues (%) 9.1 10.1 10.0 10.4 9.1 8.4 7.6

Operating cash flow 546 806 801 563 858 771 890

Free Cash flow 136 632 736 533 814 778 912

FCF per share 0.6 2.6 3.0 2.2 3.3 3.2 3.7

FCF yield (%) 2.3 8.6 10.9 7.5 9.0 8.6 10.1


Dividend yield (%) 8.3 8.4 7.3 7.0 6.9 7.4 8.4


Initiating coverage with OUTPERFORM and NT$47 TP

We initiate coverage on VPEC with an OUTPERFORM rating and a target price of NT$47,

based on 18x 2016E EPS. Historically, its shares have been trading at 4-30x 12M forward

P/E, with the mid-cycle average at 18-19x. As we forecast a 13% operating profit CAGR

over 2015-17, higher than 5% over 2013-15, we think the mid-cycle average is not a

demanding target for valuation. In addition, if we exclude the net cash position that VPEC

holds on hand, our target multiple will be 16x 2016E P/E. We think the market has been

giving valuation premiums for VPEC vs foundry players, because (1) it supplies epiwafers

to both IDMs and foundry players, so should have less significant order adjustment vs

foundry players when the end demand turns softer, and (2) it is migrating to the optical

communication space, where investors are willing to pay higher valuation (leader in the

optical communication epiwafer suppliers in Taiwan Landmark Opto trading at 23x 2016E

consensus P/E now, with 60-65% gross margin since 2014).

Downside risks

Downside risks to our positive view on VPEC would include:

■ Weaker growth in the RF component space: As VPEC supplies compound

semiconductor epiwafers to both IDM and foundry players and ~85% of its revenues

are from the RF component space, weaker end demand for RF components should

lead to a slower growth for VPEC.

■ Slower traction in the optical communication expansion. VPEC has been diversifying

into the optical communication space, which should account for ~15% of its total

revenues in 2015 and higher beyond. Though management does not disclose the

margin from orders in this application. The 60-65% gross margins reported by the peer

Landmark (3081.TWO) in since 2014 might be a reference. Though VPEC's margin in


per share in 2015-17

105% dividend payout in

2015-17

11 November 2015


this business could be lower than Landmark on its different product mix and smaller

scale, we believe orders from the optical communication should be above its corporate

average. Thus, slower traction in this space could negatively impact its earnings.

■ High-than-expected adoption of the silicon-based RF components impacting the total

addressable market. Any disruptive technology that could lead to better performance

of silicon-based RF components at the expense of compound semiconductor could

lead to a smaller total addressable market for VPEC.

Figure 88: VPEC—12-month forward P/E Figure 89: VPEC—12-month forward P/B

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

Jan-0

9

May-0

9

Sep-0

9

Jan-1

0

May-1

0

Sep-1

0

Jan-1

1

May-1

1

Sep-1

1

Jan-1

2

May-1

2

Sep-1

2

Jan-1

3

May-1

3

Sep-1

3

Jan-1

4

May-1

4

Sep-1

4

Jan-1

5

May-1

5

Sep-1

5

Average = 19

+1 Std dev = 24

-1 Std dev = 13

+2 Std dev = 29

-2 Std dev = 8

VPEC PE (x)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Jan-0

9

May-0

9

Sep-0

9

Jan-1

0

May-1

0

Sep-1

0

Jan-1

1

May-1

1

Sep-1

1

Jan-1

2

May-1

2

Sep-1

2

Jan-1

3

May-1

3

Sep-1

3

Jan-1

4

May-1

4

Sep-1

4

Jan-1

5

May-1

5

Sep-1

5

Average = 2.6

+1 Std dev = 3.3

-1 Std dev = 1.9

+2 Std dev = 4.0

-2 Std dev = 1.2

VPEC PB (x)





assumptions, CS HOLT would value VPEC at NT$47.77, similar to our target price.



11 November 2015


Company profile

Company background

Founded in November 1996, VPEC has been focusing on the compound semiconductor

epiwafer fabrication. VPEC exited the LED epiwafer business in 2009 due to unfavourable

market environments and currently focuses on the GaAs epiwafer space. It sells epiwafers

to downstream IDM or foundry players for the RF components. VPEC now operates

around 40 MOCVD tools in Taoyuan (Taiwan). It was listed on Taiwan Stock Exchange

(TWSE) in January 2002.


VPEC has a fairly diversified shareholder structure, as top ten shareholders only own

15.9% of the company. Management owns 8-9% of the company. The QFII holdings at

VPEC has been rising from ~5% in April 2013 to 15.0% currently.

Figure 91: VPEC—QFII holdings vs share prices Figure 92: VPEC—major shareholders

0

10

20

30

40

50

60

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

Jan-11 Jul-11 Jan-12 Jul-12 Jan-13 Jul-13 Jan-14 Jul-14 Jan-15 Jul-15

NT

$

QFII Holdings Share price (NT$, RHS)

Shareholder Stake

1 Deutsche Bank A.G. 2.4%

2 HSBC (Taiwan) & UBS AG Investment Accoun 2.0%

3 Standard Chartered Bank & Standard Chart 1.6%

4 HSBC (Taiwan) Commercial Bank & Goldman 1.6%

5 HSBC (Taiwan) & UBS Securities & Singapo 1.5%

6 Datum & Central Bank Of Norseland Inv. 1.4%

7 Ch'en Chien Liang-A 1.4%

8 Ch'en Mao Ch'ang--a 1.4%

9 CITI Bank & Berkeley Trust Account 1.3%

10 Chang Sun Tsui 1.3%


Figure 93: VPEC—management profile


Chairman Maochang Chen Chairman at VPEC National Hualien Industrial Vocational Senior High School

CEO Chaohsing Huang CEO at VPEC Ph.D. of Electrical Engineering from National Taiwan University

Senior VP Hsuenchung Chen Director at Coretronic Mexico Dab Director at TS Metal

National Hualien Industrial Vocational Senior High School

VP Changming Wu Director at VPEC NA

VP Yuchung King VP at VPEC Ph.D. of Electronics from National Taiwan University

VP Jinglong Hsieh VP at VPEC Master of Electrical Engineering from National Ocean University Master of BA from Chung Yuan Christian University

Director of Finance Jingling Chung Director at VPEC Master of Accounting from Chung Yuan Christian University


40 MOCVD tools in

operation

11 November 2015


Figure 94: VPEC—key milestones

Time Milestone

Nov-1996 Founded in Taoyuan, Taiwan

Oct-1997 First MOCVD installed for epiwafers

Jan-1998 LED epiwafer reached the ultra-brightness grade

Feb-1998 Pilot production of HBT epiwafers for qualification

Mar-1998 LED epiwafers sent for qualification

May-2000 Pingcheng factory started operation and pHEMT pilot production

Jun-2000 HBT epiwafers qualified by the leading customer

Oct-2000 High power LED epiwafers in mass production

Jan-2002 Listed on TWSE

Oct-2002 InP HBT qualified by customers

Sep-2004 Private placement of 3.4mn shares

Mar-2005 Private placement of 8.25mn shares

Jul-2005 Private placement of 8.35mn shares

Jun-2007 Granted US patent for HBT technology

Oct-2007 Right issue of 8mn shares

Nov-2007 Granted US patent for HB LED having reflective layer

Jan-2009 Sold LED-related equipment

Dec-2013 Pingcheng factory II completed

Feb-2015 Granted US patent for BiHEMT device having a stacked separating layer


11 N

ov

em

ber 2

015

Taiw

an

Co

mp

ou

nd

Sem

ico

nd

ucto

r Secto

r 4

2

Figure 95: VPEC—quarterly income statement

Year-end 31 Dec (NT$mn) 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15E 1Q16E 2Q16E 3Q16E 4Q16E 2013 2014 2015E 2016E

Sales 376 527 605 565 525 650 590 538 501 705 755 629 2,150 2,073 2,302 2,590

COGS 256 349 390 357 320 399 379 347 326 433 461 403 1,446 1,352 1,445 1,624

Gross profits 120 178 215 208 205 251 211 191 175 271 294 226 704 721 857 967

Total operating exp. 44 48 48 57 51 54 55 51 51 60 61 55 183 196 211 227

Operating exp. promotion 3 3 4 5 3 3 2 3 4 4 3 3 12 15 11 13

Operating exp. administrative 16 20 21 22 20 24 23 22 22 26 27 25 76 78 90 99

Operating expense R&D 26 25 23 30 28 26 30 27 26 30 31 27 95 103 110 114

EBIT 76 130 168 151 154 197 156 140 124 211 233 172 521 524 647 740

Non-oper. income (Loss) 9 (3) 10 24 (2) 4 43 1 1 1 1 1 32 40 46 5

Interest expenses - - - - - - - - - - - - - - - -

Interest income 1 2 1 (4) 1 2 - 1 1 1 1 1 - - 4 5

Other non-op. income/(loss) 8 (5) 9 28 (3) 2 43 - - - - - 32 40 42 -

Pre-tax Income 85 127 178 174 153 201 198 141 126 213 234 173 553 565 693 745

Income taxes exp./(gains) 11 40 22 23 20 28 28 27 19 32 35 26 92 96 102 112

Net income before extraord. 74 88 155 151 133 173 171 114 107 181 199 147 461 468 590 633

Net income 74 88 155 151 133 173 171 114 107 181 199 147 461 468 590 633

EPS (NT$) 0.30 0.36 0.63 0.61 0.54 0.70 0.69 0.46 0.43 0.73 0.81 0.60 1.87 1.90 2.39 2.57

Aveage shares 247 247 247 247 247 247 247 247 247 247 247 247 247 247 247 247

EBITDA 130 184 221 205 206 249 208 192 178 265 287 226 735 740 855 956

Margins (%)

EBITDA margin 34.6 34.9 36.4 36.3 39.2 38.3 35.3 35.8 35.5 37.6 38.0 36.0 34.2 35.7 37.1 36.9

Gross margin 31.9 33.8 35.5 36.8 39.0 38.6 35.8 35.5 35.0 38.5 38.9 36.0 32.7 34.8 37.2 37.3

Operating margin 20.2 24.7 27.7 26.7 29.4 30.3 26.4 26.0 24.8 30.0 30.8 27.3 24.2 25.3 28.1 28.6

Tax rate 12.8 31.1 12.6 13.4 12.9 14.0 13.9 19.0 15.0 15.0 15.0 15.0 16.7 17.0 14.8 15.0

Net margin 19.8 16.7 25.6 26.7 25.3 26.5 28.9 21.2 21.3 25.7 26.3 23.3 21.4 22.6 25.6 24.4

QoQ (%)

Sales (8) 40 15 (7) (7) 24 (9) (9) (7) 41 7 (17)

COGS (12) 36 12 (8) (10) 25 (5) (8) (6) 33 6 (13)

Gross profit 2 48 21 (3) (1) 22 (16) (10) (8) 55 8 (23)

Operating profit 13 71 29 (10) 2 28 (21) (10) (11) 70 10 (26)

Net profit 20 18 77 (3) (12) 30 (1) (33) (6) 69 10 (26)

YoY (%)

Sales (30) (27) 25 39 40 23 (3) (5) (4) 8 28 17 (4) (4) 11 13

COGS (30) (24) 17 23 25 14 (3) (3) 2 9 22 16 (3) (6) 7 12

Gross profit (31) (31) 40 76 70 41 (2) (8) (14) 8 39 19 (7) 2 19 13

Operating profit (44) (39) 57 125 103 51 (7) (7) (19) 7 49 23 (8) 1 23 14

Net profit (44) (53) 94 144 79 97 10 (25) (20) 5 17 29 (0) 2 26 7


11 November 2015



Advanced Wireless Semiconductor Company (8086.TWO / 8086 TT)

INITIATION

Standing on the shoulders of the giant

■ Initiating coverage with NEUTRAL. We initiate coverage on AWSC with a

NEUTRAL rating and a target price of NT$88, based on 11x 2016E EPS. We

believe AWSC is a quality name operating in a duopolistic competitive

landscape, though the high reliance on Skyworks (>80% of total revenues)

could be a mixed blessing, as the benefit of the close relationship could be

coupled with higher business volatility. Although the new business and new

customer addition could help the outlook into 2016, we believe the shares

are fairly valued at this level.

■ Heavy reliance on Skyworks a mixed blessing. As the second largest

GaAs foundry player with 34% market share, AWSC generated over 80% of

its total revenues from its top customer Skyworks. We see the high reliance

on Skyworks a mixed blessing. On the positive side, by leveraging Skyworks'

experiences and equipment expertise to shorten the learning curve and

alleviate the capex burden. However, on the negative side, there could

potentially be higher business volatility, as AWSC might act as the buffer

capacity for Skyworks to balance own utilisation over seasonality or cycles.

■ Emerging business opportunities in 2016. On top of the GaAs foundry

service to Skyworks, AWSC will add more value and start to do the copper

pillar bumping after the wafer process to help simplify the logistic process for

the customer, which should be positive to the product ASP for AWSC in

2016. In addition, AWSC had secured a new Japanese IDM customer for the

RF component in the GPS application and expects to see more revenue

contribution in 2016.

■ Valuation. Our TP of NT$88 is based on 11x 2016E EPS. Due to the

earnings volatility in the past few years, AWSC does not have a stable P/E

valuation history for reference. We thus benchmark our target multiple to its

peer Win Semi (we use 12x P/E for TP), though with a modest discount to

factor in the potential single client risk and higher business volatility.

Share price performance

0

500

1000

0

50

100

150

200





10/11/15


Performance over 1M 3M 12M Absolute (%) -2.4 -0.5 156.9 — Relative (%) -3.4 -3.6 162.5 —


Year 12/14A 12/15E 12/16E 12/17E Revenue (NT$ mn) 2,669.4 4,437.4 4,916.5 5,455.6 EBITDA (NT$ mn) 625.4 1,478.9 1,645.6 1,843.2 EBIT (NT$ mn) 396.1 1,234.4 1,379.2 1,544.9 Net profit (NT$ mn) 374.7 1,078.8 1,173.8 1,315.4 EPS (CS adj.) (NT$) 2.76 7.83 8.37 9.37 Change from previous EPS (%) n.a. Consensus EPS (NT$) n.a. 7.27 8.09 — EPS growth (%) n.m. 183.3 6.8 12.1 P/E (x) 28.4 10.0 9.4 8.4 Dividend yield (%) 0 2.5 6.8 7.5 EV/EBITDA (x) 16.1 6.2 5.3 4.5 P/B (x) 4.8 3.4 3.1 2.7 ROE (%) 18.7 40.0 34.7 34.2 Net debt/equity (%) Net cash Net cash Net cash Net cash


Rating NEUTRAL Price (10 Nov 15, NT$) 78.60 Target price (NT$) 88.00¹ Upside/downside (%) 12.0 Mkt cap (NT$ mn) 10,756.9 (US$ 329.0) Enterprise value (NT$ mn) 9,163 Number of shares (mn) 136.86 Free float (%) 86.0 52-week price range 128.0 - 29.4 ADTO - 6M (US$ mn) 37.9




[V] = Stock considered volatile (see Disclosure Appendix).

Research Analysts

Derrick Yang

886 2 2715 6367

[email protected]

Jerry Su

886 2 2715 6361

[email protected]

11 November 2015


Advanced Wireless Semiconductor Company 8086.TWO / 8086 TT Price (09 Nov 15): NT$79.70, Rating:: NEUTRAL [V], Target Price: NT$88.00, Analyst: Derrick Yang


Scenario TP %Up/Dwn Assumptions Upside 125.00 56.84 15x P/E Central Case 88.00 10.41 11x P/E Downside 50.00 (37.26) 6x P/E


Mobile 2,160 3,607 3,988 4,376 WiFi 438.3 710.0 796.5 923.0 Others — — — — — — — — 71.5 120.1 131.5 156.8


Sales revenue 2,669 4,437 4,916 5,456 Cost of goods sold 2,092 2,956 3,273 3,628 SG&A 70.2 89.7 99.4 106.6 Other operating exp./(inc.) (118.0) (87.7) (101.7) (121.9) EBITDA 625 1,479 1,646 1,843 Depreciation & amortisation 229.3 244.5 266.4 298.3 EBIT 396 1,234 1,379 1,545 Net interest expense/(inc.) (2.7) (3.5) 0.7 0.7 Non-operating inc./(exp.) 33.4 21.1 (1.6) (1.6) Associates/JV 1.6 1.6 1.6 1.6 Recurring PBT 434 1,261 1,379 1,544 Exceptionals/extraordinaries — — — — Taxes 59.1 181.7 204.7 228.8 Profit after tax 375 1,079 1,174 1,315 Other after tax income — — — — Minority interests — — — — Preferred dividends — — — — Reported net profit 375 1,079 1,174 1,315 Analyst adjustments — — — — Net profit (Credit Suisse) 375 1,079 1,174 1,315


EBIT 396 1,234 1,379 1,545 Net interest — — — — Tax paid — — — — Working capital (124.2) 57.0 61.2 3.4 Other cash & non-cash items 207.8 89.0 61.0 68.7 Operating cash flow 480 1,380 1,501 1,617 Capex (318.2) (360.3) (360.0) (360.0) Free cash flow to the firm 162 1,020 1,141 1,257 Disposals of fixed assets — — — — Acquisitions — — — — Divestments 4.4 0.0 — — Associate investments — — — — Other investment/(outflows) 6.6 (11.1) 0.3 (2.6) Investing cash flow (307.1) (371.4) (359.7) (362.6) Equity raised 50.9 151.2 — — Dividends paid — (274.1) (755.2) (821.7) Net borrowings (39.3) (77.6) 2.8 5.3 Other financing cash flow — — — — Financing cash flow 11.6 (200.5) (752.4) (816.3) Total cash flow 184.2 808.4 389.4 438.1 Adjustments — — — — Net change in cash 184.2 808.4 389.4 438.1


Cash & cash equivalents 925 1,734 2,123 2,561 Current receivables 251.6 259.0 236.2 241.4 Inventories 378.6 420.2 481.8 553.1 Other current assets 117.5 146.2 155.5 178.7 Current assets 1,673 2,559 2,996 3,534 Property, plant & equip. 1,081 1,272 1,370 1,437 Investments 0.50 0.50 0.50 0.50 Intangibles 6.6 4.3 4.3 4.3 Other non-current assets 135.2 73.7 68.7 66.6 Total assets 2,897 3,910 4,440 5,042 Accounts payable 246.1 224.1 310.4 356.4 Short-term debt 111.5 90.9 92.7 96.2 Current provisions — — — — Other current liabilities 202.5 359.1 382.0 439.2 Current liabilities 560.1 674.0 785.1 891.7 Long-term debt 105.6 48.6 49.6 51.5 Non-current provisions — — — — Other non-current liab. 9.6 9.6 9.6 9.6 Total liabilities 675.3 732.3 844.4 952.8 Shareholders' equity 2,221 3,177 3,596 4,090 Minority interests — — — — Total liabilities & equity 2,897 3,910 4,440 5,042



2.8 7.8 8.4 9.4 DPS (NT$) — 1.95 5.38 5.86 BVPS (NT$) 16.4 23.1 25.6 29.1 Operating CFPS (NT$) 3.5 10.0 10.7 11.5


12/14A 12/15E 12/16E 12/17E

Growth(%) Sales revenue 150 66 11 11 EBIT 699 212 12 12 Net profit 606 188 9 12 EPS 606 183 7 12 Margins (%)

EBITDA 23.4 33.3 33.5 33.8 EBIT 14.8 27.8 28.1 28.3 Pre-tax profit 16.3 28.4 28.0 28.3 Net profit 14.0 24.3 23.9 24.1 Valuation metrics (x) P/E 28.8 10.2 9.5 8.5 P/B 4.86 3.46 3.11 2.73 Dividend yield (%) — 2.45 6.75 7.35 P/CF 22.5 8.0 7.4 6.9 EV/sales 3.82 2.10 1.82 1.56 EV/EBITDA 16.3 6.3 5.4 4.6 EV/EBIT 25.7 7.5 6.5 5.5 ROE analysis (%) ROE 18.7 40.0 34.7 34.2 ROIC 24.3 68.2 73.4 80.0 Asset turnover (x) 0.92 1.14 1.11 1.08 Interest burden (x) 1.10 1.02 1.00 1.00 Tax burden (x) 0.86 0.86 0.85 0.85 Financial leverage (x) 1.30 1.23 1.23 1.23 Credit ratios Net debt/equity (%) (31.9) (50.2) (55.1) (59.0) Net debt/EBITDA (x) (1.13) (1.08) (1.20) (1.31) Interest cover (x) (147) (357) 1,966 2,119


11 November 2015


Standing on the shoulders of the giant We initiate coverage on AWSC with a NEUTRAL rating and a target price of NT$88, based

on 11x 2016E EPS. We believe AWSC is a quality name operating in a duopolistic

competitive landscape, though the high reliance on Skyworks could be a mixed blessing,

as the benefit of the close relationship could be coupled with higher business volatility.

Although the new business and new customer addition could help the outlook into 2016,

we believe the shares are fairly valued at this level.

Heavy reliance on Skyworks a mixed blessing

Over 80% of revenues from Skyworks

AWSC is the second largest GaAs foundry player in the world, with 34% market share.

Though the RF component end market is dominated by a few players, AWSC has an

exceptionally high exposure to Skyworks. AWSC has seen its revenue contribution from

the top customer increasing from ~55-65% in 2011-14 to over 80% in 2014, which we

believe is reflective of the higher outsourcing ratio from Skyworks.

Figure 96: AWSC—34% share in the GaAs foundry space Figure 97: AWSC—over 80% revenues from Skyworks

Win Semi 63%

AWSC 34%

GCS 2% Others 1%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

0

500

1,000

1,500

2,000

2,500

3,000

2011 2012 2013 2014

NT

$mn

AWSC revenues Revenues from Skyworks Revenue controbution from Skyworks

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

Standing on the shoulders of the giant

We see AWSC's heavy reliance on Skyworks as a mixed blessing. There are several

positive implications from different fronts: Firstly, we think AWSC can leverage Skyworks'

existing knowledge on production facilities. As the equipment for the compound

semiconductor needs relatively high levels of customisation and different configuration

could lead to different equipment costs, we think AWSC has been benefiting from this, by

sharing the equipment design and procurement scale.

If we look at the capex-to-sales ratio, we estimate it should be an average of 8% over

2013-17, while we estimate the same ratio to be 20% on average for its major competitor

Win Semi. Similar trend in the depreciation-expenses-to-sales ratio, we estimate the

depreciation expenses to be 7% of its total sales on average over 2013-17, while the same

ratio is estimated to be 18% for Win Semi over the same period of time.

Secondly, we think the experience and knowledge from Skyworks could help AWSC

shorten the leaning learning curve in ramping up the production process for new

production and sort out potential glitches.

Thirdly, the R&D expenses could be lower on more shared activities and the marketing

expenses could be more efficiently spent on a fewer numbers of customers. This is

reflected in its operating expenses to sales ratio, where AWSC is estimated to deliver an

average of 6.1% in 2013-17, vs 10.9% for Win Semi, despite the former's much smaller

scale.

Heavy reliance on Skyworks

a mixed blessing

Over 80% revenues from

Skyworks in 2014

Leveraging Skyworks'

knowledge in production

equipment

Benefiting from Skyworks'

production know-how

Lower opex ratio vs peer

11 November 2015


Figure 98: AWSC—lower capex-to-sales (avg 13-17E) Figure 99: AWSC—lower dep exp-to-sales (avg 13-17E)

8%

20%

0%

5%

10%

15%

20%

25%

AWSC Win Semi

7%

18%

0%

5%

10%

15%

20%

25%

AWSC Win Semi


Figure 100: AWSC—lower opex ratio (avg 13-17E) Figure 101: AWSC—lower R&D as % of sales (avg 13-17E)

6.1%

10.9%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

AWSC Win Semi

3.8%

5.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

AWSC Win Semi


Higher business volatility coming along with all the benefits

Despite all the benefits from the relationship with Skyworks mentioned above, we think

there are some prices to pay as well and the primary one might be the business volatility.

Looking at the financial numbers in 2011-14, we can see that AWSC has been

experiencing a much higher volatility on both the top line and bottom line, compared to its

peer Win Semi. We think this is because Skyworks would prioritise higher utilisation at its

own capacity and use AWSC as buffer capacity to ride through seasonality and business

cycles.

Excluding the volatility and just looking at the good times like in 2015, we expect its

operating margin to be slightly lower than Win Semi's 28-29%. Despite some difference in

the product mix (Win Semi has more higher-margin WiFi and infrastructure business), we

think the magnitude still points to the fact that most of the cost benefits have been passed

through to Skyworks. Lastly, we think the close relationship with Skyworks could limit its

business opportunities in RF component players, especially bigger ones competing head-

to-head with Skyworks.

Higher volatility on both top

line and bottom line

Benefits of lower operating

cost mostly passed through

to the customer

11 November 2015


Figure 102: AWSC—more volatility in revenue growth Figure 103: AWSC—also higher OP volatility

-50%

0%

50%

100%

150%

200%

2012 2013 2014 2015E 2016E 2017E

AWSC Win Semi

-20%

-10%

0%

10%

20%

30%

40%

50%

-800%

-600%

-400%

-200%

0%

200%

400%

2012 2013 2014 2015E 2016E 2017E

AWSC (LHS) Win Semi (RHS)


Emerging business opportunities in 2016

On top of the GaAs foundry service, AWSC will also start to do the copper pillar bumping

to Skyworks. Copper pillar bumps present advantages in package size and heat

dissipation over the existing wire-bonding technology used for more RF components. The

value-added from AWSC is not only from the extra work done, but also from the simplified

logistic, as the existing outsourcing partner providing the copper pillar bumping service

could be spared. We think this development could be positive to AWSC's business due to

the higher ASP per wafer.

In addition, AWSC had secured a new Japanese IDM customer for the RF component in

the GPS application and expects to see more revenue contribution in 2016.

Figure 104: Copper pillar bumping structure Figure 105: Copper pillar bumping process

Copper pillar bump

Probing / Thinning/

Dicing / AVIExisting Process

Copper Pillar

Bumping Process

MMIC Front Side

Process with Full

Thickness

Seedlayer

DepositionPR Patterning

Bump Alloy

Deposition

PR stripingSeedlayer EtchingReflowQuality Inspection


Financial analysis

3Q15 revenues up 6% QoQ; 4Q15 to be down single digit QoQ

AWSC 3Q15 revenues were up 6% QoQ and up 42% YoY on sustaining outsourcing

orders from its major customer, though the gross margin decline 1.2 pp to 33.2% due to

the lower yield in the new product ramp. The operating expense ratio was roughly flat at

5.4% vs 2Q15 on continuous R&D efforts for the new product improvement.

Looking into 4Q15, management expects revenue to be down QoQ, though still up YoY,

off the high base in 3Q15. 3Q15 gross margin is guided to be similar to 3Q15 as the lower

utilisation should be offset by the improving yield for the new product. We forecast 4Q15

revenues to be down 6% QoQ but up 9% YoY, and the gross margin to be 32.8%.

Copper pillar bumping from

2016

New Japanese IDM

customer for the RF

component in GPS

11 November 2015


11% YoY top line growth in 2016

Looking into 2016, we model for 11% top line growth to NT$4,916 mn, driven by the

cellular RF demand, new copper pillar bumping business, WiFi spec upgrade to 802.11ac,

and increasing contribution from the new Japanese customer for the GPS application. We

expect the gross margin to be roughly flat at 33.4% in 2016 on a similar product mix,

though the operation margin should expand slightly to 28.1% in 2016 from 27.8% in 2015

due to the operating leverage.

On top of the growth along with its major customer, AWSC should see an incremental

driver from the copper pillar bumping that it will start to provide to Skyworks in 2016. For

the non-cellular business, AWSC will see more orders from its Japanese customer in 2016

for the RF component used in the GPS function.

Figure 106: AWSC—quarterly revenues and margins Figure 107: AWSC—revenue breakdown

-20%

-10%

0%

10%

20%

30%

40%

0

200

400

600

800

1,000

1,200

1,400

1,600

1Q12 3Q12 1Q13 3Q13 1Q14 3Q14 1Q15 3Q15 1Q16E 3Q16E

NT

$mn


75% 72%81% 81% 81% 80%

23%25%

16% 16% 16% 17%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2012 2013 2014 2015E 2016E 2017E

Mobile WiFi Infrastructure


Figure 108: AWSC—CS estimate vs consensus

(NT$ mn) 4Q15E 1Q16E 2015E 2016E


Revenue 1,126 1,241 990 1,194 4,437 4,536 4,916 4,988

Sequential growth -6% 4% -12% -4% 66% 70% 11% 10%

Gross profits 369 414 317 392 1,481 1,519 1,643 1,605

Operating profits 306 362 255 0 1,234 1,263 1,379 1,335

Net profits 263 297 219 277 1,079 1,076 1,174 1,133

EPS (NT$) 1.92 2.17 1.60 2.02 7.83 7.81 8.37 8.07

GM 32.8% 33.4% 32.0% 32.8% 33.4% 33.5% 33.4% 32.2%

OPM 27.2% 29.2% 25.8% 0.0% 27.8% 27.8% 28.1% 26.8%

NM 23.4% 23.9% 22.2% 23.2% 24.3% 23.7% 23.9% 22.7%



We estimate that AWSC will generate NT$1.0-1.3 bn free cash flow, or NT$7.4-9.0 free

cash flow per share over 2015-17, reflecting a free cash yield of 9.4-11.4%. The improving

free cash flow yield vs the past few years is reflecting the top line growth and margin

expansions. Capex is estimated to be NT$360 mn per year over 2015-17, implying 7-8%

capex-to-sales ratio. We estimate the depreciation-to-sales ratio to be around 5-6% in

2015-17, lower than the level in 2011-14 due to significantly higher revenues bases.

Given that AWSC has been able to generate free cash flows, we assume the dividend

payout ratio to be 70% over 2015-17, similar to the level in 2014. Based on our earnings

forecast and dividend payout assumption, we expect to see 6.8-8.3% dividend yield over

2015-17.

11% top line growth for

2016; stable margin on

similar product mix


per share over 2015-17

11 November 2015


Figure 109: Win Semi—cash flow summary (2011-17E)

(NT$ mn) 2011 2012 2013 2014 2015E 2016E 2017E

Revenues 1,651 1,611 1,066 2,669 4,437 4,916 5,456

Capex 344 155 73 318 360 360 360

Capex / Revenues (%) 20.8 9.6 6.9 11.9 8.1 7.3 6.6

Dep and Amort 203 244 235 229 245 266 298

Dep / Revenues (%) 12.3 15.1 22.0 8.6 5.5 5.4 5.5

Operating cash flow 255 292 322 480 1,380 1,501 1,617

Free Cash flow -89 137 248 162 1,020 1,141 1,257

FCF per share -0.7 1.0 1.8 1.2 7.4 8.1 9.0

FCF yield (%) -3.4 4.9 13.7 2.9 9.4 10.3 11.4


Dividend yield (%) 0.0 0.0 0.0 4.8 6.8 7.5 8.3


Initiating coverage with NEUTRAL and NT$88 TP

We initiate coverage on AWSC with a NEUTRAL rating and a target price of NT$88, based

on 11x 2016E EPS. Due to the earnings volatility in the past few years, AWSC does not

have a stable P/E valuation history for reference (only small net profits in 2012 and net

loss in 2013). We thus benchmark our target multiple to its peer Win Semi (we use 12x

P/E for TP), though with a modest discount to factor in the potential single client risk and

the accompanied higher business volatility. The 11x 12M forward P/E is also consistent

with the midpoint of the 4-15x trading range since 2H13. We thus think the shares are

fairly valued at this level.

Downside risks

Downside risks to our NEUTRAL view on ASWC would include:

■ Lower outsourcing ratios from Skyworks amid weaker end demand: As Skyworks also

has internal capacity for the compound semiconductor RF components, we believe

that in the case of weaker end demand, Skyworks will prioritise its own fab utilisation

and cut the outsourcing orders to AWSC. Given that Skyworks accounted for over

80% of AWSC's revenues, we expect this to be one of the major risks to our view.

■ Slower-than-expected LTE migration leading to softer RF component demand: Mobile

RF component accounted for ~80% of its total revenues, so we believe that slower

migration in LTE devices could pose some downside risks to our forecast.

■ Skyworks losing share in the RF component market: Given the heavy reliance on

Skyworks, AWSC would take the hit as well if Skyworks becomes less competitive and

loses shares in the RF component space. With the strong relationship with Skyworks,

it makes AWSC much more difficult to penetrate into other supply chain.

Upside risks

Upside risks to our NEUTRAL view on AWSC would include:

■ More outsourcing orders from Skyworks along with its more disciplined capacity

expansion: as Skyworks has been focusing its internal capacity expansion on filter and

assembly fronts, better-than-expect demand for its GaAs products (PA, LNA, switch,

etc.) could be mostly flowing into AWSC.

■ Better progress in the contribution from new business opportunities: With AWSC

adding one Japanese IDM into its customer base, better progress in this and other

new customers could also pose upside to its business performance.

11 November 2015


Figure 110: AWSC—12-month forward P/E Figure 111: AWSC—12-month forward P/B

0

50

100

150

200

250Se

p-1

1

De

c-1

1

Mar

-12

Jun

-12

Sep

-12

De

c-1

2

Mar

-13

Jun

-13

Sep

-13

De

c-1

3

Mar

-14

Jun

-14

Sep

-14

De

c-1

4

Mar

-15

Jun

-15

Sep

-15

NT$

Share price (NT$) 5x 10x 15x 20x 25x

0

20

40

60

80

100

120

140

Sep

-11

De

c-1

1

Mar

-12

Jun

-12

Sep

-12

De

c-1

2

Mar

-13

Jun

-13

Sep

-13

De

c-1

3

Mar

-14

Jun

-14

Sep

-14

De

c-1

4

Mar

-15

Jun

-15

Sep

-15

NT$

Share price (NT$) 1.0x 2.0x 3.0x 4.0x 5.0x





modelled assumptions, CS HOLT would value AWSC at NT$92.22, similar to our target

price.



Company profile

Company background

Founded in December 1998, AWSC has been focusing on manufacturing compound

semiconductor wafers. It started with a 4" fab and successfully converted all of its capacity

into 6" in 1H12. ~80% of its revenues are from the mobile application, 15-20% from the

WiFi application and the balance from others. It started to provide the foundry service to its

major US customer from 2002. AWSC currently has one fab in operation in Tainan

13,000/month GaAs wafer

capacity

11 November 2015


(Taiwan) with a total capacity of 13,000 GaAs wafers per month. AWSC was listed on the

Emerging Board in March 2003 and on the Taipei Exchange (former OTC) in June 2009.


Win Semi has a fairly diversified shareholder structure, as top ten shareholders only own

15.8% of the company, with management owning 6-7% of the company. The QFII holdings

at AWSC has been rising from ~3% in August 2015 to 9.3% currently.

Figure 113: AWSC—QFII holdings vs share prices Figure 114: AWSC—major shareholders

0

20

40

60

80

100

120

140

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

Jan-12 May-12 Sep-12 Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15

NT

$QFII Holdings Share price (NT$, RHS)

Shareholder Stake

1 Ch'i You Ming 2.8%

2 Taiwan Business Bank & CITC OTC Fund 1.9%

3 Capital Marathon Fund 1.7%

4 Wang Ch'ing Hua 1.6%

5 Ts'ai Wen Hui 1.5%

6 Capital Hi-Tech. Fund 1.4%

7 Yuanta Securities Co., Ltd. 1.4%

8 Labor Pension Fund Supervisory Committee 1.2%

9 Hsieh Hung Chen-A 1.2%

10 Standard Chartered Bank & Europe CREDIT 1.1%


Figure 115: AWSC—management profile


Chairman and COO Yoming Chi Manager at Hughes Aircraft GaAs Operation Manager at Rockwell Semiconductor System

Master of Electrical Engineering from the

University of Pennsylvania

General Manager Chinghua Wang Manager at Hughes Aircraft GaAs Operation Senior Manager at Intel

Ph.D. of Electrical Engineering from the

University of Southern California

Consultant Saoming Chi Consultant at PSI Master of Computer Science from Indiana

University

Manager - R&D Kuojin Huang Senior Engineer at Philips Master of Electrical Engineering from National

Cheng Kung University

Manager - Finance Lewis Liu Accounting Director at Uni-President Finance Manager at Lienhua Mobile

Master of BA from National Cheng Kung

University

Manager - Process Equipment Yongyao Hsu Engineering at Mosel Bachelor of Electronic Engineering from Feng

Chia University

Manager - Quality Assurance Shichieh Hung Engineering at Acer Semiconductor Master of Chemical Engineering from National

Cheng Kung University

Manager - Materials Chiren Chou Procurement Director at Solar Technology Bachelor of International Trade from Chinese

Culture University


11 November 2015


Figure 116: AWSC—key milestones

Tine Milestone

Dec-1998 Founded in Tainan (Taiwan)

Apr-1999 Cleanroom Completed and Certified

Jun-1999 First Yielded HBT Wafer Out

Dec-1999 Delivered First Customer Lot

Apr-2000 Qualified AlGaAs HBT Process

Sep-2000 Developed TWV Foundry Process

Nov-2000 Qualified InGaP HBT Foundry Process

Jan-2001 Delivered InGaP OEIC Wafers to Customers

Mar-2001 HBT Octagonal Layout Patent Granted by Taiwan

May-2001 InGaP OEIC Foundry Process Qualified

Jul-2001 Passed Customer Audit

Dec-2001 New Technology Released to Public

Mar-2002 Passed Skyworks Process Qualification

May-2002 Passed Quality Audit by Skyworks

Jul-2002 Started Production Ramp for Skyworks

Oct-2002 Achieved Production Yield and Cycle Time Targets

Nov-2002 InP Government Project Granted

Jan-2003 Over 1 Million Cellular Power Amplifier Chips Delivered

Feb-2003 Production Ramp Up for WLAN Products

May-2003 Over 2 Million WLAN Power Amplifier Chips Delivered

May-2004 Completed the First Phase of InP Government Project

Jul-2004 Over 10 Million Cellular Power Amplifier Chips Delivered

Jul-2004 Over 15 Million WLAN Power Amplifier Chips Delivered

Aug-2004 Offer Monthly Low NRE Shuttle Mask Services

Sep-2004 Offer pHEMT Switch Process

Dec-2004 Over 37 Million Power Amplifier Chips Delivered

Jan-2005 Offer InP Power Process

Jan-2005 Offer GaAs PIN Diode Process

Jun-2005 Achieve BEP in Month of June

Sep-2005 More than 2,700 Wafers Shipped in One Month

Mar-2006 More than 4,200 Wafers Shipped in One Month

Feb-2007 Completed FAB Expansion; Total Capacity of 1600 4" HBT Wafers

Sep-2008 Completed 6" Pilot Line

Jun-2009 Listed on Taipei Exchange (former OTC)

Sep-2009 Passed first tier customer's 6" product qualification

Apr-2010 Complete 6" FAB Expansion; Total Capacity of 1,600 Wafers in One Month

May-2010 More than 7,000 Wafers Shipped in One Month

Jun-2010 More than 1,300 pHEMT Wafers Shipped in One Month

Sep-2010 Completed 3rd Floor Construction

Jan-2011 Completed 6" GaAs CPV Solar Cell FAB

Apr-2011 Shipped More Than 2,000 6" Wafers in April, 2011

Jul-2011 Qualified BiFET Process

Sep-2011 Shipped More Than 2,000 6" CPV Wafers

Mar-2012 Fully Converted to 6" Wafer Production

Dec-2013 Completed BiHEMT, E/D pHEMT Process Development

Dec-2014 Capacity Up To 3,000 Wafers Out Per Week


11 N

ov

em

ber 2

015

Taiw

an

Co

mp

ou

nd

Sem

ico

nd

ucto

r Secto

r 5

3

Figure 117: AWSC—quarterly income statement Year-end 31 Dec (NT$mn) 1Q13 2Q13 3Q13 4Q13 1Q14 2Q14 3Q14 4Q14 1Q15 2Q15 3Q15 4Q15E 1Q16E 2Q16E 3Q16E 4Q16E 2013 2014 2015E 2016E

Sales 246 350 246 225 268 523 843 1,035 1,016 1,098 1,197 1,126 990 1,334 1,395 1,197 1,066 2,669 4,437 4,916

COGS 228 275 252 212 263 429 665 734 680 721 800 756 673 876 917 807 967 2,092 2,956 3,273

Gross profits 18 75 (7) 12 5 94 178 301 337 377 398 369 317 458 479 390 99 578 1,481 1,643

Total operating exp. 45 45 35 39 37 43 47 54 61 59 64 63 61 68 67 68 165 181 247 264

Operating exp. promotion 3 7 3 4 3 3 4 5 5 5 5 6 7 7 6 7 18 16 19 26

Operating exp. administrative 11 12 10 11 11 12 14 17 16 18 18 18 18 19 17 20 44 54 70 74

Operating expense R&D 31 27 21 24 23 27 29 32 40 37 41 39 37 43 45 41 103 111 157 165

EBIT (28) 30 (41) (27) (32) 51 131 246 276 318 334 306 255 389 412 323 (66) 396 1,234 1,379

Non-oper. income (Loss) 8 2 (4) 4 10 (2) 9 19 (3) (2) 31 (0) (0) (0) (0) (0) 10 38 26 (1)

Interest expenses (1) (1) (1) (1) (1) (1) (1) (0) (0) (0) - (0) (0) (0) (0) (0) (4) (2) (0) (1)

Interest income 1 1 1 1 1 1 1 1 2 2 - - - - - - 4 5 4 -

Other non-op. income/(loss) 8 2 (4) 4 10 (2) 9 18 (4) (5) 31 - - - - - 10 35 23 -

Pre-tax Income (20) 32 (45) (23) (22) 49 140 266 274 316 365 306 255 389 411 322 (56) 434 1,261 1,379

Income taxes exp./(gains) (0) 18 (2) 2 (4) 8 21 34 37 51 50 43 36 66 58 45 18 59 182 205

Net income before extraord. (20) 14 (44) (24) (18) 41 119 232 236 265 314 263 219 323 354 277 (74) 375 1,079 1,174

Net income (20) 14 (44) (24) (18) 41 119 232 236 265 314 263 219 323 354 277 (74) 375 1,079 1,174

EPS (NT$) (0.15) 0.10 (0.32) (0.18) (0.13) 0.30 0.88 1.71 1.73 1.93 2.24 1.88 1.56 2.30 2.52 1.98 (0.55) 2.76 7.83 8.37

Aveage shares 136 136 136 136 136 136 136 136 137 137 140 140 140 140 140 140 136 136 138 140

EBITDA 31 89 18 31 25 108 189 303 335 379 396 370 319 455 479 392 169 625 1,479 1,646

Margins (%)

EBITDA margin 12.7 25.4 7.1 13.9 9.4 20.7 22.4 29.3 33.0 34.5 33.0 32.8 32.2 34.1 34.3 32.8 15.8 23.4 33.3 33.5

Gross margin 7.1 21.5 (2.7) 5.5 1.9 17.9 21.1 29.1 33.1 34.4 33.2 32.8 32.0 34.3 34.3 32.6 9.3 21.6 33.4 33.4

Operating margin (11.3) 8.6 (16.8) (12.0) (12.0) 9.7 15.5 23.8 27.2 29.0 27.9 27.2 25.8 29.2 29.5 27.0 (6.2) 14.8 27.8 28.1

Tax rate 0.2 56.0 3.5 (7.3) 17.8 17.2 14.9 12.6 13.6 16.2 13.8 14.0 14.0 17.0 14.0 14.0 (31.7) 13.6 14.4 14.8

Net margin (8.1) 4.0 (17.8) (10.8) (6.7) 7.8 14.2 22.5 23.3 24.1 26.3 23.4 22.2 24.2 25.4 23.2 (7.0) 14.0 24.3 23.9

QoQ (%)

Sales (26) 43 (30) (9) 19 95 61 23 (2) 8 9 (6) (12) 35 5 (14)

COGS (23) 20 (8) (16) 24 63 55 10 (7) 6 11 (5) (11) 30 5 (12)

Gross profit (54) 329 (109) NM (58) 1,692 90 69 12 12 5 (7) (14) 44 5 (18)

Operating profit NM NM (238) NM NM NM 157 88 12 15 5 (8) (17) 53 6 (22)

Net profit NM NM (413) NM NM NM 192 95 2 12 19 (16) (17) 47 10 (22)

YoY (%)

Sales (14) (29) (51) (32) 9 49 243 360 279 110 42 9 (3) 21 17 6 (34) 150 66 11

COGS 2 (32) (39) (28) 15 56 164 246 158 68 20 3 (1) 22 15 7 (28) 116 41 11

Gross profit (71) (12) (107) (68) (70) 24 NM 2,312 6,345 303 124 23 (6) 21 20 6 (64) 485 156 11

Operating profit (1,754) 30 (253) NM NM 70 NM NM NM 525 155 24 (8) 22 23 5 (263) NM 212 12

Net profit NM (34) (397) NM NM 193 NM NM NM 547 163 13 (7) 22 13 5 (1,560) NM 188 9


11 November 2015


Companies Mentioned (Price as of 10-Nov-2015)

ANADIGICS Inc. (ANAD.OQ, $0.23) Advanced Wireless Semiconductor Company (8086.TWO, NT$78.6, NEUTRAL[V], TP NT$88.0) Alcatel-Lucent (ALUA.PA, €3.78) Analog Devices Inc. (ADI.OQ, $60.96) Apple Inc (AAPL.OQ, $120.57) Avago Technologies Ltd. (AVGO.OQ, $126.5) Coretronic Corp (5371.TWO, NT$32.7) Furukawa Electric (5801.T, ¥246) GCS Holdings (4991.TWO, NT$78.2) HTC Corp (2498.TW, NT$82.6) Hitachi Cable (5812.T^F13, ¥181) Hitachi Cable (5812.T^F13, ¥181) Hitachi Cable (5812.T^F13, ¥181) Hitachi Cable (5812.T^F13, ¥181) IQE (IQE.L, 24.25p) Intel Corp. (INTC.OQ, $33.35) IntelliEPI (4971.TWO, NT$115.0) LG Electronics Inc (066570.KS, W52,500) LPI (2369.TW, NT$10.0) LandMark (3081.TWO, NT$425.0) MACOM (MTSI.OQ, $34.51) Macronix (2337.TW, NT$4.74) Maxtek Tech (3315.TW, NT$16.3) Microchip Technology Inc. (MCHP.OQ, $47.82) Murata Manufacturing (6981.T, ¥19,225) Nokia (NOK.N, $7.43) Philips (PHG.AS, €24.86) QUALCOMM Inc. (QCOM.OQ, $52.94) Qorvo (QRVO.OQ, $54.09) Richwave (4968.TWO, NT$71.53) Samsung Electronics (005930.KS, W1,321,000) Shunsin (6451.TW, NT$125.5) Skyworks Solutns (SWKS.OQ, $84.89) Sony (6758.T, ¥3,463) Sumitomo Chemical (4005.T, ¥684) THEIL (6271.TW, NT$79.7) Taiwan Semiconductor Manufacturing (2330.TW, NT$139.5) Uni-President Enterprises (1216.TW, NT$53.8) United Microelectronics (2303.TW, NT$12.05) Visual Photonics Epitaxy Co., Ltd (2455.TW, NT$36.5, OUTPERFORM, TP NT$47.0) Win Semiconductors Corp (3105.TWO, NT$48.5, OUTPERFORM, TP NT$62.0) ZTE Corporation (0763.HK, HK$18.38)

Disclosure Appendix

Important Global Disclosures

I, Derrick Yang, certify that (1) the views expressed in this report accurately reflect my personal views about all of the subject companies and securities and (2) no part of my compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report.

The analyst(s) responsible for preparing this research report received Compensation that is based upon various factors including Credit Suisse's total revenues, a portion of which are generated by Credit Suisse's investment banking activities

As of December 10, 2012 Analysts’ stock rating are defined as follows:

Outperform (O) : The stock’s total return is expected to outperform the relevant benchmark*over the next 12 months.

Neutral (N) : The stock’s total return is expected to be in line with the relevant benchmark* over the next 12 months.

Underperform (U) : The stock’s total return is expected to underperform the relevant benchmark* over the next 12 months.

*Relevant benchmark by region: As of 10th December 2012, Japanese ratings are based on a stock’s total return relative to the analy st's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms represen ting the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ra tings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representi ng the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin Am erican and non-Japan Asia stocks, ratings are based on a stock’s total return relative to the average total return of the relevant country or regional benchmark; prior to 2nd October 2012 U.S. and Canadian ratings were based on (1) a stock’s absolute total return potential to its current share price and (2) the relative attractiveness of a stock’s total return potential within an analyst’s coverage universe. For Australian and New Zealand stocks, the expected total return (ETR) calculation includes 1 2-month rolling dividend yield. An Outperform rating is assigned where an ETR is greater than or equal to 7.5%; Underperform where an ETR less than or equal to 5%. A Neutral may be assigned where the ETR is between -5% and 15%. The overlapping rating range allows analysts to assign a rating that puts ETR in the context of associated risks. Prior to 18 May 2015, ETR ranges for Outperform and Underperform ratings did not overlap with Neutral thresholds between 15% and 7.5%, wh ich was in operation from 7 July 2011.

11 November 2015


Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances.

Volatility Indicator [V] : A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward.

Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation:

Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months.

Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months.

Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months.

*An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cover multiple sectors.

Credit Suisse's distribution of stock ratings (and banking clients) is:

Global Ratings Distribution

Rating Versus universe (%) Of which banking clients (%)

Outperform/Buy* 59% (34% banking clients)

Neutral/Hold* 26% (35% banking clients)

Underperform/Sell* 13% (23% banking clients)

Restricted 2%

*For purposes of the NYSE and NASD ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, an d Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdin gs, and other individual factors.

Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein.

Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: http://www.csfb.com/research-and-analytics/disclaimer/managing_conflicts_disclaimer.html

Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties.

Price Target: (12 months) for Advanced Wireless Semiconductor Company (8086.TWO)

Method: Our target price of NT$88 for Advanced Wireless Semiconductor Company is based on 11x 2016E EPS (earnings per share) vs the historical trading range of 4-15x since 2H13. Due to the earnings volatility in the past few years, AWSC does not have a stable P/E valuation history for reference. We thus benchmark our target multiple to its peer Win Semi (we use 12x P/E for TP), though with a modest discount to factor in the potential single client risk and higher business volatility.

Risk: Risks that could cause the share price to diverge from our NT$88 target price for ASWC include: (1) lower outsourcing ratios from Skyworks amid weaker end demand, (2) slower than expected LTE migration leading to softer RF component demand, (3) Skyworks losing share in the RF component market. Upside risks include: (1) More outsourcing orders from Skyworks along with its more disciplined capacity expansion; (2) Better progress in the contribution from new business opportunities.

Price Target: (12 months) for Visual Photonics Epitaxy Co., Ltd (2455.TW)

Method: Our target price of NT$47 for VPEC is based on 18x 2016E EPS (earnings per share), vs the historical average of 18-19x and the range of 4-30x 12M forward P/E (price-to-earnings).

Risk: Risks that could impede achievement of our target price of NT$47 for VPEC include: (1) weaker growth in the RF component space, (2) slower traction in the optical communication expansion, (3) higher-than-expected adoption of the silicon-based RF components impacting the total addressable market.

Price Target: (12 months) for Win Semiconductors Corp (3105.TWO)

Method: Our target price of NT$62 for Win Semiconductors Corp is based on 12x 2016E EPS (earnings per share), vs the historical average of 12x and the range of 6-20x 12M forward P/E (price-to-earnings).

11 November 2015


Risk: Risks that could impede achievement of our NT$62 target price for Win Semi include: (1) slower shipment growth in the LTE-enabled devices, (2) lower capex among telecom operators for LTE infrastructure due to macro uncertainties, (3) weaker WiFi demand on delayed spec upgrade to 802.1ac and softer IoT device proliferation.

Please refer to the firm's disclosure website at https://rave.credit-suisse.com/disclosures for the definitions of abbreviations typically used in the target price method and risk sections.

See the Companies Mentioned section for full company names

The subject company (3105.TWO, AVGO.OQ, ADI.OQ, MCHP.OQ, 2330.TW, AAPL.OQ, 0763.HK, NOK.N, ALUA.PA, 2498.TW, 005930.KS, 066570.KS, 2303.TW, INTC.OQ, 1216.TW) currently is, or was during the 12-month period preceding the date of distribution of this report, a client of Credit Suisse.

Credit Suisse provided investment banking services to the subject company (AVGO.OQ, ADI.OQ, MCHP.OQ, AAPL.OQ, NOK.N, ALUA.PA, 2303.TW, INTC.OQ) within the past 12 months.

Credit Suisse provided non-investment banking services to the subject company (INTC.OQ) within the past 12 months

Credit Suisse has managed or co-managed a public offering of securities for the subject company (AAPL.OQ, 2303.TW) within the past 12 months.

Credit Suisse has received investment banking related compensation from the subject company (AVGO.OQ, ADI.OQ, MCHP.OQ, AAPL.OQ, NOK.N, ALUA.PA, 2303.TW, INTC.OQ) within the past 12 months

Credit Suisse expects to receive or intends to seek investment banking related compensation from the subject company (3105.TWO, AVGO.OQ, 6981.T, ADI.OQ, MCHP.OQ, 4005.T, 5801.T, QCOM.OQ, AAPL.OQ, NOK.N, ALUA.PA, 2498.TW, 005930.KS, 066570.KS, 2303.TW, INTC.OQ, 1216.TW) within the next 3 months.

Credit Suisse has received compensation for products and services other than investment banking services from the subject company (INTC.OQ) within the past 12 months

As of the date of this report, Credit Suisse makes a market in the following subject companies (MCHP.OQ, QCOM.OQ, AAPL.OQ, 6758.T, INTC.OQ).

As of the end of the preceding month, Credit Suisse beneficially own 1% or more of a class of common equity securities of (2330.TW, 0763.HK, ALUA.PA, 2498.TW, 2303.TW, 1216.TW).

Credit Suisse has a material conflict of interest with the subject company (AVGO.OQ) . Credit Suisse is acting as financial advisor to Avago Technologies Ltd (AVGN) on its announced potential acquisition of Broadcom Corp. (BRCM).

Credit Suisse has a material conflict of interest with the subject company (MCHP.OQ) . Credit Suisse is acting as financial advisor to Micrel Inc. (MCRL) in relation to its proposed sale to Microchip Technology Inc. (MCHP).

Credit Suisse has a material conflict of interest with the subject company (2330.TW) . Credit Suisse is acting as the financial advisor to Motech Industries Inc in relation to the share subscription by Taiwan Semiconductor Manufacturing Co., Ltd.

Credit Suisse has a material conflict of interest with the subject company (005930.KS) . Credit Suisse is acting as exclusive financial advisor to Samsung Electronics and Samsung Fine Chemicals in relation to the proposed sale of their ownership stakes in the semiconductor wafer joint ventures with SunEdison, SMP Ltd and MEMC Korea Company Ltd, to SunEdison.

Credit Suisse has a material conflict of interest with the subject company (INTC.OQ) . Credit Suisse Securities (USA) LLC is acting as financial advisor to Intel Corp (INTL) on its announced proposed acquisition of LSI’s Axxia Networking Business from Avago Technologies Limited (AVGO).

As of the date of this report, an analyst involved in the preparation of this report has the following material conflict of interest with the subject company (AAPL.OQ). A Credit Suisse analyst involved in the preparation of this report has a long position in the common stock of AAPL.

For other important disclosures concerning companies featured in this report, including price charts, please visit the website at https://rave.credit-suisse.com/disclosures or call +1 (877) 291-2683.

Important Regional Disclosures

Singapore recipients should contact Credit Suisse AG, Singapore Branch for any matters arising from this research report.

The analyst(s) involved in the preparation of this report may participate in events hosted by the subject company, including site visits. Credit Suisse does not accept or permit analysts to accept payment or reimbursement for travel expenses associated with these events.

Restrictions on certain Canadian securities are indicated by the following abbreviations: NVS--Non-Voting shares; RVS--Restricted Voting Shares; SVS--Subordinate Voting Shares.

Individuals receiving this report from a Canadian investment dealer that is not affiliated with Credit Suisse should be advised that this report may not contain regulatory disclosures the non-affiliated Canadian investment dealer would be required to make if this were its own report.

For Credit Suisse Securities (Canada), Inc.'s policies and procedures regarding the dissemination of equity research, please visit https://www.credit-suisse.com/sites/disclaimers-ib/en/canada-research-policy.html.

The following disclosed European company/ies have estimates that comply with IFRS: (ANAD.OQ, ALUA.PA, PHG.AS).

11 November 2015


Credit Suisse has acted as lead manager or syndicate member in a public offering of securities for the subject company (ADI.OQ, AAPL.OQ, ALUA.PA, 2303.TW, 1216.TW) within the past 3 years.

As of the date of this report, Credit Suisse acts as a market maker or liquidity provider in the equities securities that are the subject of this report.

Principal is not guaranteed in the case of equities because equity prices are variable.

Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that.

Taiwanese Disclosures: This research report is for reference only. Investors should carefully consider their own investment risk. Investment results are the responsibility of the individual investor. Reports may not be reprinted without permission of CS. Reports written by Taiwan based analysts on non-Taiwan listed companies are not considered recommendations to buy or sell securities under Taiwan Stock Exchange Operational Regulations Governing Securities Firms Recommending Trades in Securities to Customers.

To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important disclosures regarding any non-U.S. analyst contributors: The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts listed below may not be associated persons of CSSU and therefore may not be subject to the NASD Rule 2711 and NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account.

Credit Suisse AG, Taipei Securities Branch ........................................................................................................................ Derrick Yang ; Jerry Su

Important Credit Suisse HOLT Disclosures

With respect to the analysis in this report based on the Credit Suisse HOLT methodology, Credit Suisse certifies that (1) the views expressed in this report accurately reflect the Credit Suisse HOLT methodology and (2) no part of the Firm’s compensation was, is, or will be directly related to the specific views disclosed in this report.

The Credit Suisse HOLT methodology does not assign ratings to a security. It is an analytical tool that involves use of a set of proprietary quantitative algorithms and warranted value calculations, collectively called the Credit Suisse HOLT valuation model, that are consistently applied to all the companies included in its database. Third-party data (including consensus earnings estimates) are systematically translated into a number of default algorithms available in the Credit Suisse HOLT valuation model. The source financial statement, pricing, and earnings data provided by outside data vendors are subject to quality control and may also be adjusted to more closely measure the underlying economics of firm performance. The adjustments provide consistency when analyzing a single company across time, or analyzing multiple companies across industries or national borders. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes the baseline valuation for a security, and a user then may adjust the default variables to produce alternative scenarios, any of which could occur.

Additional information about the Credit Suisse HOLT methodology is available on request.

The Credit Suisse HOLT methodology does not assign a price target to a security. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes a warranted price for a security, and as the third-party data are updated, the warranted price may also change. The default variable may also be adjusted to produce alternative warranted prices, any of which could occur.

CFROI®, HOLT, HOLTfolio, ValueSearch, AggreGator, Signal Flag and “Powered by HOLT” are trademarks or service marks or registered trademarks or registered service marks of Credit Suisse or its affiliates in the United States and other countries. HOLT is a corporate performance and valuation advisory service of Credit Suisse.

For Credit Suisse disclosure information on other companies mentioned in this report, please visit the website at https://rave.credit-suisse.com/disclosures or call +1 (877) 291-2683.

11 November 2015


References in this report to Credit Suisse include all of the subsidiaries and affiliates of Credit Suisse operating under its investment banking division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who-we-are This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse AG or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates. The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk. This report is issued and distributed in Europe (except Switzerland) by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. This report is issued and distributed in Europe (except Switzerland) by Credit Suisse International, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. This report is being distributed in Germany by Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). This report is being distributed in the United States and Canada by Credit Suisse Securities (USA) LLC; in Switzerland by Credit Suisse AG; in Brazil by Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; in Mexico by Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); in Japan by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau (Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; elsewhere in Asia/ Pacific by whichever of the following is the appropriately authorised entity in the relevant jurisdiction: Credit Suisse (Hong Kong) Limited, Credit Suisse Equities (Australia) Limited, Credit Suisse Securities (Thailand) Limited, regulated by the Office of the Securities and Exchange Commission, Thailand, having registered address at 990 Abdulrahim Place, 27th Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok 10500, Thailand, Tel. +66 2614 6000, Credit Suisse Securities (Malaysia) Sdn Bhd, Credit Suisse AG, Singapore Branch, Credit Suisse Securities (India) Private Limited (CIN no. U67120MH1996PTC104392) regulated by the Securities and Exchange Board of India as Research Analyst (registration no. INH 000001030) and as Stock Broker (registration no. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777, Credit Suisse Securities (Europe) Limited, Seoul Branch, Credit Suisse AG, Taipei Securities Branch, PT Credit Suisse Securities Indonesia, Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore branch to overseas investors (as defined under the Financial Advisers Regulations). By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the "FAA"), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore branch may provide to you. This information is being distributed by Credit Suisse AG, Dubai Branch, duly licensed and regulated by the Dubai Financial Services Authority (DFSA), and is directed at Professional Clients or Market Counterparties only, as defined by the DFSA. The financial products or financial services to which the information relates will only be made available to a client who meets the regulatory criteria to be a Professional Client or Market Counterparty only, as defined by the DFSA, and is not intended for any other person. This research may not conform to Canadian disclosure requirements. In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-U.S. customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. U.S. customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the U.S. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority or in respect of which the protections of the Prudential Regulation Authority and Financial Conduct Authority for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case are available upon request in respect of this report. CS may provide various services to US municipal entities or obligated persons ("municipalities"), including suggesting individual transactions or trades and entering into such transactions. Any services CS provides to municipalities are not viewed as "advice" within the meaning of Section 975 of the Dodd-Frank Wall Street Reform and Consumer Protection Act. CS is providing any such services and related information solely on an arm's length basis and not as an advisor or fiduciary to the municipality. In connection with the provision of the any such services, there is no agreement, direct or indirect, between any municipality (including the officials, management, employees or agents thereof) and CS for CS to provide advice to the municipality. Municipalities should consult with their financial, accounting and legal advisors regarding any such services provided by CS. In addition, CS is not acting for direct or indirect compensation to solicit the municipality on behalf of an unaffiliated broker, dealer, municipal securities dealer, municipal advisor, or investment adviser for the purpose of obtaining or retaining an engagement by the municipality for or in connection with Municipal Financial Products, the issuance of municipal securities, or of an investment adviser to provide investment advisory services to or on behalf of the municipality. If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice by Credit Suisse to the clients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any liability whatsoever for any direct or consequential loss arising from their use of this report or its content. Principal is not guaranteed. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that.

Copyright © 2015 CREDIT SUISSE AG and/or its affiliates. All rights reserved.

Investment principal on bonds can be eroded depending on sale price or market price. In addition, there are bonds on which investment principal can be eroded due to changes in redemption amounts. Care is required when investing in such instruments. When you purchase non-listed Japanese fixed income securities (Japanese government bonds, Japanese municipal bonds, Japanese government guaranteed bonds, Japanese corporate bonds) from CS as a seller, you will be requested to pay the purchase price only.

TW GaAs sector_initiation_post panel.doc

https://www.credit-suisse.com/who-we-are

Taiwan Compound Semiconductor Sector

Documents

Transcript of Taiwan Compound Semiconductor Sector