10 March 2017 Asia Pacific/Japan

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

10 March 2017 Asia Pacific/Japan Equity Research

Technology

Technology sector Research Analysts

Hideyuki Maekawa

81 3 4550 9723

[email protected]

Akinori Kanemoto

81 3 4550 7363

[email protected]

Mika Nishimura

81 3 4550 7369

[email protected]

COMMENT

Asia feedback (Semiconductor/SPE): mounting risks

■ Summary: As noted in our latest hardware outlook based on our Asia feedback (Hardware): Spring hasn't yet come published on 10 March, excessive

inventory in PCs and some Chinese smartphones are leading to cuts in production plans. We think the startup for new Chinese smartphones in spring and beyond will lag versus expectations, and note a component glut in the supply chain due to production plan slippages triggered by application processor (AP) supply delays. As we anticipated, risk appears to have mounted for a cutback in semiconductor production in the second half of 2Q 2017. There appears to be some excess inventory in both PC DRAM and mobile RAM, the supply–demand balance for which was tight previously, and a glut due to cuts in demand projections, so we anticipate a peak-out in DRAM prices. With the predominant outlook pointing to two new iPhone models, one with an increase to 64GB and the other with 256GB, we continue to look for an increase in installed NAND density on smartphones. In addition, enterprise demand appears to be robust; as this could lead to a sustained tightness in NAND supply–demand, we anticipate ongoing price increases not only in 1Q but also 2Q. As for semiconductor capex, Taiwan foundries are yet to start 7nm node mass production investment; we thus note a muted equipment order trend than anticipated previously. Meanwhile, 3D NAND investment appears to be firm, with concurrent development in multiple projects. To sum up, we see a limited number of positives, but expect mounting risks in semiconductor production, DRAM prices, and foundry SPE orders.

■ Key takeaways from our channel check: (1) Taiwan foundries’ 300mm

capacity utilization remains high at 90-95%. No change in plans to fully utilize the 28nm process, which is used mainly on Chinese smartphones. (2) The predominant outlook points to two new iPhone models, one with an increase to 64GB in installed NAND density and the other with 256GB. (3) An outlook for NAND price increases of 5–30% in 1Q and 5–20% in 2Q for all applications. (4) In terms of 64-layer 3D NAND development, SSD sample shipments are one to two months behind schedule, even at the leading makers. (5) With PC DRAM inventory above warehouse capacity in some cases, Chinese smartphone makers have also revised down 1H 2017 demand projections. (6) A sharp rise in server DRAM demand in tandem with the change in CPU platform and the resultant shift from 16GB to 32GB modules. (7) Concrete negotiations yet to begin on Taiwan foundries’ 7nm mass production. (8) Some initial inquiries for Chinese 3D NAND; technology appears to be Korean in origin. (9) Request for the postponement of delivery dates by Taiwanese OSATs due to delays in the startup of new APs. (10) After the new iPhone models, Chinese smartphones now looking at iris scanner; this could lead to back-end SPE (dicer) demand.

■ Stock calls: We forecast lackluster performance across the technology sector

due to mounting risks of a correction in semiconductors. We expect a correction in components to play the role of a catalyst. As for stocks, we downgrade Tokyo Electron and Hitachi Kokusai Electric to NEUTRAL, as the stocks lack catalysts for SPE orders. Once risks have played out and share prices have corrected, we see investors favoring back-end SPEs (Disco) with a good investment theme. Toshiba’s NAND business remains brisk, and we believe business conditions continue to be favorable for inviting bids for its memory business.

https://plus.credit-suisse.com/r/V65luV2AD-WEp6sJ

https://plus.credit-suisse.com/r/V65luV2AD-WEp6sJ

10 March 2017

Technology sector 2

Executive summary and investment implications

Semiconductors: Operating rates still high at Taiwan

foundries; DRAM supply–demand varies by

application/customer, but some inventories appear

excessive; further tightness in NAND supply

The leading foundry and LSI makers operating at full capacity; corrections begin at mid-size foundries

Since our November survey, we note signs of a rebound in Taiwan foundries’ 300mm

capacity utilization (input basis), one of the advance indicators we use for technology

stocks. Operating rates, having reached 90-95% for overall, and 28nm process mainly for

smartphone is 100% recently, are likely to remain high heading into 2Q. However, the risk

of inventory adjustments appears to have escalated due to sluggish PC production in 1Q,

Chinese smartphone inventory adjustments, and delays in new product launches initially

scheduled for 2Q. We view the current 100% operating rates and end-product production

trends as unbalanced.

Also, given swings in the market share of their customers, we think it is rather odd the

leading TSMC and Samsung Electronics are operating at full capacity. Considering

(1) lower volume on the 28nm node line due to Qualcomm’s shift to the 14nm node, (2) a

change in the 14nm node’s foundry from TSMC to Samsung, and (3) market share swings

in Chinese smartphones (from Mediatek to Qualcomm), it would be normal for TSMC to

cut back production, but the 28nm node line is currently operating at full capacity and

some observers have even mentioned increased output. Meanwhile, rivals such as UMC

and SMIC have faced corrections in their 300mm operating rates since the start of 2017,

which suggests a potential for market share gains by TSMC. However, we believe overall

demand is in a downtrend, delaying AP shipment for new products and see inventory

adjustments as highly likely unless Chinese smartphones register a seasonal recovery.

Also, the tightness in the discrete device demand-supply balance reported in our previous

survey is persisting. We also note negotiations for raising price in some products.

PC DRAM and some mobile DRAM inventories in over-supply; server DRAM in short supply

As we anticipated in our previous survey, the DRAM supply–demand balance began to

crumble starting in the latter part of 1Q 2017. For some manufacturers, PC DRAM

inventory has risen to fill the available warehouse space. We also note some glut in mobile

RAM due to twin impacts from Chinese smartphone inventory adjustments and delays in

new product launches scheduled for 2Q. Meanwhile, server DRAM demand has shifted

from 16GB to 32GB modules in tandem with the shipment of Intel’s Skylake CPU platform

and we note supply shortages due to a sharp rise in this demand. There appears to be

some variation in supply–demand conditions by application and customer, but overall the

supply is currently unbalanced. Despite of excessive inventory at downstream, as DRAM

maker inventories and market inventories are still low, we see little risk of a price

deterioration and still see room for a 10% increase in DRAM prices heading into 2Q.

Ongoing NAND supply tightness; prospects for raising price again in 2Q

NAND supply tightness has continued in place due to reduced 2D NAND supplies.

Increases in installed NAND density on smartphones and data center-use SSDs are

driving NAND demand. Prices have risen 5–30% QoQ in 1Q and should rise another 5–

20% in 2Q.

10 March 2017

Technology sector 3

The next generation iPhone 8 is likely to come in 64GB and 256GB NAND density

configurations (we had expected a 512GB version, but we believe this is yet to be

confirmed). With the majority of high-end Chinese smartphones equipped with 128GB, we

note a move from 32GB to 64GB on mid-range and low-end smartphones and anticipate

ongoing bit growth in smartphones in 2H 2017 (we previously forecast smartphone bit

growth to underperform NAND bit growth overall).

We expect 3D NAND adoption to increase with the spring model release of Chinese

smartphones and look for it to become mainstream in the autumn models. In the spring

models, supplies will be limited to that from Samsung, but we see prospects for Toshiba to

begin supplies for the autumn models. In addition to Samsung, Micron appears to be

moving towards approval of cSSDs for PCs and eSSDs for servers. Since Toshiba is

counting on 64-layer NAND, it plans to submit cSSD samples in 2Q. Overall, we note a

delay of one to two months in 64-layer NAND sample shipments at companies and expect

adoption to be limited to 48-layer NAND (32-layer NAND at some makers) for SSD

applications in 2017. Also, with regard to mass production of 64-layer and 72-layer NAND,

both Samsung and Toshiba have established a batch etching process; but SK Hynix and

Micron likely to use a two-pass etching process on two stacks of 32 and 36 layers with

inferior technical and cost performance.

Investment implications: Risk of a growing likelihood of semiconductor inventory adjustment across the technology sector; DRAM stocks bear close monitoring; continued supply–demand tightness in NAND stocks

Growing inventory adjustment risk at Taiwanese foundries is a negative for all technology

sector stocks. A QoQ sales decline heading into 2Q at Taiwanese foundries and

corrections in Chinese smartphones represent the consensus view in Asian equity

markets. However, until the severity of the corrections has been ascertained, we think it

will be difficult to regard the onset of inventory adjustments as the end of negative news.

For Japanese stocks, we do not specifically incorporate the risk of an adjustment in the

semiconductor inventory cycle, but if Taiwanese foundries pare back inventory,

subcontractors whose business is closely tied to the operating rates of Taiwanese

foundries will bear the largest risk of inventory adjustments.

The DRAM supply–demand situation varies considerably by application and customer, but

we do not see the current conditions as something that could trigger a major price collapse

overall. We also see no particular negative impact for the equity markets due to DRAM

prices. However, the market view, which has been overwhelmingly bullish on DRAM prices

until now, is getting increasingly divided. We believe this factor could restrict share price

upside.

We also note instances of NAND prices raised heading into 2Q due to further supply–

demand tightness. This presents a never-again opportunity for Toshiba’s memory

business (whose sale is currently under competitive bidding) to be assessed at the

maximum value.

10 March 2017

Technology sector 4

SPE: Foundry investments pause for front-end SPE,

but 3D NAND carries on. Initial inquiries from China

memory-related. Corrections in smartphone-related

orders for back-end SPE, but investment in

fingerprint/iris sensors

Front-end SPE: Drop-off period in Taiwan foundry investment before 7nm process investment; companies roll-out 3D NAND investment; a major inquiry from Chinese NAND

Orders linked to TSMC’s 10nm node mass production line capex, which have continued

without a break since March 2016, are almost at an end and we expect deliveries to be

completed within 1H 2017. We note investment in 7nm pilot lines, but have yet to see a

concrete plan for establishing a mass production line. However, as Taiwanese foundry

makers are planning to proactively invest in the 7nm process, we regard the present

merely as a drop-off period in business discussions.

Samsung is continuing investment in new 3D NAND capacity at the Pyeongtaek plant, but

we also note investment by Toshiba, SK Hynix, and Intel Dalian. Spending appears to be

steady and there are no major companies' planning changes here since our previous

survey. We understand that there are initial equipment inquiries related to Chinese NAND

investment and expect concrete negotiations to begin soon.

We have yet to confirm a proactive stance on DRAM investment as companies seem

cognizant of the risk that the startup for leading-edge processes may not be as swift as in

the past.

Judging from trends in package substrate material trends, we believe Intel’s 10nm process

start-up may have been delayed as we had anticipated; there is even the likelihood of the

third generation 14nm process (tick-tock-tock or perhaps tick-tock-tock-tock) being

prioritized.

Back-end SPE: Solid investment in fingerprint/iris sensors

While there is a seasonal rebound overall, we note requests for postponement of delivery

dates by OSATs due to delays in the start-up of new smartphone application processors

(APs), making market trends look mixed. However, as seen in 2016, fingerprint sensor

investment continues to track at high levels. Investment in iris sensors, a function we

expect on the 2017 model of the leading North American smartphone maker, has

commenced in 1Q 2017. Also, Chinese LED-related investment appears to have resumed

after a gap of a few years.

In FO-WLP/PLP-related investment, development is currently underway at different

companies, but we understand that there are no major investments for APs due in 2017 (in

line with our previous outlook). Korea’s SEMCO plans to invest in a new FO-PLP line in

late-3Q 2017 for PMICs. We note moves on a new 450mm development project in the

second-generation InFO-WLP (rolled out at AP mass production). In FO-PLP

development, OSATs and Taiwanese LCD panel makers have apparently begun

discussions with equipment makers as 3.5G line equipment can be used by both.

Investment implications: near-term preference order: memory front-end SPE, assembly back-end SPE, tester back-end SPE, foundry front-end SPE

We think there is a high risk of semiconductor inventory adjustments that would indirectly

limit upside for SPE-related stocks. Judging by current capex trends, business conditions

are not conducive to a major order recovery (except along specific themes) in either front-

end or back-end SPE. Overall, we see little reason to invest, although based on the

10 March 2017

Technology sector 5

business climate, our preference order would be memory front-end SPE (Tokyo Electron,

Hitachi Kokusai Electric), assembly back-end SPE (Disco), tester back-end SPE

(Advantest: 6857), and foundry front-end SPE (SCREEN: 7735).

We downgrade Tokyo Electron and Hitachi Kokusai from Outperform to NEUTRAL partly

due to a lack of new catalysts.

Back-end SPE stocks with promising company-specific themes would likely appeal after a

correction triggered by semiconductor inventory adjustment.

Figure 1: Semiconductor/SPE valuations (as of 9 March)

Source: Credit Suisse estimates, Thomson Reuters Datastream

Closing Price: 03/09/17Closing

Price:

Mkt Price EPS

Cap Cur TP % to TP P/E, x

3/16 3/17 3/18 3/16 3/17 3/18 3/17 3/17 3/17 3/17 3/17 3/17

A E1 E2 A E1 E2 % % x x x %

Semiconductor

Toshiba 6502 OUTPERFORM 867 205 460 124.6 -108.6 48.8 47.7 -2.0 4.2 4.3 0.0 47.9 1.6 0.3 3.4 96.7 -14.8 -56.0 0.6 -17.4 -57.9 -16.1

SPE

Hitachi Kokusai Electric 6756 NEUTRAL 268 2,608 2,560 -1.8 126.5 89.6 129.5 10.7 29.1 20.1 1.1 9.3 2.6 1.3 10.9 -50.4 4.9 21.1 109.6 2.3 19.2 93.0

Tokyo Electron 8035 NEUTRAL 1,906 11,615 10,500 -9.6 461.1 612.6 918.6 15.9 19.0 12.6 2.6 17.1 3.1 2.2 10.2 -39.5 2.9 13.8 59.5 0.3 11.9 42.8

SCREEN Holdings 7735 NEUTRAL 358 7,590 7,380 -2.8 79.4 548.2 651.6 56.1 13.8 11.6 1.0 19.5 2.4 1.2 9.5 -13.6 7.7 19.0 79.0 5.1 17.0 62.3

Disco 6146 NEUTRAL 635 17,720 16,000 -9.7 646.1 672.3 742.3 14.8 26.4 23.9 2.0 13.8 3.5 4.1 14.2 -46.2 9.2 35.1 75.4 6.7 33.1 58.8

Advantest 6857 UNDERPERFORM 347 1,978 1,390 -29.7 38.3 69.8 72.6 27.1 28.4 27.2 1.2 12.2 3.2 1.8 14.4 -54.8 -2.3 26.4 87.5 -4.8 24.5 70.8

JEOL 6951 OUTPERFORM 56 576 510 -11.5 42.3 -11.9 9.3 13.5 -48.3 62.2 1.2 -3.8 1.9 0.7 24.7 39.3 3.6 15.7 13.8 1.1 13.7 -2.9

NuFlare Technology 6256 NEUTRAL 86 7,190 5,820 -19.1 767.7 803.2 259.4 6.5 9.0 27.7 1.7 17.0 1.4 1.7 5.4 -1.3 12.2 27.9 35.2 9.6 26.0 18.5

Lasertec 6920 NEUTRAL 70 2,951 1,710 -42.1 143.2 134.2 136.6 8.7 22.0 21.6 1.6 13.3 2.8 3.3 13.1 -26.9 10.4 47.3 92.5 7.9 45.4 75.8

TOWEA 6315 OUTPERFORM 49 1,939 1,680 -13.4 71.6 129.9 173.8 10.0 14.9 11.2 0.5 14.4 2.1 1.7 8.7 -11.3 20.1 47.9 166.7 17.5 46.0 150.0

Micronics Japan 6871 UNDERPERFORM 39 962 775 -19.4 144.7 13.2 27.6 7.5 72.7 34.8 1.6 2.4 1.8 1.1 9.0 -45.5 2.8 -10.9 -9.2 0.2 -12.8 -25.8

Tokyo Seimitsu 7729 NEUTRAL 152 3,675 2,740 -25.4 234.6 223.4 224.7 9.4 16.5 16.4 1.8 11.3 1.8 1.7 7.9 -37.4 -2.1 15.2 74.3 -4.7 13.3 57.6

Company

3M 12M

Relative performance

%

1M 3M 12M 1M

EV /

SalesCS EPS

JPY, bn JPY JPY %

EV /

EBITDAROE PBR

Net

D/E

Absolute

performance %Code Rating

Dvd

Yld

10 March 2017

Technology sector 6

Figure 2: Japanese technology stocks' performance over the past six

months

Source: Thomson Reuters

-39.0%-11.0%

-4.2%-4.1%-2.5%

0.0%1.1%2.8%2.9%3.0%4.2%4.4%5.0%5.6%10.5%12.1%13.9%15.7%16.1%16.4%17.4%18.7%19.2%20.3%21.2%23.1%23.2%23.3%24.7%25.0%26.2%26.5%27.0%27.6%27.6%28.1%28.9%29.0%29.1%29.3%29.5%32.2%33.7%34.5%35.8%35.8%36.2%37.1%37.7%38.1%43.5%43.9%44.2%45.5%48.1%49.4%49.9%50.9%55.7%

69.9%165.7%

-100% 0% 100% 200%

TOSHIBAMEDIATEK

QUALCOMMPIONEER

JAPAN AVNS.ELTN.IND.SANDISK

INTELTAIWAN SEMICON.MNFG.

MITSUMINASDAQ

SONYNECTDK

CASIO COMPUTERNIKON

CIRRUS LOGICTOPIX ELECTRIC APPLIANCES

NIKKEI 225 STOCK AVERAGEJVC KENWOOD

NIDECSCREEN HOLDINGS

TEXAS INSTRUMENTSNIKKEI 225

HIROSEPANASONIC

MURATASK HYNIX

FUJITSUASML HOLDING

KYOCERATOKYO ELECTRONNISSHA PRINTING

QORVOSOXX

SAMSUNG ELECTRONICSMITSUBISHI ELECTRIC

LAM RESEARCHHITACHI

ALPINE ELECTRONICSSEAGATE TECH.APPLIED MATS.

KLA TENCORCITIZEN HDG.

APPLEADVANTEST

SHINKOIBIDEN

CLARIONTAIYO YUDEN

BROADCOMNGK SPARK PLUG

WESTERN DIGITALTERADYNE

SKYWORKS SOLUTIONSHITACHI KOKUSAI

DISCOSEIKO HOLDINGS

ALPSWACOM

JAPAN DISPLAYSHARP

6 month stock returns

10 March 2017

Technology sector 7

Semiconductor market outlook

Logic makers, foundries, CIS

Key takeaways

■ We understand inventory adjustment at Chinese smartphone makers (risks include

Qualcomm/Mediatek's APs, Omnivision's CIS) had no impact on semiconductor

production. Inventories of Qualcomm's MSM8996 AP for high-end smartphones were

excessive and production was adjusted, but there were no changes in production of

semiconductors for mid-range and low-end smartphones.

■ Capacity utilization at Taiwanese foundries (300mm input basis) remained high at 90-

95%. We expect a slight decline in utilization in 2Q but it should remain high. We

understand the largest maker won share away from the mid-tier manufacturer, which

adjusted capacity utilization from February.

■ TSMC continued to operate at full capacity in 28nm processes, which are heavily

weighted toward smartphones. However, utilization in 16nm/40nm processes is

expected to decline slightly heading into 2Q.

■ Korean LSI makers have been operating at full capacity at 300mm, input recently

having been very strong at 110% of production capacity. Qualcomm commenced

inventory adjustment in January, but this was offset by firm demand for 14nm

processes (for the company's own smartphones) and the take-off of 10nm process

production.

■ Mass-production at leading-edge 10nm processes is due to start in March for the new

iPhone while small-lot mass-production for the Galaxy S8 is due to start in 4Q. We

estimate AP yields using 10nm processes for the Galaxy S8 have been tracking at

around 40–50%.

■ Mediatek's Helio P35/X35 (10nm process) has been cancelled. The P30 is due to be

developed using 12nm processes (improved 16nm).

■ Korean LSI makers plan to expand new lines for 10nm processes and CIS production

capacity.

■ Delays in Intel's CPU technology roadmap. Selection of the substrate material for 10nm

processes was also part of the original 1Q 2017 mass-production plan, but we gather

no progress has yet been made. Conversely, we understand there is developing third-

generation materials for 14nm processes.

■ With regard to high-quality raw materials used by logic makers and foundries, there

have been production adjustments at midsize foundries, but we gather there are no

plans to reduce utilization in line with demand to build up raw material inventories. We

understand raw materials supply–demand has remained tight.

Utilization at Taiwanese foundries: Plan to maintain high capacity utilization

TSMC’ capacity utilization at 300mm (figure 3) remained high in 1Q at 90-95%. We gather

they expect no major adjustments in 2Q, utilization remaining high.

However, capacity utilization at UMC/SMIC, especially those using 28/40nm processes,

has started to decline by 10ppt or more, which we mainly attribute to corrections in

demand and market share losses. However, TSMC plans to maintain high capacity

utilization, helped by market share gains.

Taiwan foundries still at full tilt in 28nm

processes (largely smartphones); mid-tier

foundries starting to adjust

10 March 2017

Technology sector 8

By process, capacity utilization in 28nm processes (mainly Chinese smartphone makers)

is full operation in 1Q and 2Q. Utilization of production capacity in 16nm processes

declined due to cuts in iPhone 7 production, while in 40nm and wider processes, foundry

makers' production plans call for a slight adjustment in 2Q (figure 5). However, we think

there is probably risk of production adjustments in areas such as OmniVision's CIS,

production of which was ramped up on expectations of sales to vivo, which started

inventory adjustment, and APs for Chinese smartphones (mainly manufactured using

28nm processes).

We continue to forecast risk of production adjustments at foundries from mid-2Q as we

have for some time. However, we now confirm new concerns, including the impact of

higher bill of materials (BOM) costs on hardware sales prices, postponement of new

Chinese smartphone products in 2Q, and the state of Chinese smartphone makers'

inventories, and we think risk of substantial production adjustments has increased. We

think there will likely be major adjustments in 2017 in tandem with the two-year foundry

utilization cycle we believe exists.

With regard to mass-production using 10nm processes at TSMC, mass-production of APs

for new iPhones is due to start in March while Hi Silicon's 10nm process (Kirin 970) is due

to start being used for production in 2Q. Meanwhile, the development roadmap for

Mediatek's Helio P35/X35, which were to be mass-produced using 10nm processes has

been cancelled while the P30/X30 are due to be manufactured using 12nm processes

(improved 16nm process). As anticipated demand has been lost, we think spare capacity

will open up in 10nm processes. We think this is probably one of the reasons why the

orders for equipment for 7nm process mass-production investment from Taiwanese

foundries, which we discuss later, have not materialized.

Figure 3: Projected capacity utilization for Taiwanese foundry 300nm/200nm lines (input basis)

Source: Company data, Credit Suisse estimates

50%

60%

70%

80%

90%

100%

110%

120%

1Q

10

2Q

10

3Q

10

4Q

10

1Q

11

2Q

11

3Q

11

4Q

11

1Q

12

2Q

12

3Q

12

4Q

12

1Q

13

2Q

13

3Q

13

4Q

13

1Q

14

2Q

14

3Q

14

4Q

14

1Q

15

2Q

15

3Q

15

4Q

15

1Q

16

2Q

16

3Q

16

4Q

16

1Q

17

2Q

17

50%

60%

70%

80%

90%

100%

110%

120%

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.Jul.

Au

g.

Se

p.

Oct.

Nov.

Dec.

Jan

.F

eb.

Mar.

Ap

r.M

ay.

Jun

.

2010 2011 2012 2013 2014 2015 2016 2017

TS

MC

Quart

erly U

tiliz

ation R

ate

s

Month

ly U

tiliz

ation r

ate

(300/2

00m

m)

300mm (Input) 200mm (Input) TSMC total utilization (sales)

FORECAST

iPhone-related 10nm production likely to

begin in March

10 March 2017

Technology sector 9

Figure 4: Projected quarterly utilization rates for

major Taiwanese foundry 300mm lines (input basis)

Figure 5: Projected quarterly utilization rates for

major Taiwanese foundry 300/200nm lines, by

process (input basis)

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

Samsung Electronics LSI: Operating at 110% capacity, mainly for own smartphones

Samsung Electronics' 300mm lines continue to operate at full capacity, as was also

revealed by the last survey (105-110% input vs. 14/10nm process production capacity).

Increased adoption in its own smartphone models by Exynos (Figure 7) and increased

adoption in production for Chinese smartphones at 14nm processes, which Qualcomm

won away from TSMC, have helped 14nm capacity utilization reach a high level. In

addition, production of 10nm Exynos 8895, which is due to be used in the Galaxy S8, and

the Qualcomm MSM8998 started in small lots in 4Q last year. Yields are still low (around

40–50%), but due in part to the early start of production, we think there will not be any

bottlenecks in terms of parts supply ahead of the Galaxy S8 launch.

However, inventories of the Qualcomm MSM8996 (AP for high-end smartphones) made

using 14nm processes have built up to over 20mn units. Production cuts started in

January, but as demand is firm (as noted above) it has not led to public capacity utilization

adjustments.

Moreover, if adoption of 14nm process Qualcomm APs in mid-range and low-end Chinese

smartphones increases, we expect shortages in 300mm LSI production capacity. In

addition, 10nm process mass-production has started and we expect adoption in high-end

units this year and in mid-range smartphones from next year. Accordingly, we understand

Samsung Electronics is considering expanding capacity at Line 17 (allotting a quarter of

the total space).

With regard to foundry partners, Qualcomm has decided on Samsung Electronics for

14nm/10nm mass-production, but has not yet made a decision as regards 7nm processes.

The stock market believes Qualcomm will likely select a TSMC that is a year ahead in

terms of 7nm process mass-production. We think the 7nm process foundry partner will be

decided based on Qualcomm's 7nm process roadmap and when Samsung Electronics will

be able to commence mass-production at 7nm. Qualcomm has not disclosed its 7nm

process roadmap, but according to package substrate makers, as current package

substrate structures are expected to continue from end-2018 through 1H 2019 release

products, we think this means the 10nm process generation will continue until end-2018.

91%92%

78%76%

88%

99%95%

91%91%

91%

75%

78%

89%

99%

94%

99%

94%

93%

30%

40%

50%

60%

70%

80%

90%

100%

110%

1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q

2013 2014 2015 2016 2017

300mm Fab

30%

40%

50%

60%

70%

80%

90%

100%

110%

120%

1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q

2013 2014 2015 2016 2017

90nm 65nm40nm 28nm Poly-SiN28nm HKMG 20/16nm10nm

Brisk trends for 14/10nm processes

10 March 2017

Technology sector 10

Accordingly, we expect 7nm process APs to be released in 2019 and that Samsung

Electronics LSI maker will have put a mass-production structure in place by then. If

Samsung Electronics LSI can match Qualcomm's roadmap, we think Samsung Electronics

LSI maker, which has the largest customer for Qualcomm's high-end APs in its group, will

likely be the foundry.

Samsung Electronics LSI is also due to expand production capacity for CIS, supply–

demand for which has become tight due to demand for dual cameras in smartphones. As

there is no space for expansion on existing lines, we think the company will partly convert

Line 11, its oldest 300mm DRAM line (50k wafers/month capacity). We gather the

company already started to place equipment orders in February and that 300mm capacity

is due to be expanded by around 20% (+7k wafers).

Figure 6: Samsung Electronics LSI 14/20nm process capacity utilization

Source: Credit Suisse estimates

Figure 7: Samsung Electronics' share of smartphone AP


0%

20%

40%

60%

80%

100%

120%

Ap

r

Ma

y

Jun

Jul

Au

g

Se

p

Oct

Nov

Dec

Jan

Fe

b

Ma

r

Ap

r

Ma

y

Jun

Jul

Au

g

Se

p

Oct

Nov

Dec

Jan

Feb

Ma

r

Ap

r

Ma

y

Jun

2015 2016 2017

22%

41%50%

38%

39% 23%

34%

19%

9%

6%2%

17%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2015 2016 2017

Samsung Qualcomm Spreadtrum Marvell Mediatek

10 March 2017


Figure 8: Smartphone application processor maker technology roadmaps


NAND: Two new iPhones, possibly 256GB and 64GB

Survey points (market conditions)

■ Declining 2D NAND capacity has tightened supply across all applications again in 1Q.

■ Smartphone storage density continues expanding in 2017. China smartphone demand

is shifting to 128GB for eMMC for high-end/mid-range handsets and to 64GB for eMCP

for mid-range/low-end handsets.

■ The new iPhone is expected to offer two storage options, 256GB and 64GB (512GB

option also possible if Apple launches three models).

■ China smartphone manufacturers are to expand uptake of 3D NAND from spring

models.

■ The sufficient rate (bit supply relative to total customer demand) for China smartphones

is worsening quarterly and has fallen to less than 50% in some cases. Smartphone

manufacturer inventories remain depleted.

■ NAND pricing is forecast to rise 5–30% QoQ in 1Q and 5–20% QoQ in 2Q.

Survey points (3D NAND)

■ 3D NAND certification is proceeding for individual applications, and the product’s full-

fledged adoption in China smartphones is expected with spring models. Following

Samsung Electronics and Intel, Micron Technology’s China SSD certification is now

moving.

■ Development of 64-layer 3D NAND for the SSD market has been delayed by about one

month at all manufacturers. (The latest plans have Japanese majors sample-shipping

cSSD at end-April and Korean majors at end-May.)

■ Among Chinese SSD users, Alibaba and Lenovo are considering/have begun in-house

SSD assembly.

■ Although some PC manufacturers are behind their 1Q production schedules, SSD

procurement is unchanged, as their stagey is to build up inventory for 2H.

1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q

Samsung

Mediatek

Hi Silicon

2014 2015 2016 2017

Apple

Qualcomm

2018

MSM8994

(20nm)

Exynos 7420

(14nm)

MSM8998

(10nm)

A8

(20nm)

A9

(16nm)

A11

(10nm)

A10

(16nm)

Kirin 950

(16nm)

X20/MT6797

(20nm)

X10/MT6795

(28nm)

MSM8996

(14nm)

X30

(10nm)

Kirin 930

(28nm)

Kirin 920

(28nm)

Exynos 8895

(10nm)

Exynos 8890

(14nm)Exynos 5422

(28nm)

A10X

(10nm)

A12

(7nm ?)

Exynos (7nm)

Kirin 960

(16nm FF+)

Kirin 970

(10nm)

10 March 2017


■ NAND manufacturers use two processes for 64- and 72-layer 3D NAND: Samsung

Electronics and Toshiba have one stack of 64 layers (one memory-hole etching

process), while SK Hynix and Micron Technology have two (two processes). The two

manufacturing processes’ performances and costs are expected to diverge sharply.

■ Smartphones are the main OEM application for 64-layer 3D NAND this year. Owing to

development delays, 3D NAND’s adoption for SSD is not expected to expand until

2018 (volume production and supply to start in 4Q 2017).

■ Adoption of 3D NAND for the MacBook, the largest cSSD user, has been postponed

until 2018 or after.

■ SK Hynix’s initial 3D NAND investment is in the M14 fab. The company is expected to

spend on retooling 2D NAND capacity to position it for 48- and 72-layer products, and

is considering stepping up retooling investment if 2D NAND demand shrinks more

rapidly than projected.

Survey points (SSD)

■ cSSD is to shift to 64 layers in 2H 2017, while eSSD is to remain at 48.

■ eSSD for the Skylake platform uses 48-layer 3D NAND. Skylake for enterprises is

scheduled for release in July 2017.

■ eSSD’s per-gigabyte price is expected to be about 30% higher than cSSD’s.

■ Non-Volatile Memory express (NVMe) is the main SSD interface for China data centers

this year (Serial AT Attachment (SATA) was previously). Legacy customers (e.g.,

banks) tend to prefer SATA.

■ Samsung Electronics and Intel have monopolized China’s eSSD market. Micron

Technology was accredited recently and has also entered. Toshiba supplies discrete

NAND.

NAND supply–demand outlook: 2D NAND supply to remain tight; new iPhone expected in 256GB and 64GB options

NAND supply is expected to continue tightening in through 2017 owing to increased

storage density for China smartphones and the iPhone 7 and reduced 2D NAND supply in

bit terms as manufacturers shift to 3D NAND. NAND manufacturers’ sufficient rate is

expected to remain at 70–80% in 1Q. Supply is becoming increasingly tight: 2D NAND

supply in bit terms has decreased 5–10% quarterly, and the sufficient rate for

manufacturers supplying mainly 2D NAND has dipped below 50% for China smartphones.

While some of the smartphone components that had been in short supply since 2H 2016

are now either fully or more readily available, NAND is still the component in shortest

supply.

3D NAND demand is also growing, and its adoption in smartphones and SSDs is

increasing. Manufacturers currently use 48- and 32-layer 3D NAND, and we expect 3D

NAND supply to remain tight, as capacity for these generations has not expanded.

Adoption of 64-layer 3D NAND is expected to increase with new smartphones (led by the

iPhone) launching in 2H 2017, but yields suggest supply will likely remain tight in 2H. We

also see oversupply risk in 2018, but think this depends on manufacturing yields.

Tighter supply-demand due to reduced planar NAND supply; greater

3D NAND uptake for smartphones, SSDs

10 March 2017


Figure 9: NAND supply-demand model: Greater-than-expected tightening due to iPhone 8’s use of

64GB/256B storage


Figure 10: 2D/3D NAND bit supply forecasts

Figure 11: 2D/3D NAND production capacity

forecasts

Source: Credit Suisse estimates Source: Credit Suisse estimates

On the demand side, the new iPhone is expected to offer two NAND storage density

options: 256GB and 64GB. (The iPhone 7 offers 32GB, 128GB, and 256GB.) As the

iPhone 7 volume breakdown is 43% low density, 41% medium density and 16% high

density, the new iPhone is expected to increase 256GB’s share to 57%. We estimate

demand for NAND for the new iPhone to rise 88% YoY if production volume for the new

handsets is unchanged in 3Q and 4Q, and view this as a positive change given our

previous view that demand for NAND for the iPhone would not grow. In previous iPhone

cycles, storage density was unchanged for second-generation handsets, but we expect

this model to break down. Some observers have questioned whether Apple may also

release a 512GB handset to preserve its three-model paradigm, but we see no supporting

evidence at this stage. We had previously forecast 3D NAND to be used for 128GB in

addition to 256GB, but the two options of 256GB and 64GB suggest that it will continue to

be used for 256GB and that NAND manufacturers will have the discretion to determine

whether 2D NAND or 3D NAND is used for 64GB. Put differently, as the per-gigabyte price

-5%

14%

8%4%

-6%

11%

18%

5%

-8%

4%

13%8%

5%

13%11%

-11%

-25%

-11%-10%

-16%

-28%

-40%

-30%

-20%

-10%

0%

10%

20%

30%

1Q

07

2Q

07

3Q

07

4Q

07

1Q

08

2Q

08

3Q

08

4Q

08

1Q

09

2Q

09

3Q

09

4Q

09

1Q

10

2Q

10

3Q

10

4Q

10

1Q

11

2Q

11

3Q

11

4Q

11

1Q

12

2Q

12

3Q

12

4Q

12

1Q

13

2Q

13

3Q

13

4Q

13

1Q

14

2Q

14

3Q

14

4Q

14

1Q

15

2Q

15

3Q

15

4Q

15

1Q

16

2Q

16

3Q

16

4Q

16

1Q

17

2Q

17

3Q

17

4Q

17

Oversupply

Supply shortage

22,186 22,436 23,098 21,681 19,965 18,500 17,363 16,181

1,966 3,221 4,636 6,954 9,504 14,503

19,530 24,040

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

1Q16E 2Q16E 3Q16E 4Q16E 1Q17E 2Q17E 3Q17E 4Q17E

Pro

ductio

n (

mn G

B)

2D NAND 3D NAND

0%

10%

20%

30%

40%

50%

60%

70%

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

1Q16E 2Q16E 3Q16E 4Q16E 1Q17E 2Q17E 3Q17E 4Q17E

3D

NA

ND

% o

f to

tal capacity

3D

Mass p

roductio

nC

apacity (

K w

pm

)

2D Capacity 3D Capacity % of total NAND capacity

New iPhone likely to offer two storage options: 64GB or

256GB

10 March 2017


is based on 3D NAND cost, we see 3D NAND adoption increasing. However, if we

assume Toshiba and Samsung Electronics are the main NAND suppliers for 256GB and

SK Hynix is the main supplier for 64GB, SK Hynix’s 3D NAND volume-production system

suggests 2D NAND could be preferred.

Figure 12: Weighting by iPhone generation, storage option (based on four

quarters since launch)

Source: IDC, Credit Suisse

Figure 13: Shipment volume by iPhone generation, storage option (based on

four quarters since launch); equivalent volumes given for two quarters for

iPhone 8 to enable comparison with iPhone 7.


49%

54%

68% 66%

51% 50%57%

51%

27%22% 22%

43%

1%

6%11% 12%

36% 38%

43%

42%

13% 12%

41%

16%

57%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

iPhone4

iPhone4s

iPhone5

iPhone5s

iPhone6

iPhone6s

iPhone7

iPhone8

iPhoneSE

256GB

128GB

64GB

32GB

16GB

1GB

0

20

40

60

80

100

120

140

160

180

iPhone4

iPhone4s

iPhone5

iPhone5s

iPhone6

iPhone6s

iPhone7

iPhone8

iPhoneSE

iPh

on

e s

hip

me

nt u

nits (

mn

)

256GB

128GB

64GB

32GB

16GB

8GB

1GB

2 quarters only

10 March 2017


Figure 14: Volume of NAND demand by iPhone generation, storage option

(based on four quarters since launch); NAND demand calculation based on

same two-quarter assumption for iPhone 8 to enable comparison with iPhone 7.


China smartphone manufacturers are also increasing their adoption of 128GB for eMMC

for high-end/mid-range handsets, and we expect adoption of 64GB for mid-range/low-end

handsets to expand this year. With the 3D NAND supply structure becoming the

mainstream, 3D NAND’s minimum density is 32GB per chip (256Gb), and as two stacks

are needed for performance, minimum density is 64GB. Although 2D NAND still supports

smartphones with storage density of 32GB or less, we expect unit pricing to rise, as

tightening supply makes it difficult to secure volume.

We see production adjustments and bill of materials (BOM) cost growth as potential risks

for China smartphones. On the production side, the ongoing acute NAND supply shortage

makes it difficult to foresee the supply–demand balance worsening if we include inventory

demand as well. On the cost side, continuing competition to improve specs and

manufacturers’ growing adoption of 3D NAND suggest little risk of installed NAND density

growing less rapidly or decreasing because of BOM cost growth. Our view is that installed

density will continue expanding.

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

iPhone4

iPhone4s

iPhone5

iPhone5s

iPhone6

iPhone6s

iPhone7

iPhone8

iPhoneSE

NA

ND

co

nsu

mp

tio

n m

n G

B

256GB

128GB

64GB

32GB

16GB

1GB

2 quarters only

10 March 2017


Figure 15: Chinese smartphone NAND density ratio trend estimates

Source: IDC, Credit Suisse estimates

Figure 16: NAND density: iPhone, Samsung, Huawei, Chinese smartphone

average

Source: IDC, Credit Suisse estimates

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0

100.0

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q

2013 2014 2015 2016 2017

<1GB 1GB 4GB

8GB 16GB 32GB

64GB 128GB+ Ave density (GB)

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

90.0

100.0

1Q

13

2Q

13

3Q

13

4Q

13

1Q

14

2Q

14

3Q

14

4Q

14

1Q

15

2Q

15

3Q

15

4Q

15

1Q

16

2Q

16

3Q

16

4Q

16

Avera

ge N

AN

D D

ensity

GB

/unit

Huawei Samsung iPhone Oppo China Average

10 March 2017


Figure 17: NAND total demand bit growth, mobile Phone bit growth


NAND prices: likely to keep rising in 2Q

NAND prices for smartphone and SSD applications are apparently up 5–30% in 1Q.

Increases of 5–20% QoQ are now under negotiation, meaning this uptrend is likely to

persist in 2Q.

We understand that while US makers discounted prices for low-end 3D NAND, prices for

SSD applications swung upward after product approvals, leading to a positive overall

trend.

Figure 18: 128Gb TLC NAND contract price trend

Source: DRAMeXchange, Credit Suisse

63%

132%

106%

90%

60%

21%

32%

55%

86%

77%

54%

34% 32%38%

97%

109%

76% 74%

33%

52%

39%45% 43% 40%

35%

26%30%

41%47% 44%

37%

25%32%

36%

63%67%

55% 57% 39%

52%

0%

20%

40%

60%

80%

100%

120%

140%

1Q

13

2Q

13

3Q

13

4Q

13

1Q

14

2Q

14

3Q

14

4Q

14

1Q

15

2Q

15

3Q

15

4Q

15

1Q

16E

2Q

16E

3Q

16E

4Q

16E

1Q

17E

2Q

17E

3Q

17E

4Q

17E

Mobile Phone Bit Growth (%) NAND Total Demand Bit Growth (%)

iPhone6 128GB

iPhone7 256GBChina 32/64/128GB

iPhone864/256GB

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

Nov-1

2

Jan

-13

Ma

r-13

Ma

y-1

3

Jul-1

3

Se

p-1

3

Nov-1

3

Jan

-14

Ma

r-14

Ma

y-1

4

Jul-1

4

Se

p-1

4

Nov-1

4

Jan

-15

Ma

r-15

Ma

y-1

5

Jul-1

5

Se

p-1

5

Nov-1

5

Jan

-16

Ma

r-16

Ma

y-1

6

Jul-1

6

Se

p-1

6

Nov-1

6

Jan

-17

US

D

NAND contract price TLC 128Gb 16384Mx8

5–20% price increases likely in 2Q

10 March 2017


3D NAND demand/development/production: greater US uptake in SSD applications

In terms of 3D NAND demand, approvals were limited to a few SSD applications in 2016,

but look likely to extend to a full-range of applications, including smartphones, cSSD and

eSSD in 2017 when 3D NAND uptake for mass-production lines should pick up. Nearly all

key Chinese smartphone makers look set to step up usage of 3D NAND for spring models

onward. South Korean makers have not switched yet, partly as few of their models have

high storage density. However, we understand they are mulling 3D NAND for high-storage

(128GB) flagship models due to come out in 2H 2017.

From the standpoint of cSSD demand among notebook makers, we note key PC makers

look likely to cut production targets but appear to be building up SSD inventory in

anticipation of 2H demand, so we expect limited negative impact on NAND supply–

demand.

On the demand-side, we understand the anticipated adoption of 3D NAND for the

MacBook this year has been postponed to “hedge” supply risk, as use of such chips in the

new iPhone put heavy upward pressure on demand. Those fearing oversupply following

maker investment in 3D NAND capacity should remember that the upcoming MacBook,

iPad and other products are likely to become major sources of demand.

On the supply-side, makers are responding with mass-production for 32- and 48-layer 3D

NAND, and are expected to mass-produce 64-layer chips in 2H 2017. However, in the

area of 64-layer development, we note cSSD/eSSD development is behind schedule at

the largest South Korean maker and its Japanese peer. Both companies have submitted

samples for use in the new iPhone model, which are currently under evaluation. Current

plans call for the Japanese maker and the largest South Korean maker to ship cSSD

samples at end-April and end-May respectively, meaning their products could be used as

early as in the autumn notebook models. We understand that eSSDs will rely mainly on

48-layer technology this year and shift to 64-layer chips from next year.

Lower-tier makers developing 64- and 72-layer chips are likely to use double-stack

designs (two stacks of 32-layer or 36-layer chips; the memory-hole etching process is

conducted twice) with inferior technical and cost performance to single-stack designs. We

thus see a wide gap opening up between the top and bottom makers of 64-layer products.

Development of 96-layer chips is unlikely to advance to mass production until makers

secure the appropriate single-etch memory hole technology. Some have development

roadmaps for 96-layer technology but we doubt they have set a date for mass production.

China server, data center demand: in-house SSD assembly at Alibaba, Lenovo

Although the Chinese server/datacenter market amounts to only one third of the demand

generated by US hyperscale datacenters, it continues to grow.

The eSSDs being shipped now are compatible with Broadwell platforms but are likely to

shift to Skylake architecture this year. A new Skylake CPU is slated for release in July that

should be compatible with 48-layer 3D NAND.

The typical eSSD interface has been SATA but looks likely to become NVMe in 2017.

The two main distribution channels for eSSD in China involve: (1) NAND makers supplying

SSDs directly to customers; and (2) NAND makers supplying NAND chips to local SSD

makers or to customers’ in-house SSD assembly plants. Samsung Electronics and Intel

dominated SSD supply, but Micron has started to secure approvals recently (for 32-layer

3D NAND products). Meanwhile, Alibaba and Lenovo in particular have plans for in-house

SSD assembly. Due to differing business models, Samsung Electronics is apparently not

providing support to local SSD makers (Memblaze, Shannon Systems) and in-house SSD

assembly plants, but Toshiba and Micron look likely to.

Greater 3D NAND uptake in cSSD/eSSD

64-layer NAND SSD development delayed at

Toshiba, Samsung Electric

Costs cloud prospects for 96-layer mass-

production

10 March 2017


We understand that Lenovo has begun in-house SSD assembly, while Alibaba is

considering it, and that Lenovo is making controllers in-house.

Figure 19: Enterprise SSD supply chain in China


Figure 20: Enterprise storage market (worldwide) Figure 21: Enterprise storage market (US)

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

Discrete NAND SSD in-house

Hyper Scale

Tencent Intel

Alibaba Shannon Toshiba, Micron

Baidu Intel, (Lenovo ?)

Huawei Toshiba, Micron

SSD maker

Memblaze Toshiba

Shannon Toshiba

Set maker

Lenovo Toshiba, Micron

1.21 1.34 1.58

1.97

2.01

2.25

2.30 2.91

0.0

10.0

20.0

30.0

40.0

50.0

60.0

20

13Q

1

20

13Q

2

20

13Q

3

20

13Q

4

20

14Q

1

20

14Q

2

20

14Q

3

20

14Q

4

20

15Q

1

20

15Q

2

20

15Q

3

20

15Q

4

20

16Q

1

20

16Q

2

20

16Q

3

20

16Q

4

Exabyte

s

HDD Flash

0.40 0.43 0.50 0.67

0.65

0.83

0.79 1.05

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.020

13Q

1

20

13Q

2

20

13Q

3

20

13Q

4

20

14Q

1

20

14Q

2

20

14Q

3

20

14Q

4

20

15Q

1

20

15Q

2

20

15Q

3

20

15Q

4

20

16Q

1

20

16Q

2

20

16Q

3

20

16Q

4

Exa

byt

es

HDD Flash

Trend toward in-house SSD assembly

10 March 2017


Figure 22: Enterprise SSD shipments by capacity Figure 23: Enterprise SSD shipment mix by capacity

Source: TSR, Credit Suisse Source: TSR, Credit Suisse

Figure 24: Enterprise SSD shipments by maker Figure 25: Enterprise SSD share by maker


Figure 26: PC SSD shipments by capacity Figure 27: PC SSD shipment mix by capacity


0

200

400

600

800

1,000

1,200

1,400

Jan

Fe

bM

ar

Apr

Ma

yJun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Aug

Se

pO

ct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

eJuly

Au

gS

ep

Oct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

Enterprise shipment by capacity

<100GB 101-199GB 200-399GB 400-599GB

600GB< 600-999GB 1TB-1.9TB >2TB

0%

20%

40%

60%

80%

100%

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Apr

Ma

yJun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

Nov

De

cJan

Feb

Ma

rA

pr

Ma

yJun

eJuly

Au

gS

ep

Oct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

Enterprise shipment by capacity

<100GB 101-199GB 200-399GB 400-599GB

600GB< 600-999GB 1TB-1.9TB >2TB

0

200

400

600

800

1,000

1,200

1,400

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

June

July

Aug

Sep

Oct

Nov

Dec

Jan

2013 2014 2015 2016 2017

Units (

k p

cs)

Enterprise shipment by supplier

Samsung IntelSandisk (Fusion-io) WDC (STEC)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Jan

Fe

bM

ar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Fe

bM

ar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Jan

Fe

bM

ar

Apr

May

June

July

Aug

Sep

Oct

Nov

Dec

Jan

2013 2014 2015 2016 2017

Units (

k p

cs)

Enterprise shipment by supplier

Samsung Intel Sandisk (Fusion-io) WDC (HGST/Sandisk) WDC (STEC) Others

0

1,000

2,000

3,000

4,000

5,000

6,000

Jan

Fe

bM

ar

Apr

Ma

yJun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Feb

Ma

rA

pr

Ma

yJun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

eJuly

Au

gS

ep

Oct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

PC by density

<32GB 33-64GB 65-128GB 129-256GB 256GB 512GB >1TB

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

No

vD

ec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

No

vD

ec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

No

vD

ec

Jan

Feb

Mar

Apr

May

June

July

Aug

Sep

Oct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

PC by density

<32GB 33-64GB 65-128GB 129-256GB 256GB 512GB >1TB

10 March 2017


Figure 28: PC SSD shipments by maker Figure 29: PC SSD share by maker


Chinese makers’ entry into NAND market

Yangtze River Storage Technology (YRST) still seems to have not yet obtained foreign

support in the form of capital tie-ups and technology licensing for its entry into the NAND

market. Even so, the company began plant construction at end-2016 and we understand it

is now in initial talks with equipment makers (for 100k-wafer capacity).

We thought setting up NAND production would be difficult without technology licensing,

but equipment lists, plant layouts and other info have seemingly leaked out of South

Korea. Access to such info would likely put YRST on track to produce NAND. However,

the victim of the leak has apparently asked equipment makers not to sell the listed items;

their response will be key, in our view.

We also gather that YRST is developing 32-layer technology, and plans trial production

runs in 2H 2018, followed by mass production in 2019 (Figure 30).

Figure 30: NAND makers' technology roadmap


0

1,000

2,000

3,000

4,000

5,000

6,000

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Feb

Ma

rA

pr

Ma

yJun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Apr

Ma

yJun

Jul

Au

gS

ep

Oct

Nov

De

cJan

Fe

bM

ar

Ap

rM

ay

Jun

eJuly

Au

gS

ep

Oct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

PC shipment by supplier

SanDisk WDC (HGST/Sandisk)Intel Toshiba

0%

20%

40%

60%

80%

100%

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Apr

May

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

Jun

Jul

Au

gS

ep

Oct

No

vD

ec

Jan

Fe

bM

ar

Ap

rM

ay

June

July

Aug

Se

pO

ct

No

vD

ec

Jan

2013 2014 2015 2016 2017

Un

its (

k p

cs)

PC shipment by supplier

SanDisk WDC (HGST/Sandisk)Intel Toshiba

1Q 14 2Q 14 3Q 14 4Q 14 1Q 15 2Q 15 3Q 15 4Q 15 1Q 16 2Q 16 3Q 16 4Q 16 1Q 17 2Q 17 3Q 17 4Q 17 1Q 18 2Q 18 3Q 18 4Q 18 1Q 19 2Q 19 3Q 19 4Q 19

3D NAND 32 Layers

48 Layers

64 Layers

72 Layers

96 Layers

Leak of equipment list and technology

10 March 2017


DRAM: Supply–demand eased depending on

applications and customers

Key takeaways

■ Excluding inventory demand, PC DRAM and mobile DRAM were in oversupply in 1Q

2017. Server DRAM supply–demand was tight. There were therefore differences in

supply–demand in terms of customers and applications.

■ DRAM makers flexibly change their product mixes. We expect Samsung Electronics

plans to increase PC/server DRAM supply from March output.

■ DRAM inventories are in the market, but have dried up at DRAM makers. However,

some PC makers have secured sufficient inventories.

■ With regard to PC DRAM, PC demand is lower than expected, but PC makers are

building up inventories ahead of expected tight supply–demand in 2H 2017.

■ In mobile RAM, as inventories at some Chinese smartphone makers are excessive,

they are passing some of their allocation to midsize and small manufacturers. Supply

of mobile RAM to Chinese smartphone makers is balanced between actual demand

and demand to build up inventories.

■ In server DRAM, due to the adoption of the Skylake platform, DRAM module capacity

has expanded sharply from 16GB to 32GB. Density per server has also increased.

■ There is scope for an increase in DRAM prices of nearly 10% (the price of 4GB

modules rising from $25–26 to $27–28).

■ Investment in leading-edge processes by DRAM makers looks weak and is limited.

DRAM supply–demand: Different depending on applications, customers

DRAM supply–demand has turned around from being tight at one point, and due to lower-

than-expected PC production and inventory adjustment at some Chinese smartphone

makers, PC DRAM and mobile RAM is in oversupply on an actual demand basis. We are

seeing adjustments at some Chinese smartphone makers due to excess DRAM and

product inventories. The only product area in which supply–demand is still tight is server

DRAM, as supply is limited whereas demand has risen sharply. We think that to date, the

largest maker in particular, based on a focus on earnings in 1Q 2017 concentrated on

supply allocation for servers, which carry the lowest margins, aiming to maximize earnings

derived from setting aside some of its allocation for PC DRAM and mobile RAM. However,

we gather there was a production shift from mobile RAM to PC/server DRAM in January,

and that there are concerns about the impact on PC DRAM supply–demand of the output

contribution in the second half of March. We expect supply demand continue to be

mismatched as PC DRAM and server DRAM data width are different.

On the demand front, with regard to DRAM for servers, as Intel's Skylake platform was

released last year and DRAM modules were upgraded from 16GB to 32GB, DRAM per

server in servers for data centers has increased sharply.

With 3GB and 4GB now mainstream in Chinese smartphones, we think bit growth

henceforth will likely be gradual. In smartphones, the spread of LP DDR4 is a theme this

year, but we think it will be adopted in mid-range smartphones (priced at RMB2,000 and

higher) only in 2H 2017 at the earliest.

On the supply front, excluding Samsung Electronics, companies were struggling to

improve yields at 20nm processes, but we understand they are seeing improvement. We

did not find a single DRAM maker that was bullish on capex, as cutting costs due to

process migration has become difficult due to protracted yield improvement.

Glut due to slower mobile-RAM demand;

closer look may reveal excess PC DRAM

inventory

10 March 2017


Figure 31: DRAM supply-demand model


Figure 32: Mobile RAM supply-demand model Figure 33: PC DRAM supply-demand model

Source: Credit Suisse estimates Source: Credit Suisse estimates

6%

1%

-4%

12%

7%7%

-4%

12%

5%

3%

-5%

14%

1%

0%

-13%-15%

-10%

-5%

0%

5%

10%

15%

20%

1Q

04

2Q

04

3Q

04

4Q

04

1Q

05

2Q

05

3Q

05

4Q

05

1Q

06

2Q

06

3Q

06

4Q

06

1Q

07

2Q

07

3Q

07

4Q

07

1Q

08

2Q

08

3Q

08

4Q

08

1Q

09

2Q

09

3Q

09

4Q

09

1Q

10

2Q

10

3Q

10

4Q

10

1Q

11

2Q

11

3Q

11

4Q

11

1Q

12

2Q

12

3Q

12

4Q

12

1Q

13

2Q

13

3Q

13

4Q

13

1Q

14

2Q

14

3Q

14

4Q

14

1Q

15

2Q

15

3Q

15

4Q

15

1Q

16

2Q

16

3Q

16

4Q

16

1Q

17

2Q

17

3Q

17

4Q

17

Oversupply

Supply shortage

13%

24%

14%

-9%

5%6%

14%

0%

4%2%

3%

-1%

11%

3%

-3%

-8%

0%

-6%

1%

4%

21%

-4%

-1%

-8%

-10%

-5%

0%

5%

10%

15%

20%

25%

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

1Q

12

3Q

12

1Q

13

3Q

13

1Q

14

3Q

14

1Q

15

3Q

15

1Q

16

3Q

16

1Q

17

3Q

17

1G

b e

quiv

. m

n u

nits

Demand Supply Demand/supply sufficiency rate

13%

24%

5%0%

15%

6%

-9%

-26%

-2%

8%

0%

-6%

13%10%

13%

-2%

22%

16%

4%

-13%

8%5%

0%

-19%

-30%

-20%

-10%

0%

10%

20%

30%

40%

0

2,000

4,000

6,000

8,000

10,000

12,000

14,0001Q

12

3Q

12

1Q

13

3Q

13

1Q

14

3Q

14

1Q

15

3Q

15

1Q

16

3Q

16

1Q

17

3Q

17

1G

b e

quiv

. m

n u

nits

Demand Supply Demand/supply sufficiency rate

10 March 2017


Figure 34: Chinese smartphone RAM density trend


Figure 35: Server DRAM average density

Source: DRAMeXchange

DRAM inventories: Some customers have excess inventories

DRAM maker and distributor inventories are very low. However, due to sluggish sales,

mobile RAM inventories at some Chinese smartphone makers are excessive. We

understand that in mobile RAM, these makers have been passing some of their allocations

on to midsize and small smartphone makers, which have low inventories. In PC DRAM,

PC makers purchase inventory when demand is sluggish. Already some manufacturers

have been saying they have no room for more inventories also in PC DRAM.

DRAM inventory may be quietly accumulating at downstream producers, meaning their

inventory levels may become more apparent when demand slumps.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

6GB 4GB 3GB 2GB 1GB 512MB Ave DRAM (GB)

117.9

133.0 141.9 144.8 151.2

173.7

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

180.0

200.0

1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q

2014 2015 2016 2017

Serv

er

DR

AM

density/

box (

GB

)

10 March 2017


DRAM prices: We see room for prices to rise by around 10%, but this is broadly the peak

DRAM makers plan to raise PC DRAM prices to $27–28 (from around $25–26 now).

However, this appears to be the upper limit given PC BOM costs and DRAM market

conditions.

We think it will be difficult to raise DRAM price, as we have heard examples of demand

having slowed down and some PC makers have raised prices due to higher BOM costs.

Figure 36: DDR4 4GB module large lot price trend


LP DDR4 adoption

LP DDR4 is being adopted in high-end smartphones. We think adoption in mid-range

Chinese smartphones (priced at RMB2,000 or around $290 and higher) will likely increase

in tandem with the launch of APs that support LP DDR4. However, we think this will only

happen in 2H 2017. However, there are as yet no signs of moves to use LP DDR4 in low-

end smartphones. Compared with when LP DDR2 was replaced by LP DDR3, support is

modest in migration to LP DDR4 and the pace of change feels slow.

Several major Chinese smartphone makers have expressed interest in high-speed (3GHz

to 4GHz) LP DDR4 (LP DDR4X) and we think adoption will start. However, we think North

American and Korean smartphone makers do not yet have any adoption plans. The

technology looks set for use in spring models onward, along with the new iPhone.

10.00

12.00

14.00

16.00

18.00

20.00

22.00

24.00

26.00

Au

g

Se

p

Oct

No

v

De

c

Jan

Fe

b

Ma

r

Ap

r

Ma

y

Jun

July

Au

g

Se

p

Oct

No

v

De

c

Jan

Fe

b

2015 2016 2017

US$

We think LP DDR4 will be adopted in mid-

range smartphone only in 2H 2017 at the

earliest; expecting high-speed LP DDR4 to be adopted in high-end Chinese smartphones

10 March 2017


Figure 37: Chinese smartphone breakdown by ASP

Source: IDC

DRAM 1xnm process migration

In terms of 1xnm processes at the largest DRAM maker, PC ODM was approved in

November last year and mass-production has started, but there are no signs of a mass

shift to 1xnm. Compared with processes to date, the yield improvement curve has been

gradual and productivity has declined, so we think we cannot expect ROI as high as in the

past.

0%

20%

40%

60%

80%

100%

200

7Q

1

200

7Q

3

200

8Q

1

200

8Q

3

200

9Q

1

200

9Q

3

201

0Q

1

201

0Q

3

201

1Q

1

201

1Q

3

201

2Q

1

201

2Q

3

201

3Q

1

201

3Q

3

201

4Q

1

201

4Q

3

201

5Q

1

201

5Q

3

201

6Q

1

201

6Q

3

$0<$50

$50-$100

$100-$200

$200-$300

$300-$400

$400-$500

$500-$600

$600<

No sign of bullish stance on 1xnm

process migration

10 March 2017


SPE capex trends

Front-end SPE capex: Taiwanese foundries ease up;

3D NAND firm; initial inquiries picking up in China

Key points

■ In leading-edge technology investment among TSMC, 10nm process expansion is set

to continue through 2Q 2017, but then come to an end. A mini-line (several thousand

wafers/month) is scheduled to start up in 2Q 2017 at a 7nm process mass-production

plant. We understand no concrete discussions are underway so far for the period

beyond this.

■ TSMC operated 300mm operations in Nanjing (China) are moving from plants in

Hsinchu and Tainan (both Taiwan) around July.

■ Samsung Electronics is expanding at its Austin (US) plant (14/10nm) and newly adding

10nm production on Line 17. CIS production capacity is also slated for expansion.

■ There are no major changes in investment by Samsung Electronics in new 3D NAND

plant, and we expect such investment flows to remain steady. An expansion of 80,000–

90,000 wafers (we had forecast 70,000) is planned by end-2017. The bulk of mass-

production plans are for 64-layer products.

■ SK Hynix also plans to build 36-layer 3D NAND capacity on its M14 line. Its plans also

include converting existing 2D NAND lines to 3D NAND if 2D demand drops.

■ Additional investment in Nanya Fab3N is going ahead for the purpose of expanding

monthly capacity from the current 10,000 wafers to 38,000 by end-3Q 2017.

■ We understand there are additional delays in 10nm development at Intel.

■ Initial SPE inquiries have started coming in from Chinese 3D NAND makers. Early

indications point to demand for production capacity of 100,000 wafers/month.

TSMC capex trends: Plan for 7nm mass-production line still undisclosed

TSMC built 35,000 wafers of monthly 10nm production capacity for the iPhone/iPad,

HiSilicon, and Mediatek up to end-2016. Another 20,000 wafers of capacity is planned for

1H 2017 and construction still appears to be progressing on the schedule we confirmed

three months ago. This totals about 55,000 wafers of 10nm capacity added. In the

intervening three months, however, Mediatek's Helio P35 has cancelled a 10nm project

and it now looks as though production for the iPhone could be peaking out, thus leading to

an unexpected surplus.

Equipment for next-generation 7nm process development and test lines amounting to

several thousand wafers/month has already been delivered. A 7nm process production

line is planned for Taichung, and investment in a mini-line (several thousand

wafers/month) for plant certification at Fab 15 P5/6 is planned for 2Q 2017. That much is

unchanged from three months ago, but no concrete expansion plans for the resulting

production line have thus far appeared. This is probably no more than a matter of timing,

but our sense is that such specifics are by now somewhat overdue. Mediatek’s above-

mentioned project cancellation necessitated an earlier reuse of equipment from 10nm

process for 7nm process, and that is consistent with a delay to the start of new

negotiations.

That said, we believe that preparation for 7nm process products for the new 2018 iPhones

will require ordering production-line equipment by no later than 3Q 2017.

Taiwanese foundries plan for 7nm mass-production line still

undisclosed

10 March 2017


Samsung Electronics capex trends: Expanding 10nm process; additional CIS line expansion

As mentioned above, production using leading-edge technology at the top Korean logic-

chip maker remains at full capacity. One reason is rising use of its proprietary Exynos AP

in its own smartphones. Another is that it won orders from Qualcomm for 14/10nm process

products. Investment to expand 14/10nm capacity at its Austin plant has already begun,

and it also plans to launch a new line in Korea.

We understand that Line 17, which is divided into four sections, has already had DRAM

assigned to two of these sections and 3D NAND assigned to a third. The remaining

section could therefore still be assigned to logic-chip production.

Likewise in CIS, the company expects demand from adoption of dual-lens systems for

smartphones to increase still further. It has also decided to convert its oldest DRAM line,

Line 11, to the task. We understand orders from device makers have already started

coming in.

3D NAND capex trends: Investment continuing

Samsung Electronics, in addition to investing in the conversion of its existing Line 16 to 3D

NAND and taking delivery of equipment for 3D NAND production on Line 17, also started

taking equipment deliveries at its Pyeongtaek plant in December 2016. It has since been

increasing capacity step by step. We understand it plans to reach capacity of 80,000–

90,000 wafers/month by end-2017 (we had forecast 70,000). The manufacturer plans to

start up with 48-layer products, with mass production then mainly consisting of 64-layer

products. However, 64-layer SSD sample shipments are behind schedule, and we surmise

that the actual shift to 64-layer production will depend on certification timing. SK Hynix also

plans new 36-layer 3D NAND capacity for Phase 2 of M14, though this product is not cost

competitive.

Figure 38: Samsung Electronics: Pyeongtaek plant

(end-February)

Figure 39: Samsung Electronics: Pyeongtaek plant

(end-February)

Source: Credit Suisse Source: Credit Suisse

Korean logic-chip plans include expanded

10nm; increased CIS capacity

10 March 2017


Figure 40: Pyeongtaek plant near end of

construction (end-November)

Figure 41: Pyeongtaek plant near end of

construction (end-November)


Figure 42: Pyeongtaek plant during construction

(end-August)

Figure 43: Pyeongtaek plant during construction

(end-August)


Capex trends in China: Initial inquiries for Chinese NAND

There are three main current Taiwanese equipment investment plans in China. At UMC’s

Xiamen plant, about 10,000 wafers of month 40nm process capacity already reached

mass production by end-2016, and 28nm process test production has also begun, with

expansion to 10,000 wafers/month planned. Powerchip’s Hefei plant, which makes LCD

drivers, was scheduled to take equipment deliveries in 2Q 2017, but we understand this is

being pushed back into 2H. Finally, TSMC’s Nanjing plant plans to take equipment

deliveries in July 2017, though most of this is to be transferred 20/16nm process

equipment from existing lines in Taiwan.

Phase 2 investment at Intel’s Dalian plant is also starting, with plans to raise monthly

capacity to 15,000–20,000 wafers in 2H 2017.

XMC (in Wuhan), a subsidiary of Yangtze River Storage Technology, started building a

plant at end-2016. Test production is slated for 2018 followed by full-scale production in

2019. The company has already started making initial inquiries with equipment makers,

and we understand that this move indicates interest in equipment corresponding to a

capacity of some 100,000 wafers, and that production capacity could reach 50,000 wafers

by end-2018. We understand further that equipment lists, process information, and other

Chinese 3D NAND makers now apparently receiving initial general

inquiries

10 March 2017


data was leaked out from Korean makers, but so long as the products are to be used

solely within China, a lawsuit for IP infringement would be difficult to pursue, in our view.

According to the leaked information, the plans call for 32-layer 3D NAND to be followed by

48-layer and then 64-layer. As such, this would be a fairly difficult course for XMC to

pursue all on its own. According to the roadmap, the company plans 32-layer technology

for 2H 2018 and 64-layer for 2H 2019 (Figure 44).

Front-end SPE maker orders

Several companies reported record-high orders in 4Q 2016. Naturally, a reaction is

expected in 1Q 2017, but we expect orders to remain high due mainly to a build-up of

NAND, DRAM, CMOS image sensors (CIS), and LSIs.

From 2Q, we expect investment in the new NAND plant by Samsung Electronics to peak

and orders to decline to the level seen in 1H last year, due mainly to declines in several

small-lot DRAM, CIS, and other projects.

We think the focal point from now on will be the timing of moves in 7nm processes by

TSMC and Chinese 3D NAND. We expect orders associated with 7nm processes at

TSMC to start to come on stream in 3Q, and orders in Chinese 3D NAND to commence in

3–4Q. As Chinese 3D NAND investment is in 32-layer devices, the value of the investment

could be small compared with investment in the new plant by Samsung Electronics.

However, after these two projects end, the only major new plant investment on the horizon

is Toshiba's Yokkaichi Y6 fab, and we think that on an order basis, 4Q 2016–1Q 2017 and

2H 2017 will be peaks in terms of major cycles (the next will be from 2019).

Orders remained firm in Jan–Mar, especially for

3D NAND; expecting decline from Apr–Jun if foundries do not invest

in 7nm processes

10 March 2017


Figure 44: Semiconductor maker equipment ordering plans


Delay risk in Intel's 10nm logic-chip project

Our research at Semicon Japan confirmed delays in Intel’s plans to invest in 10nm

process mass production lines (adding 14nm process investments instead). This time, our

research found no further progress on selecting package substrate materials, a step that

had been planned for completion in 1Q 2017. Judging from past patterns, this suggests

that mass production will not be achieved within 2017. However, new 14nm projects are in

the works, and in terms of the tick-tock strategy used in the past, this represents a tick-

tock-tock strategy for 14nm process, which now appears on the verge of becoming tick-

tock-tock-tock. Overall, the peak for 10nm process equipment deliveries, which we had

expected for 1H 2017, looks set to arrive only later.

2016 2017 2018

1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q 1Q 2Q 3Q 4Q

Foundry/Logic 10nm FinFET Fab12 P7

10nm FinFET Fab15 P5

7nm FinFET Fab15 P6

16nm FinFET+ Nanjing

Samsung 10/14nm FinFET S2 (Austin)

10nm FinFET Line 17

Intel 10 nm Israel

UMC 40/28nm XiaMen, China

Powerchip 90nm Hefei, China

GlobalFoundries Dhengdu, China

DRAM Samsung Line 17

SK Hynix M14

Micron 1xnm ex-Elpida Hiroshima * 1xnm migration pushed out to 1Q17

ex-Rexchip * 1xnm process migration in 1Q17

Nanya Fab3AN

Winbond ?

NAND Samsung 3D NAND 48 Layers Xi-an

Line 16 ** Conversion from DRAM to Planer/3D.

3D NAND 48 Layers Line 17

3D NAND 48/64 Layers Line 18 (Pyongteak)

Toshiba 3D NAND 48 Layers Fab2

3D NAND 64 Layers

Micron 3D NAND 32 Layers Singapore

3D NAND 64 Layers Singapore ? (maybe floating gate structure)

SK Hynix 3D NAND 48 Layers

3D NAND 72 Layers

14nm M12

Intel 3D NAND Fab68, Dalian

Yangtze River 3D NAND 32 Layers Wuhan, China

Timing of order placement

Equipment installation

R&D / Pilot Production

Mass production

* 1xnm process migration in 2H 2016

* 14nm process migration

10 March 2017


Figure 45: Tokyo Electron: Quarterly order

estimates Figure 46: Screen: Quarterly order estimates


Figure 47: Hitachi Kokusai Electric: Quarterly order

estimates


106.4

76.066.9

91.3

108.297.7

151.4143.3

154.2

136.6140.5

164.3179.3

153.5

122.8

158.0

188.7

197.5195.7

267.1

200.0

170.0

200.0220.0

230.0

200.0

170.0

200.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Ja

n-1

2A

pr-

12

Ju

l-12

Oct-

12

Ja

n-1

3A

pr-

13

Ju

l-13

Oct-

13

Ja

n-1

4A

pr-

14

Ju

l-14

Oct-

14

Ja

n-1

5A

pr-

15

Ju

l-15

Oct-

15

Ja

n-1

6A

pr-

16

Ju

l-16

Oct-

16

Ja

n-1

7A

pr-

17

Ju

l-17

Oct-

17

Ja

n-1

8A

pr-

18

Ju

l-18

Oct-

18

0

50

100

150

200

250

300

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

E2Q

E3Q

E4Q

E1Q

E2Q

E3Q

E4Q

ECY2012 CY2013 CY2014 CY2015 CY2016 CY2017 CY2018

TO

PIX

rela

tive

P/B

(x)

SP

E O

rder

(¥ b

n)

Order TOPIX rel P/B

29.3

38.3

25.5 29.0

35.0 35.1 40.7

45.1

35.1 36.6 36.9

41.1 45.9

38.1 36.8

38.6

48.1 49.1 53.4

64.8

59.0

44.0 46.0

55.0 58.0

52.0

37.0 37.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Ja

n-1

2A

pr-

12

Ju

l-12

Oct-

12

Ja

n-1

3A

pr-

13

Ju

l-13

Oct-

13

Ja

n-1

4A

pr-

14

Ju

l-14

Oct-

14

Ja

n-1

5A

pr-

15

Ju

l-15

Oct-

15

Ja

n-1

6A

pr-

16

Ju

l-16

Oct-

16

Ja

n-1

7A

pr-

17

Ju

l-17

Oct-

17

Ja

n-1

8A

pr-

18

Ju

l-18

Oct-

18

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

1Q

2Q

3Q

4Q

1Q 2Q

3Q

4Q

1Q

2Q

3Q 4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

E2Q

E3Q

E4Q

E1Q

E2Q

E3Q

E4Q

E

CY2012 CY2013 CY2014 CY2015 CY2016 CY2017 CY2018

TO

PIX

rela

tive P

/B (

x)

SP

E O

rder

(¥bn)

Order TOPIX rel P/B

13.215.0

9.112.0

15.1

20.3

14.9

35.2

25.6

13.916.6

26.2

32.1

29.9

13.0

14.5

24.0

23.223.4

32.432.0

20.018.0

25.027.0

30.0

20.020.0

0.0

0.5

1.0

1.5

2.0

2.5

Ja

n-1

2A

pr-

12

Ju

l-12

Oct-

12

Ja

n-1

3A

pr-

13

Ju

l-13

Oct-

13

Ja

n-1

4A

pr-

14

Ju

l-14

Oct-

14

Ja

n-1

5A

pr-

15

Ju

l-15

Oct-

15

Ja

n-1

6A

pr-

16

Ju

l-16

Oct-

16

Ja

n-1

7A

pr-

17

Ju

l-17

Oct-

17

Ja

n-1

8A

pr-

18

Ju

l-18

Oct-

18

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

E2Q

E3Q

E4Q

E1Q

E2Q

E3Q

E4Q

E

CY2012 CY2013 CY2014 CY2015 CY2016 CY2017 CY2018

Rela

tive P

/B (

x)

SP

E o

rder

(consolid

ate

d, (¥

bn)

Order TOPIX rel P/B

10 March 2017


Back-end SPE: Sensor-related investment firm; no

significant progress in FO-WLP/PLP

Key points

■ We believe the major Taiwanese OSAT companies will request that delivery and

installation of new equipment be delayed as a result of the release of new smartphone

APs being pushed back (by around three months, we believe).

■ Assembly SPE investment among OSATs remains firm in the areas of fingerprint

sensor and NAND equipment.

■ There has been additional investment in certain types of InFO-WLP-related SPE to

cope with an increase in volume in 2017. A number of 2G InFO-WLP projects are

taking place simultaneously and a new type of 450mm FO-WLP is under consideration.

In addition, construction of new plants is scheduled to be completed in 2018.

■ LCD manufacturers AUO and Innolux are considering FO-PLP production with a view

to resuming use of existing 3.5G lines.

■ A US memory manufacturer is planning to establish a back-end processing line at its

Taichung facility. It will conduct mobile RAM testing there initially and eventually

construct an assembly line.

■ We have confirmed investment in SPE related to iris sensors, which are widely

expected to be used in the new iPhone.

■ Large-scale LED-related SPE investment (mainly in China) has resumed after a pause

of several years.

Capex trends at OSATs

Assembly SPE capacity utilization at Taiwanese OSATs is down moderately QoQ, but is

still high relative to a normal 1Q since foundry 300mm capacity utilization remains high.

However, the OSATs remain cautious in their stance toward investing in augmenting

manufacturing capacity. The release of new smartphone APs has been pushed back, and

in connection with this, we believe the OSATs have asked for the delivery of already

ordered equipment to be delayed. However, we think the delay will be short, as the

technological issues have already been solved. We do not foresee the OSATs asking for a

delay of equipment delivery or canceling orders outright because of a correction in

semiconductor market prices.

Amid these developments, capex among OSATs remains strong in the areas of SPE for

fingerprint sensors and NAND. OSATs made investments in the latter half of 2016 in

connection with a shift in production from 200mm wafers to 300mm wafers. We now

believe they are investing in equipment to meet increased demand for fingerprint sensors

for use in mid-range and low-end smartphones. (Business plans for 2017 among major

OSATs involve a YoY increase in investment related to fingerprint sensors.)

Strong investment at OSATs related to

fingerprint sensors, NAND

10 March 2017


Figure 48: OSATs: Assembly SPE capacity utilization forecasts


FO-WLP/PLP: No progress toward mass production, but LCD panel makers starting to consider adoption

The InFO-WLP manufactured by TSMC is projected to have essentially the same specs as

the 10nm-process APs going into mass production in 2017. However, with volumes rising,

we believe the Taiwanese foundries are placing additional orders for certain types of

equipment, such as molding systems. Significant spec changes are expected for 2G in

2018, and at present several projects are under consideration. This involves more than a

simple narrowing of line widths to, for example, from 5μm down to 2μm on the rewire

layer. Specifically, there are two main approaches under consideration. The first approach

is InFO+TSV technology, in which the DRAM is connected to the TSV attached to the AP,

rather than to the POP package. The second approach is a technology in which the AP

and the DRAM are mounted horizontally, rather than using a POP structure, making the

package extremely thin. In addition, 450mm FO-WLP has recently started to be

considered. In addition to existing mass production lines, new factories are currently being

built and are scheduled to go into operation in 2018.

We believe Samsung Electronics is pursuing two next-generation package development

projects, one in-house and one at subsidiary SEMCO. In the company’s own project, an

interposer PoP package similar in structure to Shinko Electric’s MCeP is currently under

development, but we believe yields are still only around 50%, so there are a number of

issues to solve before mass production can begin. In the other project, SEMCO is making

progress in FO-PLP development and expects to begin mass production of FO-PLP APs

as early as 2018. The company plans to initially gain experience in mass production of

FO-PLPs for devices with small chip sizes such as PMICs and modems, and to invest in

mass production equipment around September 2017. However, the company’s intra-group

smartphone and LSI business units have expressed no interest in adopting FO-WLP/PLP

for their APs.

Qualcomm is also reluctant about adopting FO-WLP/PLP for its APs. Qualcomm uses

Shinko Electric’s MCeP technology in its 10nm-process packages and expects to continue

doing so through its new products to be developed in 2H 2018 (for release in 1H 2019).

For that reason, we do not expect Qualcomm to have 7nm-process products until 2019 at

the earliest.

Amkor Technology Korea has finished constructing a K5 line, and we believe it has begun

full-scale SWIFT development.

40%

50%

60%

70%

80%

90%

100%

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 E

Taiwan Assemblyprocess utilization

Assembly

FO-WLP/PLP: No major progress, considering

450mm InFO

10 March 2017


Taiwanese LCD panel manufacturers AUO and Innolux have expressed interest in FO-

PLP production as a means of resuming use of their existing G3.5 lines. They have

already begun meeting with equipment manufacturers. Although AUO and Innolux are

capable of utilizing such equipment, this would be their first experience in back-end

semiconductor assembly. Since they are only at the examination stage, we think that at

this point it is hard to expect any particular level of demand for equipment.

Tester market trends: Tester investment may be peaking right now

In light of DRAM demand and technological changes, smartphone manufacturers are

increasingly shifting from LP DDR3to LP DDR4 for mobile RAM (mainly in their flagship

models). We believe they will do the same for their mid-range and low-end models in 2H

2017. We also project that Chinese smartphone manufacturers will adopt faster LP DDR4

in their spring models (moving from 3Gbps speeds to 4Gbps). Samsung Electronics made

a large-scale investment in front-end fabrication process testers in December 2016.

However, any contraction of test periods as a result of increased tester throughput would

not likely lead to demand for back-end assembly process testers in 2H.

In addition, owing to a change in testing procedures, a major Korean DRAM manufacturer

has adopted hybrid testers (made by a local Korean tester manufacturer) that carry out

both burn-in and core tests. The company also plans to reduce the number of front-end

fabrication process tests. We think it will also purchase back-end process testers primarily

from a local company.

In the wake of an increase in 3D NAND manufacturing capacity at Samsung Electronics,

demand for NAND wafer probe testing equipment increased in 1Q. As with DRAM, local

Korean tester manufacturers have a presence in this field, and we believe one such

company (YIK, formerly the Korean subsidiary of Yokogawa Electric) has a nearly 50%

share of the market for front-end fabrication process testers (with the remaining market

share belonging to Advantest).

We believe Qualcomm is likely to delay the release of new AP products and thus see a

risk of demand for logic testers being pushed back as well.

Figure 49: Disco ‒ Quarterly orders forecasts Figure 50: Advantest ‒ Quarterly orders forecasts


21.8

26.6 26.1

19.3

24.1

28.4

25.2 23.4 26.1

35.5

31.3

27.5

35.0 34.5 33.4

28.0 30.8

32.1 33.3 31.7

37.0

40.0

35.0 33.0

35.0 38.0

35.0 33.0

0.00

0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

Ja

n-1

2A

pr-

12

Ju

l-12

Oct-

12

Ja

n-1

3A

pr-

13

Ju

l-13

Oct-

13

Ja

n-1

4A

pr-

14

Ju

l-14

Oct-

14

Ja

n-1

5A

pr-

15

Ju

l-15

Oct-

15

Ja

n-1

6A

pr-

16

Ju

l-16

Oct-

16

Ja

n-1

7A

pr-

17

Ju

l-17

Oct-

17

Ja

n-1

8A

pr-

18

Ju

l-18

Oct-

18

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

50.0

1Q

2Q

3Q

4Q 1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q 4Q

1Q 2Q

3Q

4Q

1Q

E2Q

E3Q

E4Q

E1Q

E2Q

E3Q

E4Q

E

CY2012CY2013CY2014CY2015CY2016CY2017CY2018

TO

PIX

rel P

/B (

x)

Ord

er

(¥ b

n)

Order TOPIX rel P/B

34.3

46.2

25.3 24.4

29.8

39.5

21.4

27.4

39.1

49.2

38.2 38.7

50.2

44.0

27.1

39.8 40.6 43.8

28.5

44.0 39.0 44.0

38.0 35.0

40.0

37.0 35.0

43.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Ja

n-1

2A

pr-

12

Ju

l-12

Oct-

12

Ja

n-1

3A

pr-

13

Ju

l-13

Oct-

13

Ja

n-1

4A

pr-

14

Ju

l-14

Oct-

14

Ja

n-1

5A

pr-

15

Ju

l-15

Oct-

15

Ja

n-1

6A

pr-

16

Ju

l-16

Oct-

16

Ja

n-1

7A

pr-

17

Ju

l-17

Oct-

17

Ja

n-1

8A

pr-

18

Ju

l-18

Oct-

18

0.0

10.0

20.0

30.0

40.0

50.0

60.0

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

2Q

3Q

4Q

1Q

E2Q

E3Q

E4Q

E1Q

E2Q

E3Q

E4Q

E

CY2012 CY2013 CY2014 CY2015 CY2016 CY2017 CY2018

TO

PIX

rel P

/B (

x)

Ord

ers

(¥ b

n)

Order TOPIX rel P/B

Taiwanese LCD panel makers interested in

FO-PLP

DRAM testers: Faster LP DDR4, adoption of LP DDR4 for Chinese

mid-range smartphones driving

investment.

NAND testers: Demand increasing owing to

new factories

10 March 2017


Companies Mentioned (Price as of 09-Mar-2017) AUO (AUO.N, $3.84) Advantest (6857.T, ¥1,978) Alibaba Group Holding Limited (BABA.N, $103.24) Amkor Technology Inc. (AMKR.OQ, $10.55) DISCO (6146.T, ¥17,720) Hi Silicon (Unlisted) Hitachi Kokusai Electric (6756.T, ¥2,608) Innolux Corporation (3481.TW, NT$12.25) Intel Corp. (INTC.OQ, $35.82) Lenovo Group Ltd (0992.HK, HK$4.66) MediaTek Inc. (2454.TW, NT$218.0) Memblaze (Unlisted) Micron Technology Inc. (MU.OQ, $25.15) Nanya Technology (2408.TW, NT$46.55) OmniVision Tech (OVTI.OQ^A16) OmniVision Tech (OVTI.OQ^A16) Powerchip (5346q.L, $0.0) QUALCOMM Inc. (QCOM.OQ, $57.97) SCREEN (7735.T, ¥7,590) SK Hynix Inc. (000660.KS, W48,000) SMIC (0981y.F, €6.063) Samsung Electro-Mechanics (009150.KS, W58,600) Samsung Electronics (005930.KS, W2,010,000) Shannon Systems (Unlisted) Taiwan Semiconductor Manufacturing (2330.TW, NT$184.5) Tokyo Electron (8035.T, ¥11,615) Toshiba (6502.T, ¥205) United Microelectronics (2303.TW, NT$12.1) Vivo (Unlisted) XMC (Unlisted) Yangtze River Storage Tech (Unlisted) Yokogawa Electric Corp (6841.T, ¥1,793)

Disclosure Appendix

Analyst Certification Hideyuki Maekawa, Akinori Kanemoto and Mika Nishimura each certify, with respect to the companies or securities that the individual analyzes, that (1) the views expressed in this report accurately reflect his or her personal views about all of the subject companies and securities and (2) no part of his or her compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report. The analyst(s) responsible for preparing this research report received Compensation that is based upon various factors including Credit Suisse's total revenues, a portion of which are generated by Credit Suisse's investment banking activities

As of December 10, 2012 Analysts’ stock rating are defined as follows: Outperform (O) : The stock’s total return is expected to outperform the relevant benchmark* over the next 12 months. Neutral (N) : The stock’s total return is expected to be in line with the relevant benchmark* over the next 12 months. Underperform (U) : The stock’s total return is expected to underperform the relevant benchmark* over the next 12 months. *Relevant benchmark by region: As of 10th December 2012, Japanese ratings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractiv e, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ra tings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin Ame rican 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. Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances. Not Rated (NR) : Credit Suisse Equity Research does not have an investment rating or view on the stock or any other securities related to the company at this time. Not Covered (NC) : Credit Suisse Equity Research does not provide ongoing coverage of the company or offer an investment rating or investment view on the equity security of the company or related products.

Volatility Indicator [V] : A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward.

Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation: Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months. Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months. Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months. *An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cover multiple sectors.

10 March 2017


Credit Suisse's distribution of stock ratings (and banking clients) is:

Global Ratings Distribution

Rating Versus universe (%) Of which banking clients (%) Outperform/Buy* 44% (64% banking clients) Neutral/Hold* 39% (60% banking clients) Underperform/Sell* 14% (53% banking clients) Restricted 2% *For purposes of the NYSE and FINRA ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, a nd Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdings, and other individual factors.

Important Global Disclosures Credit Suisse’s research reports are made available to clients through our proprietary research portal on CS PLUS. Credit Suisse research products may also be made available through third-party vendors or alternate electronic means as a convenience. Certain research products are only made available through CS PLUS. The services provided by Credit Suisse’s analysts to clients may depend on a specific client’s preferences regarding the frequency and manner of receiving communications, the client’s risk profile and investment, the size and scope of the overall client relationship with the Firm, as well as legal and regulatory constraints. To access all of Credit Suisse’s research that you are entitled to receive in the most timely manner, please contact your sales representative or go to https://plus.credit-suisse.com . Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein. Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html . Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties.

For date and time of production, dissemination and history of recommendation for the subject company(ies) featured in this report, disseminated within the past 12 months, please refer to the link: https://rave.credit-suisse.com/disclosures/view/report?i=289184&v=65m3sqc61qguch7csos8xy3ov .

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. Principal is not guaranteed in the case of equities because equity prices are variable. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. This research report is authored by: Credit Suisse Securities (Japan) Limited ........................................................................ Hideyuki Maekawa ; Akinori Kanemoto ; Mika Nishimura To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important disclosures regarding any non-U.S. analyst contributors: The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts listed below may not be associated persons of CSSU and therefore may not be subject to the FINRA 2241 and NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account. Credit Suisse Securities (Japan) Limited ........................................................................ Hideyuki Maekawa ; Akinori Kanemoto ; Mika Nishimura

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.

https://plus.credit-suisse.com/

https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html

https://www.credit-suisse.com/sites/disclaimers-ib/en/managing-conflicts.html

https://rave.credit-suisse.com/disclosures/view/report?i=289184&v=65m3sqc61qguch7csos8xy3ov

https://rave.credit-suisse.com/disclosures/view/report?i=289184&v=65m3sqc61qguch7csos8xy3ov

10 March 2017


This report is produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who-we-are This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates.The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk.

This report is issued and distributed in European Union (except Switzerland): by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. Germany: Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). United States and Canada: Credit Suisse Securities (USA) LLC; Switzerland: Credit Suisse AG; Brazil: Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; Mexico: Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); Japan: by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau ( Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; Hong Kong: Credit Suisse (Hong Kong) Limited; Australia: Credit Suisse Equities (Australia) Limited; Thailand: Credit Suisse Securities (Thailand) Limited, regulated by the Office of the Securities and Exchange Commission, Thailand, having registered address at 990 Abdulrahim Place, 27th Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok10500, Thailand, Tel. +66 2614 6000; Malaysia: Credit Suisse Securities (Malaysia) Sdn Bhd; Singapore: Credit Suisse AG, Singapore Branch; India: Credit Suisse Securities (India) Private Limited (CIN no.U67120MH1996PTC104392) regulated by the Securities and Exchange Board of India as Research Analyst (registration no. INH 000001030) and as Stock Broker (registration no. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777; South Korea: Credit Suisse Securities (Europe) Limited,

Seoul Branch; Taiwan: Credit Suisse AG Taipei Securities Branch; Indonesia: PT Credit Suisse Securities Indonesia; Philippines:Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Additional Regional Disclaimers Hong Kong: Credit Suisse (Hong Kong) Limited ("CSHK") is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an Australian financial services licence (AFSL) and is exempt from the requirement to hold an AFSL under the Corporations Act 2001 (the Act) under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Act). Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Australia (to the extent services are offered in Australia): Credit Suisse Securities (Europe) Limited (“CSSEL”) and Credit Suisse International (“CSI”) are authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority (“FCA”) and the Prudential Regulation Authority under UK laws, which differ from Australian Laws. CSSEL and CSI do not hold an Australian Financial Services Licence (“AFSL”) and are exempt from the requirement to hold an AFSL under the Corporations Act (Cth) 2001 (“Corporations Act”) under Class Order 03/1099 published by the Australian Securities and Investments Commission (“ASIC”), in respect of the financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). This material is not for distribution to retail clients and is directed exclusively at Credit Suisse's professional clients and eligible counterparties as defined by the FCA, and wholesale clients as defined under section 761G of the Corporations Act. Credit Suisse (Hong Kong) Limited (“CSHK”) is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Credit Suisse Securities (USA) LLC (CSSU) and Credit Suisse Asset Management LLC (CSAM LLC) are licensed and regulated by the Securities Exchange Commission of the United States under the laws of the United States, which differ from Australian laws. CSSU and CSAM LLC do not hold an AFSL and is exempt from the requirement to hold an AFSL under the Corporations Act under Class Order 03/1100 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Corporations Act). Malaysia: Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. Singapore: This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore Branch to overseas investors (as defined under the Financial Advisers Regulations). Credit Suisse AG, Singapore Branch may distribute reports produced by its foreign entities or affiliates pursuant to an arrangement under Regulation 32C of the Financial Advisers Regulations. Singapore recipients should contact Credit Suisse AG, Singapore Branch at +65-6212-2000 for matters arising from, or in connection with, this report. By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore Branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the “FAA”), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore Branch may provide to you. UAE: This information is being distributed by Credit Suisse AG (DIFC Branch), duly licensed and regulated by the Dubai Financial Services Authority (“DFSA”). Related financial services or products are only made available to Professional Clients or Market Counterparties, as defined by the DFSA, and are not intended for any other persons. Credit Suisse AG (DIFC Branch) is located on Level 9 East, The Gate Building, DIFC, Dubai, United Arab Emirates. EU: This report has been produced by subsidiaries and affiliates of Credit Suisse operating under its Global Markets Division In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-US customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. US customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the US. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority or in respect of which the protections of the Prudential Regulation Authority and Financial Conduct Authority for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case are available upon request in respect of this report. CS may provide various services to US municipal entities or obligated persons ("municipalities"), including suggesting individual transactions or trades and entering into such transactions. Any services CS provides to municipalities are not viewed as "advice" within the meaning of Section 975 of the Dodd-Frank Wall Street Reform and Consumer Protection Act. CS is providing any such services and related information solely on an arm's length basis and not as an advisor or fiduciary to the municipality. In connection with the provision of the any such services, there is no agreement, direct or indirect, between any municipality (including the officials,management, employees or agents thereof) and CS for CS to provide advice to the municipality. Municipalities should consult with their financial, accounting and legal advisors regarding any such services provided by CS. In addition, CS is not acting for direct or indirect compensation to solicit the municipality on behalf of an unaffiliated broker, dealer, municipal securities dealer, municipal advisor, or investment adviser for the purpose of obtaining or retaining an engagement by the municipality for or in connection with Municipal Financial Products, the issuance of municipal securities, or of an investment adviser to provide investment advisory services to or on behalf of the municipality. If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice by Credit Suisse to the clients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any liability whatsoever for any direct or consequential loss arising from their use of this report or its content. Principal is not guaranteed. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Copyright © 2017 CREDIT SUISSE AG and/or its affiliates. All rights reserved.

Investment principal on bonds can be eroded depending on sale price or market price. In addition, there are bonds on which investment principal can be eroded due to changes in redemption amounts. Care is required when investing in such instruments. When you purchase non-listed Japanese fixed income securities (Japanese government bonds, Japanese municipal bonds, Japanese government guaranteed bonds, Japanese corporate bonds) from CS as a seller, you will be requested to pay the purchase price only.

https://www.credit-suisse.com/who-we-are

Documents

10 March 2017 Asia Pacific/Japan