Cisco Systems Inc. (CSCO) - Credit Suisse | PLUS

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

CREDIT SUISSE SECURITIES RESEARCH & ANALYTICS BEYOND INFORMATION®

Client-Driven Solutions, Insights, and Access

18 September 2013

Americas/United States

Equity Research

Networking Equipment

Cisco Systems Inc. (CSCO) INITIATION

Software Defined Disruption ■ Initiate with Underperform—Now for the Hard Part: Over the past year,

Cisco has seen the benefits of improving execution, gross margin stability, and increased capital redeployment, all of which are to be applauded. However, the impact on the networking market from Software Defined Networking (SDN) architectures should be very real and accelerate over time, leading to further margin compression. We initiate with an Underperform rating and a target price of $21.

■ SDN to Threaten the Most Profitable Part of the Stack: Currently, on the hardware side, networking stands as one of the least competitive and highest margin industries in the IT stack. SDN, by allowing software to be separate from the physical infrastructure, will allow competition at multiple points in the network, which was not previously possible. While the impact will take time, the threat will be very real, shrinking gross profit dollars for the industry. Irrespective of whether Cisco executes well in SDN environments, the company is in a vulnerable position amidst this transition.

■ Margins to Continue Their Long-Term Trend Downward: Based on five factors, we estimate that gross margins will decrease to ~60% and operating margins to 26% over the long term, with limited growth in organic operating income. We note the following: (1) our proprietary segmentation analysis shows gross margins by business line; within that UCS is materially dilutive at ~30% GMs; (2) while software and services can offset some of the headwinds, gaps in Cisco's IT stack mean that success in wider IT environments cannot be guaranteed; (3) SDN has the potential to shrink gross profit dollars available to vendors by as much as 70%; (4) Huawei will have an effect on the service provider market and internationally; and (5) the opex base is quite lean and further optimization will likely prove challenging.

■ Valuation—Target Price of $21: While we do not see significant downside risk on Cisco, we also do not see significant upside potential. Our TP of $21 is a blended average of P/E, DCF, and Credit Suisse HOLT

® valuation.

Share price performance

16

21

26

Sep-12 Dec-12 Mar-13 Jun-13 Sep-13

Daily Sep 17, 2012 - Sep 17, 2013, 9/17/12 = US$19.16

Price Indexed S&P 500 INDEX

On 09/17/13 the S&P 500 INDEX closed at 1704.76

Quarterly EPS Q1 Q2 Q3 Q4 2012A 0.43 0.47 0.48 0.47 2013A 0.48 0.51 0.51 0.52 2014E 0.51 0.51 0.53 0.55

Financial and valuation metrics

Year 07/12A 07/13A 07/14E 07/15E EPS - (Excl. ESO) (US$) 1.85 2.02 2.10 2.13 EPS (CS adj.) (US$) 1.85 2.02 2.10 2.13 Prev. EPS (CS adj.) (US$) — — — — P/E (CS adj., x) 13.1 12.0 11.6 11.4 P/E rel. (CS adj., %) — 73.7 75.4 82.4 Revenue (US$ m) 46,061.0 48,607.0 50,822.9 53,655.8 EBITDA (US$ m) 15,338.0 16,033.0 16,289.8 16,322.9 Net debt (US$ m) 6,529 8,286 2,833 -2,316 OCFPS (US$) 2.13 2.40 2.47 2.47 P/OCF (x) 7.5 10.6 9.9 9.9

Number of shares (m) 5,361.55 Price/sales(x) 2.62 BV/share (Next Qtr., US$) — P/BVPS (x) 2.2 Net debt (Next Qtr., US$ m) — Dividend (current, US$) — Dividend yield (%) —

Source: Company data, Credit Suisse estimates.

Rating UNDERPERFORM* Price (17 Sep 13, US$) 24.37 Target price (US$) 21.00¹ 52-week price range 26.37 - 16.82 Market cap. (US$ m) 130,660.97

*Stock ratings are relative to the coverage universe in each

analyst's or each team's respective sector.

¹Target price is for 12 months.

Research Analysts

Kulbinder Garcha

212 325 4795

[email protected]

Andrew Ruben

212 325 4798

[email protected]

Vlad Rom

212 325 5442

[email protected]

Matthew Cabral

212 538 6260

[email protected]

Talal Khan, CFA

212 325 8603

[email protected]

Ray Bao

212 325 1227

[email protected]

18 September 2013

Cisco Systems Inc. (CSCO) 2

Table of Contents Proprietary Profits: How Long? 3 Networking to See Moderate Growth 5

A Nexus of Forces at Play for Networking 5 Carrier—Capex Moderate Growth 9 Enterprise Switching and Routing: +3.5% CAGR 17 Service Provider Switching & Routing: +5.5% CAGR 21

Scorecard—Cisco wins 26 What Makes a Networking Vendor Successful? 30

Wireless LAN—Secular Growth 33 Competitive Dynamics—Cisco Dominates 35

SDN—Reorienting Networking 39 SDN—Gradual but real adoption 42 Software Disrupting Networking 45

Networking—Is It So Special? SDN May Change That 46 Hardware and Software Decouple—Lessons from Sun 49 SDN Drives a Major Headwind on Profit Dollars 52 Mean Reversion on Margins 55

UCS—Gaining Share in Servers 57 x86 Server Outlook—Pockets of Strength 58 x86 Server Market—Positioning Critical 60 Cisco Gaining Share with UCS 63

Aiming for the IT Stack 68 Chasing TAM Expansion Is the Norm 68

IT Strategies in Evolution 72 The Industry Says…Cisco, IBM, and MSFT 73

Major Headwinds on Margins 76 Segmentation Analysis—UCS Dilutive 79 Building Software and Services Some Challenges 80 Mean Reversion on Margins 84 SDN Drives a Major Headwind on Profit Dollars 85 ASIC Strategy a Differentiator in an SDN World? 89 Huawei a Threat Not to Be Underestimated 90 Cost Management Has Been Strong of Late 94

Portfolio Refresh Helping Share 97 Switching Portfolio Refresh 97

Capital Distributions Could Move Higher 100 Could Cisco Do More? 100

Valuation—FV of $21 Per Share 104 Risks: Where Could We Be Wrong? 110 Management Team 113 Financial Statements 114

18 September 2013


Proprietary Profits: How Long? If there has been one constant in networking over the past 15 years, it has been Cisco.

Through multiple waves of technologies and competitors, Cisco has been able to

out-execute its rivals. At the same time, the company, along with Juniper, has benefited

from a highly vertically integrated structure in networking, where hardware, software, and

services are bundled. Combining this with the perceived value of networking within the IT

space and its impact on other areas has shown that network administrators are loathed to

disrupt it too rapidly. All these factors have led to an oligopolistic industry structure, where

vendors have been able to extract margins that resemble more of a monopoly. One

example of this is the gross margins in switching, which are estimated at 65-70%, notably

higher than other IT hardware categories such as servers and storage.

Exhibit 1: Cisco Summary Income Statement in US millions, unless otherwise stated

Revenue by segment FY12 FY13 FY14E FY15E FY16E

Switching 14,531 14,741 15,239 15,734 16,049

Routing 8,425 8,230 8,522 9,136 9,501

SP Video 3,858 4,852 4,836 4,806 4,857

DC 1,298 2,073 2,516 2,893 3,182

Wireless 1,669 2,166 2,708 3,077 3,385

Security 1,344 1,347 1,374 1,429 1,472

Collaboration 4,182 3,956 3,963 4,083 4,207

Other 1,019 664 398 359 323

Services 9,735 10,578 11,267 12,139 12,989

Revenue 46,061 48,607 50,823 53,656 55,965

yoy 6.6% 5.5% 4.6% 5.6% 4.3%

Consensus 50,944 53,895 57,639

Gross Profit 28,763 30,400 31,563 32,867 33,700

Gross Margin (%) 62.4% 62.5% 62.1% 61.3% 60.2%

Operating Income 12,736 13,682 14,431 14,654 14,727

Operating Margin (%) 27.7% 28.1% 28.4% 27.3% 26.3%

Net Income 10,017 10,866 11,485 11,714 11,815

Share Count 5,405 5,369 5,462 5,502 5,542

EPS ($) 1.85 2.02 2.10 2.13 2.13

yoy 14.4% 9.2% 3.9% 1.3% 0.1%

Consensus 2.11 2.29 2.49

Source: Factset, Company data, Credit Suisse estimates.

Simply put, we believe that the gradual onset of Software Defined Networking (SDN) will

radically change networking economics, driving a more competitive structure and

sustaining the long-term downward trend on gross margins. This implies that, while Cisco's

operating margin shrinks over time, EPS can remain flat driven by capital distribution.

Given this, we initiate coverage on Cisco with an Underperform rating and target price of

$21.

SDN—Some real drivers. The onset of SDN is driven by a very real and genuine

phenomena, which mainly includes changing and rapidly growing datacenter traffic and

the need for more flexible and automated application-specific networks. At its basic level,

the onset and uptake of SDN means that, while previous networking architectures

consisted of bundled hardware and software, these can now be separated effectively into

multiple planes, with functionality being increasingly abstracted away from hardware to

enable aggregation and management, functionality, and automation. To be clear, we

believe that the shift will be gradual, as our survey shows most enterprises currently are in

the proof-of-concept stage, with initial deployments over 18 months away, and production

18 September 2013


in scale to be a 2-5 years away. However, this new way of operating will bring with it new

competition at multiple levels, including at the hardware/data plane, the controller,

applications, and orchestration levels. The end result, in our view, is a much more

competitive structure for the networking markets, and consequently, lower margins.

Margins—Downward pressure. We reviewed the outlook for margins through five different

approaches, and the bias to profitability remains to the downside. As shown in Exhibit 104,

we project 26% long-term operating margins, given several factors:

Segmentation analysis—Negative UCS mix shift against the software and services upsell.

Our proprietary segmentation analysis for Cisco’s gross margins by business segment

reveals several major conclusions. First, and not surprisingly, UCS is materially gross

margin dilutive; we believe margins are around 30%, half of corporate levels, given that

growth will increasingly come from gaining share and, given the company's server

exposure, higher scale seems unlikely. Second, approximately 44% of incremental gross

profit dollars that Cisco targets in the medium-term come from the ability to sell additional

services. This strategy centers around both creating software-based solutions and

upselling incremental software functionality, which may prove difficult, especially given the

historically bundled nature of networking spend. Third, we forecast that there will be a

disruption to networking gross margins over time from SDN, which will result in profit

compression.

SDN could make networking look more like other areas of IT spend. Over a five-year view,

we believe SDN will alter the dynamics of the networking market, whereby at its most

basic level, hardware, software, and services may be delivered in a far less integrated

manner. Although it remains to be seen how the traditional networking vendors adapt to

this change, based on our analysis, we believe that SDN will present a major change and

headwind to gross profit dollars. Indeed, the networking industry structure in the longer

term could more closely resemble the server or storage industry which are more

competitive, with gross margins between 15 and 30 percentage points lower. We

compared the cost and total cost of ownership, ex-labor, running costs, etc. for a basic

Cisco datacenter switch (Nexus 3064) to an alternative in an SDN world (Cumulus Linux

software and top-of-line rack bare metal switch). We demonstrate that the Cumulus option

is some 70% cheaper per year, and as a result, gross profits of hardware and software are

a combined 74% lower. In other words, even if Cisco were to embrace SDN, the gross

profit dollars available would significantly shrink. While this is simplistic in many ways, it

highlights the headwinds facing the traditional margin structure of Cisco and something

that the software and services upsell may not easily offset. To be clear, we do not assume

such drastic declines in gross margins, but equally we do believe the pressure over time is

downwards.

Cost management has been solid but in danger of perennial restructuring. Cisco has been

continuously focused on managing its operating costs, highlighted by the recent 4,000

employee workforce rebalancing. Given these efforts, we assume a long-term opex to

sales ratio of 34% is sustainable. We would also note that on a revenue-per-head basis,

compared with other major IT companies, Cisco looks lean.

In traditional networking, Cisco should continue to dominate. In the traditional enterprise

networking, especially in switching, Cisco’s strategy is by far the most robust. Cisco has

continued to out execute its rivals consistently in the networking over time. Our scorecard

also shows that customers have long favored Cisco, with the company ranked first on our

overall scorecard, including number one positions on six metrics, such as roadmap,

salesforce, service offering, and IT portfolio. This level of dominance, though, is arguably

already reflected in its market share. Equally interestingly, on the most important metrics

of price and performance and reliability, Cisco is essentially on par with Juniper and

Huawei. This suggests Cisco’s incumbency, rather than technology, is the key

differentiator for the company. The real concern, as we discuss in detail, is that SDN

threatens the very structure of the industry in which Cisco operates.

18 September 2013


Networking to See Moderate Growth Within the context of a cyclical spending recovery, our bottom-up model suggests that the

networking industry, in aggregate, will grow at a rate of 4.2% per year. Given the different

purchasing patterns and products, we segment the market into enterprise and service

provider, and within each of these, look at three distinct product lines: Switching, Routing,

and WLAN. In aggregate, these three markets have an addressable opportunity of some

$53bn in the long term. Within this market we several important growth drivers:

A nexus of forces at play. Fundamentally, both service provider and enterprise networks

have a significant number of forces to drive revenue growth over time. These include: (1)

smartphone and tablet driven data traffic; (2) growth of internet based video; (3) elastic

cloud computing and business analytics; and (4 aggregate trends in carrier capex. We do

believe that through cycle these will produce a networking segment (defined as enterprise

and service provider switching and routing) growth rate of 4.2% per year, modestly higher

that aggregate IT. We note that IDC expects IT spending growth of approximately 5% per

annum going forward; however, as detailed in our May 30, 2013 report, titled "IT

Hardware: "Un"clouding Storage – EMC Shines," we expect this to be significantly less

given the compression effects of cloud.

Enterprise Ethernet switching, multiple mix changes. In 2012, the enterprise switching

market amounted to $19.8 billion in sales and has grown at a rate of 3% per annum over

the past five years, something which we believe will continue in the long term. Within this

we see 10GbE adoption driving both port and revenue growth, stimulated by Intel's Xeon

server chip release code-named "Romley" as well as server virtualization. Among Layer 2,

Layer 3, and chassis, we see 10GbE port adoption to rise to 25% in 2017, up from 6%

currently and only 0.3% five years ago. Additionally, we see Layer 3 switches continuing to

gain wallet share at the expense of Layer 2. We believe the added intelligence of Layer 3

switches removes the need for the manual configuration historically required within

enterprise networks, which is especially important in virtualized infrastructures, as L3

switches are able to abstract away a myriad of VLANs. Finally, we see chassis switches

under macro and secular pressure.

Service provider routing. In examining trends within the service provider (SP) market, we

begin by focusing on routers that accounted $11.3 billion out of $14.2 billion in total service

provider spend in 2012. In revenue terms, the market has shown significant growth over

the past five years, expanding at a pace of 7.9% per annum, which is far faster than

growth on the enterprise side. Within this segment, we see core routers losing wallet share

within spend, as spending within the core of the network tends to be a function of capacity,

and recently service providers have been trying to prolong spend within the area as long

as possible. We see edge routing as the primary source of growth, rising 6.3% in the long

term. The tailwind to edge routing spend has been the tremendous growth in smartphones

and the resulting growth in mobile data consumption. We believe a key driver here will be

LTE that enables both voice and data to travel across IP networks, leading to converged

IP solutions across service providers' networks.

A Nexus of Forces at Play for Networking

We see several fundamental drivers at play for the networking market, as the major

secular trends of social, mobile, cloud, and big data analytics place considerable strains

on today’s networking infrastructure. These forces are playing out on both the carrier side

and enterprise side, essentially creating what Gartner describes as a nexus of forces. This

suggests that there remains healthy underlying growth ultimately driven by IP traffic growth

Global IP traffic growing at the rate of 23%. The Cisco Visual Networking Index highlights

that growth of aggregate IP traffic in the long term should be at a rate of 23% per year,

rising to 120 EB in 2017. Of this traffic, 68% will still reside on the fixed internet, whereas

23% will be on managed IP networks, and finally 9% will be driven from mobile data.

18 September 2013


Exhibit 2: Global IP Traffic Distribution... Exhibit 3: ...A Skew Towards Consumer Use

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

2012 2013 2014 2015 2016 2017Fixed Internet Managed IP Mobile data

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

2012 2013 2014 2015 2016 2017Consumer Business

Source: Cisco Visual Networking Index. Source: Cisco Visual Networking Index.

Smartphone and tablet adoption will be key. Long-term (2012-17) mobile data is expected

to grow at a CAGR of some 66%. Of this segment, 72% of this traffic will come from

smartphones with a further 13% from tablets. Here, the driver is not only the impact of user

growth of the installed base, but also the impact of the multiplier effect. For example,

compared with a non-smartphone user, a typical smartphone user may use 50 times as

much traffic, whereas a tablet user may use 120 times as much. (See Exhibit 5.)

Exhibit 4: Smartphones Lead Traffic Growth Exabytes per month

Exhibit 5: Summary of Per Device Usage Growth in MB per Month

0.0

2.0

4.0

6.0

8.0

10.0

12.0

2012 2013E 2014E 2015E 2016E 2017E

Exab

yte

s p

er

mo

nth

Smartphones Tablets Other Portables Laptops M2M Non-Smartphones

Device Type 2012

2012*

Multiple2017

2017*

Multiple

Non-Smartphones 7 1.0x 31 1.0x

M2M Module 64 9.4x 330 10.6x

Smartphone 342 50.3x 2,660 85.8x

4G Smartphone 1,302 191.5x 5,114 165.0x

Tablet 820 120.6x 5,387 173.8x

Laptop 2,503 368.1x 5,731 184.9x

* Multiple of non-smartphone device usage Source: Cisco Visual Networking Index. Source: Cisco Visual Networking Index.

Video highlights. It is expected that by 2017 consumer internet video traffic will be 69% of

all consumer internet traffic, up from 57% in 2012. If one were to include video exchanged

through P2P sharing, VoD, and TV, the range would be 80-90% of all global consumer

traffic by 2017. Within this, internet video to TV is expected to increase 5 times by 2017

and VoD traffic in 2017 is expected to be equivalent to 6bn DVDs per month. By 2017 65%

of all Internet video traffic is expected to cross content delivery networks in 2017, up from

53% in 2012.

18 September 2013


Exhibit 6: Internet Video Growing to ~70% of All Consumer Internet Traffic by 2017

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2012 2013 2014 2015 2016 2017

Internet video Web, Email and Data File Sharing Online Gaming

Source: Cisco Visual Networking Index.

Networking lifecycle. Like all IT hardware, appliances have an essential useful life. Several

factors determine the useful life of an asset including replacement costs, ongoing

operating costs, industry innovation, and vendor support. Based on Gartner data,

highlighted in Exhibit 7, the useful life for edge LAN switching is the longest at seven to ten

years, with core switching and routing coming in at a useful life of five to seven years.

Datacenter networking has a modestly lower useful life of four to seven years, which

makes sense given more disruptive changes.

Exhibit 7: Core Switching and Routing Product Life Is Somewhat Longer Than

Datacenter Networking Product

-

2.0

4.0

6.0

8.0

10.0

12.0

Useful Life Depreciation

Source: Gartner.

This corresponds with our own survey work that suggests that the average installed base

of enterprise switching equipment is 3.5 years. (See Exhibit 8.) As the age of the installed

base continues to move toward the useful life, in a normalized environment, we would

expect to see a replacement cycle across both switching and routing spend. However, it is

important to keep in mind that across the various end markets within IT hardware

18 September 2013


(Servers, Storage, PCs, Printers, etc.), organizations are choosing to run their existing

equipment longer, which prolongs the upgrade cycle.

Exhibit 8: On Average, the Installed Base of Ethernet Switches Is ~3.5 Years

Respondents asked the average life of an Ethernet switch

2.5

2.7

2.9

3.1

3.3

3.5

3.7

3.9

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Less than 1year

1-2 years 3-5 years 5-7 years 7-9 years 10 years ormore

Year

s

Perc

en

tage

of

resp

on

den

ts

Source: Credit Suisse Networking Survey August 2013.

Credit Suisse CIO Survey supportive of networking growth into 2014. Our proprietary

survey of 40 enterprise CIOs suggests that overall IT spending growth in 2014 will likely

mirror that of 2013. (See Exhibit 9).

Exhibit 9: IT Spending Growth in 2014 to Equal 2013

Respondents asked for y/y growth rates for IT spending

1.71% 1.78%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

2013 (vs. 2012) 2014 (vs. 2013)

YoY

gro

wth

in IT

sp

end

Source: Credit Suisse IT Spending Survey August 2013.

While not the highest priority item within enterprise budgets, networking spend continues

to stand out relative to other hardware markets and falls roughly in the middle of overall

priorities. (See Exhibit 10 and Exhibit 11.)

18 September 2013


Exhibit 10: Networking in the Middle of Priorities… % of respondents ranking IT segments as top spending areas

Exhibit 11: …as Reflected in Growth Expectations 2013 annual revenue growth for IT spending by category

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

Printing PC Server Services Software Networking Storage Security Overall IT MobileDevices and

Services

YoY

IT s

pen

d g

row

th f

or

20

13

Source: Credit Suisse CIO Survey August 2013. Source: Credit Suisse CIO Survey August 2013.

Carrier—Capex Moderate Growth

With 38% of the networking market coming from the service provider market, capital

expenditure trends and fundamentals remain important to the outlook for switching and

routing. Following our detailed analysis of wireless capex, for which we have updated our

proprietary model (which tracks spend at over 100 leading carriers globally), we see

accelerating growth ahead driven by a recovery with Western Europe, continued high

levels of spending in the United States, and a strong roll out in China for LTE.

Europe Weak Capex Trends So Far, Are Things about to Change?

Vodafone has been gradually increasing its capex investments in Europe recently.

Vodafone has been maintaining flat capex spend for its European operations similar to its

group level trend over the past two to three years. However, in its FY13 (ending March

2013), it is worth noting that although group capex was roughly flat, its capex spend in

Europe actually grew 7% yoy, mainly driven by investments in Germany and the UK.

Figure 12: Vodafone Capex Spend in Europe Has Already Seen Some Pick Up in Early 2013 in GBP millions, unless otherwise stated; Vodafone financial year ending March

Vodafone capex (£ mn) H1 10/11 H2 10/11 H1 11/12 H2 11/12 H1 12/13 H2 12/13 FY 09/10 FY 10/11 FY 11/12 FY 12/13

Vodafone Germany 342 482 410 470 445 628 766 824 880 1,073

% change yoy 3% 11% 20% -2% 9% 34% 8% 7% 22%

Vodafone UK 178 338 219 356 231 370 494 516 575 601

% change yoy 26% -4% 23% 5% 5% 4% 4% 11% 5%

Vodafone Italy 260 330 269 352 239 328 610 590 621 567

% change yoy -13% 6% 3% 7% -11% -7% -3% 5% -9%

Vodafone Spain 220 297 147 282 152 225 543 517 429 377

% change yoy 29% -20% -33% -5% 3% -20% -5% -17% -12%

Vodafone Other Europe 521 709 497 595 417 823 1,282 1,230 1,092 1,240

% change yoy 11% -13% -5% -16% -16% 38% -4% -11% 14%

Total Vodafone Europe 1,521 2,156 1,542 2,055 1,484 2,374 3,695 3,677 3,597 3,858

% change yoy 8% -6% 1% -5% -4% 16% 0% -2% 7%

Source: Company data, Credit Suisse research

But now Vodafone is looking to break the vicious circle with £6bn investment over three

years? At the time of the Verizon deal (September 2013), Vodafone also unveiled its plan

to increase investments significantly over the next three years to enhance network quality

and drive service leadership. As a reminder, Vodafone has maintained capex at around

£6.3bn over the past couple of years, but under its new Project Spring, the carrier plans to

invest £6bnof incremental investment over the next three years with the majority of spend

directed towards capex. This would imply a capex increase of slightly over 30% yoy from

the current level. In fact, as per Vodafone, it plans to spend 45-50% of the £6bn on the

18 September 2013


mobile network, with an additional 20-25% likely spent in the area of NGN, VDSL, and

selective fiber rollouts.

Figure 13: 65-75% of Incremental Investment under Vodafone's Project Spring Will Be Directed Towards Networks

Source: Vodafone.

"Capex arms race is better than a subsidy race" as per Vodafone CEO. While it is still

early days to see how other carriers in Europe react to this aggressive stance from

Vodafone in terms of increasing investment to improve upon the quality of the network, we

would note that this is the first time in years that we have seen some real intent from a

European carrier to focus on network quality as a form of creating differentiation among

the consumer base. Judging by comments from the CEO of Vodafone, it seems that this

shift in strategy from the company may eventually force other bigger players in Europe

(such as DT, Telefonica, and France Telecom) to follow suit, which could then lead to a

situation similar to that which we saw in the United States over the past two to three years.

"I don't know what a capex arms race is. If that's the case then it's better than a

subsidy race because it requires real long-term orientation. It requires taking

money out of some commercial costs, which could be good because at the end of the

day it's a positive for the industry if everybody focuses on quality and investment as

opposed to promotion and short-term. I don't think it will happen. I think some other

competitors, the ones with bigger shoulders, will follow and I think in the end it

will help the industry structure." – Vittorio Colao, CEO of Vodafone Group

(September 3, 2013)

Vodafone represents ~25% of Western European capex. What if others follow with capex

upgrades? We estimate Vodafone represents around 25% of overall wireless capex in

Western Europe. Given Vodafone has decided to increase its capex from £6.3bn to

around £8.3bn per annum over the next three years (implies around 30% increase), we

believe this may put pressure on other operators in the region to follow suit. On the other

hand, our European Telecoms team (Justin Funnell) notes that some of the challenger

operators in Europe continue to struggle owing to declining revenues and a lack of share

gains, for example Yoigo in Spain, Wind in Italy, Everything Everywhere and 3 in the UK,

and may lack the financial resources to increase capex investments. However, we believe

this decision from Vodafone is at least likely to put pressure on bigger operators in Europe

(Deutsche Telekom, Telefonica, and France Telecom) to respond by increasing

investments to improve network quality.

18 September 2013


Figure 14: Can the Capex to Sales Ratio in W. Europe Rise Similar to the United States?

13.3%12.4%

14.2%14.8%

16.2%

18.3%

10.0% 9.6% 9.9%

11.3%12.4% 12.1%

0%

5%

10%

15%

20%

2008 2009 2010 2011 2012 1H13

Wir

ele

ss

-C

ap

ex t

o S

erv

ice

Re

ve

nu

e (

%)

US W Europe

Source: Company data, Credit Suisse research, EU Telecoms Research team (Justin Funnell).

Capex to sales significantly behind in WE when compared with the United States. Capex

trends within Europe have been lackluster over the past few years with most carriers

looking to maintain flat investment levels. Further, with service revenues having been

under pressure, the capex to sales ratio has seen a slight improvement in spite of flat

capex. When comparing this with the United States, what is striking is that service

revenues have grown at around 5% pa, but wireless capex has grown significantly faster

resulting in the capex to sales ratio rising from nearly 12.5% in 2009 to slightly over 18% in

1H13. When comparing this with Western Europe, we see that the capex to sales ratio has

only picked up from around 10% in 2009 to 12% in 1H13.

United States to Continue to See High Levels of Capex Activity

What is worth noting is that wireless capex in the United States has seen strong growth

over the past few years. In fact, based on our estimates, we believe that wireless capex is

likely to have grown from $20bn in 2009 to around $34bn in 2013 (CAGR of 15%) driven

by a move from carriers to improve the quality of their networks in order to drive

differentiation and roll out of 4G/LTE networks. (See Exhibit 15.) Looking at guidance from

carriers, we believe that wireless capex spend may remain flat in 2014. We would equally

note that it is already running at a high level of slightly over $34bn per annum. In addition,

we believe there may be risks to the upside to 2014 guidance as carriers continue to look

for ways to differentiate by improving the quality of their 4G networks.

Exhibit 15: Wireless Capex in the United States Likely to Grow to $34bn in 2013, and Remain at There for 2014 in US$ millions, unless otherwise stated

Wireless Capex (US$ mn) Q112 Q212 Q312 Q412 Q113 Q213 2009 2010 2011 2012 2013E 2014E

Verizon 1,885 2,048 2,133 2,791 1,992 2,278 7,152 8,438 8,973 8,857 9,675 9,650

AT&T 2,324 2,345 2,709 3,422 2,500 3,450 6,066 9,171 9,759 10,800 12,750 13,000

Sprint 710 1,012 1,376 1,786 1,706 1,728 1,161 1,444 2,416 4,884 7,425 7,475

T-Mobile US 747 539 717 898 1,230 1,111 3,687 2,819 2,729 2,901 4,300 4,100

Leap 146 119 106 63 26 23 680 399 442 434 150 200

MetroPCS 144 182 262 258 832 790 890 846

Total (US$ mn) 5,956 6,245 7,303 9,217 7,454 8,590 19,578 23,061 25,208 28,722 34,300 34,425

% change yoy -2% -8% 21% 46% 25% 38% 0% 18% 9% 14% 19% 0%


Wireless capex to sales ratio also rose by around 4pp in the past three years. One of the

factors that has helped this continued increase in capex investment across all U.S. carriers

18 September 2013


has been the consistent level of growth in the services revenue. In fact, we note that the

cumulative wireless services revenues for the top four carriers in the United States

(Verizon, AT&T, Sprint, and T-Mobile US) have grown around 5% pa over the past three

years. (See Figure 16.) With carriers seeing their services revenue rise, they have

continued to focus on creating differentiation with a lot of focus on network quality. As

such, in spite of growth in services revenue, capex to services revenues for wireless

operations for the top four carriers in the United States have risen from 14% in 2010 to

over 18% in 1H13. (See Figure 17.)

Figure 16: U.S. Wireless Services Revenue Growing

at 5% pa

Figure 17: Wireless Capex to Service Revenue Rising in

the United States

-10%

-5%

0%

5%

10%

15%

20%

25%

2009 2010 2011 2012 1H13

Wir

ele

ss

Serv

ices

Reve

nu

e g

rwo

th (

%)

Verizon AT&T Sprint T-Mo US Top 4 US

13.3

%

12.4

%

14.2

%

14.8

%

16.2

%

18.3

%

0%

5%

10%

15%

20%

25%

30%

2008 2009 2010 2011 2012 1H13Wir

ele

ss

-C

ap

ex t

o S

erv

ice R

eve

nu

e (

%)

Verizon AT&T Sprint T-Mo US Top 4 US

Source: Company data, Credit Suisse research. Source: Company data, Credit Suisse research.

Entry of Softbank in the U.S. market also led to capex hikes across the board. Apart from

the rollout of LTE networks, another driver for the capex hike in the United States has

been the merger of Sprint and Softbank. Sprint had plans to increase its capex from $3bn

in 2011 to around $6bn in 2012 to deal with its network modernization plans and rollout

LTE networks. However, the entry of Softbank into the U.S. market resulted in further

capex intensity among carriers in an attempt to build a better quality network to drive

differentiation. Specifically, we would note the following at each of the top four operators in

the region:

■ Sprint/Softbank have noted that they plan to spend $8bn per year over 2013/2014 in

order to roll out LTE, up from the $5.4bn spent in 2012 on capex;

■ Also AT&T had increased its capex from $20bn in 2012 to $21bn in 2013. (Initially it

had increased its capex guidance for 2013 to $22bn in November 2012, but

subsequently lowered it to $21bn in the next few months as it was on track for LTE

rollouts). This would imply 6% yoy growth in total capex; however, within this we

believe that wireless capex may be up 18% yoy (already up 27% yoy in 1H13) as the

company continues to invest more in its wireless network as opposed to wireline;

■ Verizon was initially planning to keep its capex flat at $16.2bn for 2013, but recently

on its Q213 results it noted plans to increase it slightly from $16.4bn to $16.6bn (up

2% yoy at the midpoint of the range). More specifically, we would note that wireless

capex within this is likely to grow close to 10% yoy (already up 9% yoy in 1H13) as the

carrier continues to migrate a higher proportion of group capex to its wireless

operations;

■ T-Mobile USA and Metro PCS have also embarked upon an aggressive network

strategy since their merger announcement in late 2012. Although they initially were

planning to invest $4.7-4.8bn on capex in 2013 (up from a combined $3.7bn in 2012),

they have recently lowered it to $4.2-4.4bn. However, our EU Telecoms research team

(Justin Funnell) believes that the combined entity will likely maintain capex levels of

$4.1bn for 2014.

18 September 2013


LATAM to Pick Up as LTE Networks Are Beginning to Be Deployed

With deadlines in place for a rollout of LTE networks in Brazil, we believe all four operators

in the country—Vivo, Claro, TIM, and Oi—will likely step up their efforts to deploy LTE

networks across a number of cities over the next 12-24 months. In fact, as per the

coverage plans, it is expected that operators are likely to extend LTE coverage to all cities

with more than 500,000 population by May 2014. (See Exhibit 18.)

Exhibit 18: 4G Coverage Plans in Brazil Suggest Significant Activity in the Next 12

Months

Deadline Coverage Plans

Apr 2013 All cities hosting the Confederations Cup 2013

Dec 2013 All cities hosting and co-hosting the World Cup 2014

May 2014 All capitals and cities having more than 500K inhabitants

Dec 2015 All cities having more than 200K inhabitants

Dec 2016 All cities having more than 100K inhabitants

Dec 2017 All cities having between 30K and 100K inhabitants

Source: Teleco Brazil, Credit Suisse research.

4G population coverage to expand. Looking at 3G population coverage levels, we would

note that they have almost reached 90% levels in Brazil compared with around 65% in

2009. Now with LTE population coverage at around 20%, we believe that the number is

set to rise significantly over the next 12-24 months looking at the coverage plans in the

region. Also looking at the number of cities covered, we see that currently only 20 cities

are covered with LTE as opposed to around 3,400 cities with 3G coverage.

Figure 19: LTE Coverage Is Only Starting to Happen Now Figure 20: Significant Pick Up in Coverage to Come

Cities covered in Brazil

Operator 1Q13 2Q13 1Q13 2Q13

Vivo 3,122 3,131 0 17

Claro 1,119 1,187 6 12

TIM 593 898 0 6

Oi 734 865 0 6

Total 3,370 3,436 6 20

3G coverage 4G coverage

59%65%

73%

83%88% 89% 90%

3%

18%

0%

25%

50%

75%

100%

2008 2009 2010 2011 2012 1Q13 2Q13

Po

pu

lati

on

co

vera

ge (

%)

3G 4G Source: Teleco Brazil, Credit Suisse research. Source: Teleco Brazil, Credit Suisse research.

Some Slowdown in Japan Capex, but Softbank and KDDI Still Going Strong

Softbank and KDDI still going strong with capex investments. While KDDI was looking for

a small increase in its mobile capex budget for FY14 (around low single digit), on its recent

results, it indicated towards an additional investment of JPY 30bn to augment its existing

LTE infrastructure. This would result in around 12% growth in wireless capex at KDDI for

FY14 on our numbers. Similarly, Softbank continues to emphasize the importance of

network quality and has stated that it aims to increase its capex spend for FY13 (ending

March 2014) from around JPY 700bn to JPY 780bn (implying a 12% increase).

Some slowdown at DoCoMo, is it a temporary pause? However, at NTT DoCoMo after

having seen two years of growth in capex, the company is guiding for around a 7% decline

in the capex budget for FY13 (ending March 2014), as it believes it can more than offset

the growth in LTE-related capex by declines expected in 3G and other mobile related

areas. In fact, it expects LTE-related capex to grow from JPY 220bn in FY12 to JPY 355bn

in FY13, as it plans to increase the number of LTE base stations by around 50,000 by the

end of March 2014. However, DoCoMo still has only 23% of its subscribers on the LTE

network (smartphone penetration is around 33%), which means that there is still a lot of

room for growth for LTE adoption and which may lead to incremental capacity-related

spending going forward.

18 September 2013


Figure 21: Japan Capex to Continue to Grow Mid-Single Digit in Spite of Expected Slowdown at DoCoMo

Wireless Capex (JPY bn) Mar

2011

Jun

2011

Sep

2011

Dec

2011

Mar

2012

Jun

2012

Sep

2012

Dec

2012

Mar

2013

Jun

2012

Mar

2011

Mar

2012

Mar

2013

Mar

2014E

NTT DoCoMo 200 137 175 190 225 177 184 180 213 146 668 727 754 700

Softbank 170 101 120 115 180 106 152 190 250 181 421 516 698 780

KDDI 96 49 64 79 112 68 83 85 103 69 339 304 338 380

Total 466 287 360 384 517 352 419 454 565 395 1,428 1,547 1,790 1,860

% change 11% 9% 3% 11% 23% 16% 18% 9% 12% 8% 16% 4%


Capex Spend to Pick Up in Korea after Weak 1H13 Post Spectrum Auctions

Wireless capex spend in Korea has grown from KRW 2.9tn in 2010 to KRW 5.9tn in 2012

due to increased levels of investment on existing 3G capacity and preparing for the rollout

of LTE services. In addition, within total capex we also saw a significant shift towards

wireless operations, especially at KT Corp and LG U+. However, with the impending

spectrum auctions during Q313, capex spend in Korea took a significant pause in 1H13. In

fact, total capex spend in 1H13 in Korea was down 34% yoy but more importantly

wireless-related capex was down 46% yoy.

Figure 22: Korean Capex under Pressure in 1H due to Spectrum Auctions, Expect Strong Pick Up in 2H13

Wireless Capex (KRW bn) Q111 Q211 Q311 Q411 Q112 Q212 Q312 Q412 Q113 Q213 2010 2011 2012 2013E

SK Telecom 243 472 419 708 424 555 708 867 322 223 1,390 1,842 2,554 1,600

KT Corp 363 457 394 393 661 560 262 622 247 182 1,086 1,608 2,105 1,600

LG U+ 44 93 285 424 275 309 292 322 265 256 401 846 1,198 1,100

Total 650 1,022 1,098 1,525 1,360 1,424 1,262 1,811 834 661 2,876 4,296 5,857 4,300

% change 204% 70% 51% 14% 109% 39% 15% 19% -39% -54% 49% 36% -27%


However, guidance suggests 65% of FY capex to be spent in 2H13. The spectrum auction

results in Korea were determined in the last week of August, and our Korea Telecom

analyst (Taewon Kim) views these results to be a positive for the overall telecom sector in

Korea, given reasonable levels of cost involved in spectrum purchase. While overall

wireless capex for 2013 will be down around 25% yoy given the weak spend in 1H, we still

expect a material uptick in capex spend in 2H13. In fact, we estimate that carriers have

only spent 35% of their FY capex in 1H; we believe that 2H capex spend could be around

KRW 2.8tn (up around 88% hoh).

Figure 23: LTE Penetration in Korea Now Stands at over 40%; But Still Lots to Go

7%

13%

22%

29

%

36%

42%

0%

10%

20%

30%

40%

50%

60%

Q112 Q212 Q312 Q412 Q113 Q213

LT

E p

en

etr

ati

on

(%

)

SK Telecom KT Corp LG U+ Korea Market

Source: Company data, Credit Suisse research.

18 September 2013


LTE penetration continues to rise. Korea has been one of the most advanced markets

when it comes to LTE rollouts and levels of consumer adoption. While Korea has already

seen strong LTE penetration over the past few quarters (with 42% of mobile subscribers

now on LTE networks as opposed to only 13% a year ago, as we show in Figure 23);

however, we would note that it still stands at less than 45%. This in our view could result in

continued levels of investment from Korean operators as they prepare to deal with a

growing level of LTE subscribers on their network along with rising usage patterns.

China Pushing on LTE

At its recent 1H results, China Mobile (CM) reiterated its full-year capex plans of

Rmb190bn for 2013, up from Rmb127bn in 2012. Note these numbers do not include

TD-SCDMA (3G) related capex as, it is done at the parent company level, but does

include TD-LTE related capex, which is being planned at the listed company level. If we

were to include around Rmb20bn of TD-SCDMA related capex in our numbers for both

2012 and 2013, we estimate that the total capex at China Mobile may be around

Rmb210bn, up from close to Rmb150bn in 2012.

Exhibit 24: CM Has Retained Rmb190bn of Capex

Guidance Breakdown of capex at listed company and parent company levels

Exhibit 25: Significant Capex Budget for 2H due to LTE Breakdown of listed company capex at China Mobile (1H/2H)

136.3 129.4 124.3 128.5 127.4

190.2

30.0 40.023.0 21.0 20.7

22.0

166.3 169.4

147.3 149.5 148.1

212.2

0

50

100

150

200

250

2008 2009 2010 2011 2012 2013E

Cap

ex

at

Ch

ina

Mo

bile

(R

mb

bn

)

TD-SCDMA GSM, TD-LTE and Others

6671

6367

62 63 6267

59

69

57

133

0

20

40

60

80

100

120

140

1H

08

2H

08

1H

09

2H

09

1H

10

2H

10

1H

11

2H

11

1H

12

2H

12

1H

13

2H

13EC

hin

a M

ob

ile

lis

tco

ca

pex

1H

/2H

(R

mb

bn

)

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

This planned increase in capex is mainly driven by aggressive TD-LTE network

deployment plans at China Mobile given weak subscriber uptake of its TD-SCDMA (3G)

services, as opposed to other players in the market such as China Telecom and China

Unicom.

Figure 26: Increase in Mobile Comm Capex Due to LTE Planned breakdown of capex by area of spend

Figure 27: TD-LTE Related Capex of Rmb41.5bn for 2013 Breakdown of capex for 2013 in detail

54.773.2

56.1

79.9

28.6

23.135.7

59.0

124.3 128.5 127.4

190.2

0

20

40

60

80

100

120

140

160

180

200

2010 2011 2012 2013E

Ca

pe

x a

t C

hin

a M

ob

ile

lis

tco

(R

mb

bn

)

Others

Buildings &Infrastructure

SupportSystems

BusinessDevelopment

Transmission

MobileCommunicationNetworks

44%57%

44% 42%

23%

18%

28% 31%

13%6%

6% 6%

7% 6% 8% 6%

9% 9% 11% 13%4% 4% 3% 2%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2010 2011 2012 2013E

Others Buildings & Infrastructure

Support Systems Business Development

Transmission Mobile Communication Networks

TD-LTE52%

GSM + WLAN + Core48%

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

1H spending has been weak, so pick-up expected in 2H. Out of the Rmb190bn of FY

capex guidance from China Mobile at the listed company level, it has only spent Rmb57bn

18 September 2013


in 1H13, which implies around Rmb133bn to be spent in 2H13. This is surprisingly high in

our view, given the company has not spent this level of capex in any six-month period over

the past five years. So the question remains whether the company will spend all of the

Rmb41.5bn on TD-LTE in 2H13, or some of that spending may potentially get pushed in to

2014 as we see these networks being rolled out.

TD-LTE contracts seem to have been awarded to vendors by China Mobile. Over the past

few weeks, China Mobile has been in the process of attracting bids from various domestic

and international vendors for participating in the rollout of TD-LTE networks. According to

press reports in China, it seems that the carrier recently decided on its suppliers for

procuring TD-LTE equipment, and it has awarded 65% share of contracts to China-based

suppliers (such as Huawei, ZTE, and Datang), with the remaining 35% split between

European suppliers (Alcatel-Lucent, Ericsson, and Nokia Solutions Networks), with shares

split as shown below in Exhibit 28.

Exhibit 28: Share of China Mobile's TD-LTE Network for Rollout of 207K Base Stations

Vendor Contract Share (%)

International Alcatel-Lucent Shanghai Bell 13%

Ericsson 11%

Nokia Solutions Networks 11%

China-based Huawei 25%

ZTE 20%

Datang Mobile Communications Equipment 10%

FiberHouse Technologies, Nanjing Putian, New Postcom combined 10%

Source: DigiTimes.

18 September 2013


Enterprise Switching and Routing: +3.5% CAGR

Enterprise Ethernet Switching

Diving into specific segments of networking spend, we begin by exploring the enterprise

Ethernet switch market. In 2012, in aggregate, the market was $19.8 billion and has grown

at a rate of 3.0% per annum over the past five years. Within this, shifts are underway that

have important implications for the trajectory of the industry going forward, including

broader adoption of 10GbE ports and the preference for Layer 3 over Layer 2 switches.

10GbE adoption drives both port and revenue growth. While 10 gigabit Ethernet ports

have been available for nearly ten years, Intel's Xeon server chip release code-named

"Romley" paved the way for rampant adoption as it standardized 10GbE

"LAN-on-Motherboard" across both blade and rack-mounted servers. Further, server

virtualization is a secular driver of 10GbE adoption as increased utilization of the physical

compute hardware results in higher data traffic. While many enterprises continue to

operate with a mix of legacy and modern technologies, replacement cycles coupled with

new datacenter build-outs should continue to drive adoption within the enterprise switching

market. Further, while use cases for 10GbE outside the datacenter are somewhat nascent

now, increasing use of end-user video and voice over traditional enterprise networks

should over time drive adoption within campus environments as well. As seen in Exhibit 29,

Exhibit 30, and Exhibit 31, we expect 10GbE adoption to rise to 25% in 2017, up from 6%

now and only 0.3% five years ago.

Exhibit 29: 10GbE Non-Existent in 2007 Exhibit 30: 6% of the Market Now Exhibit 31: Growing to 25% in 2017

Enterprise switch ports by port speed. Note: excludes web-managed and un-managed ports

40GbE+, 0.0%

10GbE, 0.3%

1GbE and below,

99.7%

40GbE+, 0.1%

10GbE, 6.4%

1GbE and below,

93.6%

40GbE+,

0.6%

10GbE,

24.6%

1GbE and

below,

74.8%

Source: Infonetics. Source: Infonetics. Source: Infonetics, Credit Suisse estimates

Layer 3 switches to continue gaining wallet share at the expense of Layer 2. Another

growing trend within datacenters has been an increasing use of Layer 3 (L3) switches,

often at the expense of Layer 2 (L2) switches. The added intelligence of Layer 3 switches

removes a lot of the manual configuration historically required within enterprise networks,

which is especially important within virtualized infrastructures, as L3 switches are able to

abstract away distinct, proliferating, and difficult-to-manage VLANs. Legacy datacenter

networks were built with a three-tier model with L2 switches at the top of the server rack,

the end of the server row, and then at the aggregation layer before connecting to a router.

However, in many cases the added functionality of Layer 3 switches removes the need for

one of the tiers, whether through the use of fabric extenders or otherwise. Looking forward,

we expect these growth trajectories to continue as the modest incremental cost (only 5%

for a 10GbE port) of Layer 3 switches is offset by the ease of management and ability to

utilize less hardware. Outside of datacenter environments, we expect Layer 2 switches to

continue to prevail and port growth to be a function of device growth, although it is

18 September 2013


important to keep in mind that general campus switches tend to be at a lower ASP than

their datacenter equivalents, which pressures revenue trends. Further, Layer 2 switches

are most at risk from wireless cannibalization, as they have the highest attach to end

devices.

Exhibit 32: Layer 3 Spend Continues to Grow Faster Than Layer 2 Revenue

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

Y/Y

reve

nue gro

wth

Total Layer 2 Revenue Total Layer 3 Revenue

Source: Infonetics.

Chassis switches under macro and secular pressure. Chassis switches provide Layer 3

routing primarily for aggregation and distribution within an enterprise network. Ports within

the segment have come under pressure over the past five years, declining at an average

of 5.2% per annum across the period due to both macroeconomic and secular pressures.

Built for reliability and expandability, chassis switches command a significant ASP

premium to fixed switches that also operate on Layer 3. (See Exhibit 33.) Over the past

few years broad IT spend has been under pressure, and chassis switches were certainly

not immune to that trend. This shift has been furthered by the increased functionality of

fixed Layer 3 switches as well as the aforementioned move to collapse "tiers" of switches

within enterprise environments.

Exhibit 33: Chassis Commands a Premium Port ASP to Fixed Layer 3 Switches

$0

$50

$100

$150

$200

$250

$300

2008 2009 2010 2011 2012

Po

rt A

SP

($

/P

ort

)

Chassis ASP Layer 3 Switch ASP

Source: Infonetics.

18 September 2013


Enterprise Routing

We turn now to the enterprise routing market, which is far smaller than the switching

market at $3.5 billion in 2012. In modeling this segment, we look at the market across

high-end (listed unit ASP >$30,000), midrange (between $5,000 and $30,000), branch

(between $490 and $5,000), and low-end / SOHO (<$490). WAN bandwidth consumption

continues to be a driver of the enterprise routing market, as seen in a November 2012

Infonetics survey, in which North American respondents expect their bandwidth needs to

grow 30% annually. We expect an improving macroeconomic environment to provide a

tailwind into 2014. However, we expect more modest revenue growth of 3.2% in the long

term out of the enterprise routing market.

Exhibit 34: We Expected Low-Single-Digit Revenue Growth in Enterprise Routing

0

200

400

600

800

1,000

1,200

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

Quar

terly

reve

nue (

US

D m

n)

High-end Mid-Range Branch Office Low-end / SOHO

Source: Infonetics.

Total Enterprise Switching and Routing

In modeling aggregate enterprise spend for switching and routing, we combine the secular

trends with indications from our survey work on how industry participants view the overall

spending environment. Specifically, we model both switches and routers by ports and

ASPs ($/port) by port speed across the various sub-segments identified in Exhibit 42. We

expect total enterprise spend on switching and routing to grow at a 3.4% CAGR in the long

term, with port growth of 4.8% offset by 1.3% declines in $/port.

With respect to the overall demand environment, in our survey of 40 enterprise CIOs we

posed the question of "Why are you investing in Networking?" and found that nearly 50%

of respondents view increased business demand as the most important factor. (See

Exhibit 35.) Indeed, such a response indicates to us that the overall economic backdrop is

improving, which will likely propel networking spend into 2014.

18 September 2013


Exhibit 35: Credit Suisse CIO Survey: Why Are You Investing in Networking? % of responses identified as "Most Important"

Increased business demand, 47.2%

Cost savings (OpEx), 22.2%

Application performance, 22.2%

Remain competitive, 5.6%

Regulation, 2.8%

Maintenance, 0.0%Drive new revenues,

0.0%

Source: Credit Suisse CIO Survey August 2013.

Factoring in the aforementioned secular trends at play within the enterprise market

including 10GbE adoption and a preference for fixed Layer 3 switches over Layer 2 and

chassis switches, we expect the market to grow at a CAGR of 3.4% in the long term.

Interestingly, while we expect port growth to slow versus the previous five years to 4.8%

per annum from 5.7% previously, the ASP uplift due to the shift toward 10GbE will offset

the decline and drive a modest reacceleration in the near term in revenue growth.

Exhibit 36: Enterprise Switching and Routing Revenue to Grow at a 3.4% CAGR in the Long Term US$ in millions, unless otherwise stated

Enterprise ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Unmanaged / Web managed 726.6 772.8 790.3 856.8 902.8 937.1 963.6 981.4 5.3% 4.4%

Total Layer 2 2,736.5 2,748.4 2,726.0 2,642.5 2,565.4 2,536.4 2,517.0 2,511.5 -12.4% -1.6%

Total Layer 3 7,702.0 8,843.8 9,961.8 10,892.6 11,407.3 11,961.1 12,446.7 12,898.0 20.4% 5.3%

Total Chassis 7,487.4 6,500.0 6,291.7 6,075.8 6,321.8 6,526.5 6,788.0 7,051.6 -2.8% 2.3%

Total Switching $18,652.5 $18,865.0 $19,769.8 $20,467.7 $21,197.4 $21,961.2 $22,715.3 $23,442.5 3.0% 3.5%

y/y % Change 30.4% 1.1% 4.8% 3.5% 3.6% 3.6% 3.4% 3.2%

Total 1GbE & Below 14,243.4 13,057.2 12,677.7 12,003.3 11,168.0 10,318.8 9,484.4 8,694.3 -4.1% -7.3%

Total 10GbE 4,409.1 5,786.1 6,865.4 7,814.7 9,118.4 10,607.4 12,071.4 13,491.7 36.9% 14.5%

Total 40GbE & 100GbE - 21.7 226.7 649.8 910.9 1,035.1 1,159.4 1,256.5 NA 40.8%

High-end 482.8 580.9 566.0 604.6 618.9 636.4 655.8 675.7 -2.2% 3.6%

Mid-Range 1,290.0 1,232.6 1,122.4 1,182.3 1,232.0 1,266.7 1,290.2 1,314.0 -7.8% 3.2%

Branch Office 1,565.3 1,529.7 1,492.9 1,537.5 1,538.3 1,549.9 1,619.6 1,692.5 -4.4% 2.5%

Low-end / SOHO 133.6 273.8 327.5 338.0 356.5 373.5 395.3 422.7 19.6% 5.2%

Total Routing $3,471.7 $3,617.0 $3,508.8 $3,662.3 $3,745.6 $3,826.6 $3,960.9 $4,105.0 -4.1% 3.2%

y/y % Change 13.0% 4.2% -3.0% 4.4% 2.3% 2.2% 3.5% 3.6%

Total Enterprise $22,124.2 $22,482.0 $23,278.6 $24,130.1 $24,943.0 $25,787.8 $26,676.2 $27,547.5 1.7% 3.4%

y/y % Change 27.3% 1.6% 3.5% 3.7% 3.4% 3.4% 3.4% 3.3%

Source: Infonetics, Credit Suisse estimates.

18 September 2013


Service Provider Switching & Routing: +5.5% CAGR

Service Provider Routing

In examining trends within the service provider (SP) market, we begin by focusing on

routers which totaled $11.3 billion in 2012 out of $14.2 billion in total service provider

spend. In revenue terms, the market has shown significant growth over the past five years,

expanding at a pace of 7.9% per annum, which is far faster than growth on the enterprise

side. Digging one layer deeper, there are significant divergences between the growth

within core (0.8% 2007-2012 CAGR) and edge (10.7% 2007-2012 CAGR) routers, as

service providers continue to focus on adding capacity and intelligence to the edge of their

networks at the expense of the core.

Exhibit 37: The Hierarchy of Service Provider Network Design

Edge - the boundary between the service-

provider’s premises and the customer’s

location. The concentration point where large

numbers of customer connections will be

terminated.

Aggregation - a concentration point where

data from multiple Edge locations will be

funneled. Functionally another tier of the “Edge.”

Core - the heart of the network. The major

switching locations that form the center of the

network, where data from multiple

Aggregation sites will be funneled

This is typically where one sees the

highest volume of data present in the

network.

Source: Juniper Networks, Credit Suisse research.

Core routers losing wallet share within spend. Core routers are robust, high-speed routers

built for reliability that sit in the center of a service provider's network. These routers are

critical to a network's operations and carriers pay quite a premium for that reliability as a

result. Indeed, in 2012 the average unit price of a core router was $129,000, although only

slightly more than 20,000 were sold. In aggregate, the core routing market totaled $2.7bn

in 2012 and has come under pressure, as it grew only 0.8% per annum over the past five

years and declined 10.3% y/y in 2012. Deployments tend to be 'lumpy' with large

fluctuations depending on individual service provider rollouts. (See Exhibit 38.)

Spending within the core of the network tends to be a function of capacity, and recently

service providers have been trying to prolong spend within the area as long as possible.

Indeed, recent commentary from Juniper indicates that core networks continue to be run

'hotter' i.e., closer to their capacity with an expectation of a coming upgrade cycle. On the

2Q13 conference call, Juniper CEO Kevin Johnson noted that "[Juniper] has seen the core

of the network in our analysis has been running hotter and hotter, and we're just now

starting to see the signs that investment will, in addition to being on the edge, will start to

kick in for the core and we expect to see that start to unfold later this year." We expect a

18 September 2013


cyclical uptick over the coming quarters; however, we do not see a structural shift back to

the core on a long-term basis. We model 3.5% growth through 2017 with a significant

deceleration beyond 2014.

Exhibit 38: Core Router Revenue Can be Lumpy from Quarter to Quarter y/y % change in core router revenue

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

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

Source: Company data, Infonetics

Increased focus on investment at the edge likely to continue. While core investment has

been stagnant in the aggregate over the past five years, edge routing has seen substantial

growth over the same period, rising 10.7% on average per year in revenue terms. In 2012,

the market totaled $8.6bn amid secular tailwinds of added capacity and services within

service provider networks. One such example of expanded services is the "triple play"

offerings (i.e., cable, phone, and internet) of many traditional voice or cable service

providers, as they look to drive additional ARPU from their installed base and as a result

are forced to add capacity to the edge of their networks. Additionally, overall bandwidth

consumption per subscriber driven by, for example, internet video coupled with broad

subscriber growth, has resulted in increased demand for edge routing capabilities.

A second secular tailwind to edge routing spend has been the tremendous growth in

smartphones and the resulting growth in mobile data consumption. Indeed, within North

America, smartphone penetration (defined as smartphones as a percentage of mobile

shipments) rose from 4.0% in 2006 to 49.5% in 2012, driving massive amounts of mobile

data. Globally, penetration stood at only 4.2% in 2006. Further, the move toward LTE from

2G/3G technologies within mobile networks is causing a renewed focus on a converged IP

network driving an upgrade cycle across service provider networks. Historically, 2G/3G

technologies parsed between voice and data transmissions, sending the former over

legacy circuit-switched networks while routing the data packets across IP networks.

However, LTE enables both voice and data to travel across IP networks, leading to

converged IP solutions across service providers' networks. As seen in Exhibit 39, we

expect tremendous growth in LTE smartphone units through 2017 driving a continued

focus on edge routers.

18 September 2013


Exhibit 39: LTE to See Tremendous Growth over the Next Five Years on Our Estimates Number of LTE devices sold in millions

7

86

219

298

460

604

789

0%

5%

10%

15%

20%

25%

30%

35%

40%

0

100

200

300

400

500

600

700

800

900

2011 2012E 2013E 2014E 2015E 2016E 2017E

Asia Pacific North America Western Europe Africa

CEE Latin America Middle East % of the handset market


Combining these factors, we continue to expect edge routers to outperform core within the

service provider space. We forecast 6.3% long-term revenue growth within the edge

routing market.

Service Provider Switching

In contrast to the enterprise market, the switching market is dwarfed by the routing market

within service provider spend. In 2012 the market totaled $2.9bn and has grown at an

average annual rate of 6.0% over the past five years, although last year spend was down

11.0%. As switches are primarily deployed at the edge of networks, we expect the same

secular drivers of edge routing discussed above to propel carrier switches, as we model a

5.1% growth rate in the long term.

Exhibit 40: Total Service Provider Switching and Routing to Grow 5.5% Per Year Long Term US$ in millions, unless otherwise stated

Service Provider 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Core 2,838.8 2,975.0 2,668.8 2,840.1 3,058.4 3,089.3 3,126.3 3,163.8 0.8% 3.5%

Edge 7,683.1 8,544.0 8,639.6 9,118.8 9,957.7 10,575.1 11,183.2 11,725.6 10.7% 6.3%

Total Routing $10,521.9 $11,519.0 $11,308.4 $11,958.9 $13,016.1 $13,664.4 $14,309.5 $14,889.4 7.9% 5.7%

y/y % Change 17.2% 9.5% -1.8% 5.8% 8.8% 5.0% 4.7% 4.1%

Total Switching $3,176.1 $3,257.3 $2,900.8 $2,897.3 $3,110.8 $3,323.0 $3,517.0 $3,722.4 6.0% 5.1%

y/y % Change 57.7% 2.6% -10.9% -0.1% 7.4% 6.8% 5.8% 5.8%

Total Service Provider $13,698.0 $14,776.3 $14,209.3 $14,856.2 $16,126.9 $16,987.3 $17,826.5 $18,611.8 7.5% 5.5%

y/y % Change 24.6% 7.9% -3.8% 4.6% 8.6% 5.3% 4.9% 4.4%


1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

24

Exhibit 41: Aggregate Networking: 4.2% LT Growth Driven by Edge Routers and 10GbE Adoption in Enterprise Switching

Enterprise Switching Revenue ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Unmanaged / Web managed $726.6 $772.8 $790.3 $856.8 $902.8 $937.1 $963.6 $981.4 5.3% 4.4%

Total Layer 2 $2,736.5 $2,748.4 $2,726.0 $2,642.5 $2,565.4 $2,536.4 $2,517.0 $2,511.5 -12.4% -1.6%

Total Layer 3 $7,702.0 $8,843.8 $9,961.8 $10,892.6 $11,407.3 $11,961.1 $12,446.7 $12,898.0 20.4% 5.3%

Total Chassis $7,487.4 $6,500.0 $6,291.7 $6,075.8 $6,321.8 $6,526.5 $6,788.0 $7,051.6 -2.8% 2.3%

Total 1GbE & Below 14,243.4 13,057.2 12,677.7 12,003.3 11,168.0 10,318.8 9,484.4 8,694.3 -4.1% -7.3%

Total 10GbE 4,409.1 5,786.1 6,865.4 7,814.7 9,118.4 10,607.4 12,071.4 13,491.7 36.9% 14.5%

Total 40GbE & 100GbE - 21.7 226.7 649.8 910.9 1,035.1 1,159.4 1,256.5 NA 40.8%

Total Enterprise Switching $18,652.5 $18,865.0 $19,769.8 $20,467.7 $21,197.4 $21,961.2 $22,715.3 $23,442.5 3.0% 3.5%

y/y % Change 30.4% 1.1% 4.8% 3.5% 3.6% 3.6% 3.4% 3.2%

Carrier Ethernet Switching Revenue ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Total Carrier Ethernet Switching $3,176.1 $3,257.3 $2,900.8 $2,897.3 $3,110.8 $3,323.0 $3,517.0 $3,722.4 6.0% 5.1%

y/y % Change 57.7% 2.6% -10.9% -0.1% 7.4% 6.8% 5.8% 5.8%

Enterprise Routing Revenue ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

High-end 482.8 580.9 566.0 604.6 618.9 636.4 655.8 675.7 -2.2% 3.6%

Mid-Range 1,290.0 1,232.6 1,122.4 1,182.3 1,232.0 1,266.7 1,290.2 1,314.0 -7.8% 3.2%

Branch Office 1,565.3 1,529.7 1,492.9 1,537.5 1,538.3 1,549.9 1,619.6 1,692.5 -4.4% 2.5%

Low-end / SOHO 133.6 273.8 327.5 338.0 356.5 373.5 395.3 422.7 19.6% 5.2%

Total Enterprise Routing $3,471.7 $3,617.0 $3,508.8 $3,662.3 $3,745.6 $3,826.6 $3,960.9 $4,105.0 -4.1% 3.2%

y/y % Change 13.0% 4.2% -3.0% 4.4% 2.3% 2.2% 3.5% 3.6%

Service Provider Routing Revenue ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Core 2,838.8 2,975.0 2,668.8 2,840.1 3,058.4 3,089.3 3,126.3 3,163.8 0.8% 3.5%

Edge 7,683.1 8,544.0 8,639.6 9,118.8 9,957.7 10,575.1 11,183.2 11,725.6 10.7% 6.3%

Total Service Provider Routing $10,521.9 $11,519.0 $11,308.4 $11,958.9 $13,016.1 $13,664.4 $14,309.5 $14,889.4 7.9% 5.7%

y/y % Change 17.2% 9.5% -1.8% 5.8% 8.8% 5.0% 4.7% 4.1%

Total Enterprise vs. Service Provider ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Total Enterprise 22,124.2 22,482.0 23,278.6 24,130.1 24,943.0 25,787.8 26,676.2 27,547.5 1.7% 3.4%

Total Service Provider 13,698.0 14,776.3 14,209.3 14,856.2 16,126.9 16,987.3 17,826.5 18,611.8 7.5% 5.5%

Total Revenue $35,822.2 $37,258.2 $37,487.9 $38,986.3 $41,069.9 $42,775.1 $44,502.7 $46,159.3 3.7% 4.2%

y/y % Change 26.3% 4.0% 0.6% 4.0% 5.3% 4.2% 4.0% 3.7%

1) Expect 4.2% LT revenue growth in aggregate Ethernet Switching & Routing. Looking at combined Ethernet Switching and Routing, we expect the market to grow at

4.2% per annum over the next 5 years. Looking one layer underneath those forecasts, we expect modestly faster growth out of service providers offset by modestly slower

growth out of enterprises in a challenging environment for IT budgets.

2) Enterprise switching driven by 10GbE adoption. Our forecast calls for 3.5% growth in enterprise switches, driven by 10GbE adoption which we expect to rise to 25% of

Layer 2, Layer 3, and Chassis port shipments in 2017, up from 6.4% in 2012. Overall we expect 10GbE revenue to grow at a 14.5% annualized rate long term.

3) Growth at the edge to propel SP routing. Looking at the Service Provider market, edge routers exhibited strong growth over the past 5 years while core routers have

remained relatively constant. While the growth rates out of the edge are likely to slow going forward, we continue to see a more intelligent edge as a secular tailwind to growth

within the SP routing space.


1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

25

Exhibit 42: Aggregate Networking: 10GbE Adoption to Offset Broad Pricing Pressure Modestly as Ports Expected to Grow 5.0% Per Annum

1) Ports expected to grow at 5.0% LT. We expect total ports for routing and switching to grow at a 5.0% CAGR LT driven by 13.4% growth in routing.

Within routing we forecast service provider ports to grow at 17.7% through 2017, growing to 61% of total routing ports, up from 50% in 2012.

2) Offset by $3 ASP decline at 0.8% CAGR. We forecast aggregate ASPs for ports to decline $3, or 0.8% per annum, through 2017, driven by a 10.3%

decline in service provider routing.

3) …Resulting in 4.2% aggregate revenue growth. We forecast total industry revenue to grow at 4.2% LT driven mainly by 5.7% growth in SP routing.

Revenue ($mn) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Total Enterprise Switching 18,652.5 18,865.0 19,769.8 20,467.7 21,197.4 21,961.2 22,715.3 23,442.5 3.0% 3.5%

Total Carrier Ethernet Switching 3,176.1 3,257.3 2,900.8 2,897.3 3,110.8 3,323.0 3,517.0 3,722.4 6.0% 5.1%

Total Switching $21,828.6 $22,122.2 $22,670.7 $23,365.1 $24,308.2 $25,284.2 $26,232.3 $27,164.9 3.3% 3.7%

Total Enterprise Routing 3,471.7 3,617.0 3,508.8 3,662.3 3,745.6 3,826.6 3,960.9 4,105.0 -4.1% 3.2%

Total Service Provider Routing 10,521.9 11,519.0 11,308.4 11,958.9 13,016.1 13,664.4 14,309.5 14,889.4 7.9% 5.7%

Total Routing $13,993.6 $15,136.0 $14,817.2 $15,621.2 $16,761.7 $17,490.9 $18,270.4 $18,994.3 4.2% 5.1%

Total Enterprise 22,124.2 22,482.0 23,278.6 24,130.1 24,943.0 25,787.8 26,676.2 27,547.5 1.7% 3.4%Total Service Provider 13,698.0 14,776.3 14,209.3 14,856.2 16,126.9 16,987.3 17,826.5 18,611.8 7.5% 5.5%

Total Revenue $35,822.2 $37,258.2 $37,487.9 $38,986.3 $41,069.9 $42,775.1 $44,502.7 $46,159.3 3.7% 4.2%

y/y % Change 26.3% 4.0% 0.6% 4.0% 5.3% 4.2% 4.0% 3.7%

Ports ('000) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Total Enterprise Switching 369,160.7 417,763.2 425,520.0 455,341.7 477,832.0 499,127.5 518,667.4 536,804.2 5.9% 4.8%

Total Service Provider Switching 5,892.9 6,302.9 6,371.0 6,693.0 7,333.7 7,993.7 8,633.2 9,323.9 9.7% 7.9%

Total Switching 375,053.7 424,066.1 431,891.1 462,034.7 485,165.7 507,121.2 527,300.6 546,128.1 5.9% 4.8%

Total Enterprise Routing 4,764.6 4,667.5 5,147.3 5,613.1 6,025.4 6,470.8 7,041.7 7,665.5 -4.1% 8.3%

Total Service Provider Routing 4,051.4 4,973.4 5,212.6 6,167.0 7,413.0 8,700.1 10,166.3 11,790.1 31.0% 17.7%

Total Routing Ports 8,816.0 9,640.9 10,359.9 11,780.1 13,438.4 15,170.8 17,208.0 19,455.6 6.1% 13.4%

Total Enterprise 373,925.3 422,430.7 430,667.3 460,954.8 483,857.4 505,598.3 525,709.0 544,469.7 5.7% 4.8%

Total Service Provider 9,944.3 11,276.3 11,583.7 12,860.0 14,746.7 16,693.8 18,799.5 21,114.0 16.7% 12.8%Total Ports 383,869.6 433,707.0 442,251.0 473,814.8 498,604.1 522,292.1 544,508.5 565,583.6 5.9% 5.0%

y/y % Change 25.0% 13.0% 2.0% 7.1% 5.2% 4.8% 4.3% 3.9%

ASP ($ / Port) 2010 2011 2012 2013E 2014E 2015E 2016E 2017E '07 - '12 '12 - '17E

Total Enterprise Switching 50.5 45.2 46.5 45.0 44.4 44.0 43.8 43.7 -2.7% -1.2%

Total Service Provider Switching 539.0 516.8 455.3 432.9 424.2 415.7 407.4 399.2 -3.4% -2.6%

Total Switching 58.2 52.2 52.5 50.6 50.1 49.9 49.7 49.7 -2.5% -1.1%

Total Enterprise Routing 728.6 774.9 681.7 652.5 621.6 591.4 562.5 535.5 0.0% -4.7%

Total Service Provider Routing 2,597.1 2,316.1 2,169.4 1,939.2 1,755.8 1,570.6 1,407.5 1,262.9 -17.7% -10.3%

Total Routing 1,587.3 1,570.0 1,430.2 1,326.1 1,247.3 1,152.9 1,061.7 976.3 -1.8% -7.4%

Total Enterprise 59.2 53.2 54.1 52.3 51.6 51.0 50.7 50.6 -3.8% -1.3%

Total Service Provider 1,377.5 1,310.4 1,226.7 1,155.2 1,093.6 1,017.6 948.2 881.5 -7.9% -6.4%

Total $ / Port 93.3 85.9 84.8 82.3 82.4 81.9 81.7 81.6 -2.1% -0.8%

y/y % Change 1.0% -7.9% -1.3% -2.9% 0.1% -0.6% -0.2% -0.1%


18 September 2013


Scorecard—Cisco wins What makes a successful networking player? This in itself is not a straightforward question

to answer; after all, what has been interesting is that over a prolonged period of time,

Cisco’s market share in revenue terms in the broader switching and routing market has

remained relatively stable, despite several waves of competition. (See Exhibit 69.) This

suggests that barriers to entry have, up until now, remained relatively high compared with

other areas of technology.

An additional problem for the average investor is that while presentations from most

networking companies tend to have the same promise of superior performance and

innovative technology, it is often hard to decipher what really matters. As such, we have

developed a proprietary scorecard to evaluate key networking vendors across eight

metrics that we think are important for success. We evaluate vendors across many areas,

including exposure to the faster growth segments within networking (edge routing, 10Gb

Ethernet), and including the feedback from networking administrators that reflects

customers' views. We arrive at several main conclusions.

Status quo holds for now but SDN will ensure the industry is not static. Given our

conclusions, we believe that this will shape the ability of companies such as Cisco to build

upon their already dominant position and impact fundamentals over 18+ month period.

The issue and complicating factor for networking is the disruption is not necessarily from

established vendors such as Cisco, Juniper, or Huawei, but rather on how the networking

industry could be disrupted by the uptake of SDN.

Cisco seems robust in today’s world. Cisco has continued to out execute its rivals

consistently in the networking space over time. Our scorecard shows that customers have

long favored Cisco, with the company ranked first on our overall scorecard. Cisco scored

the number one positions on six metrics such as roadmap, salesforce, service offering,

and IT portfolio. These rankings, along with its recent product refreshes, should help Cisco

maintain its share in the near term. This level of dominance, though, is arguably already

reflected in its market share. Equally interestingly, on the most important metric of

price/performance/reliability, Cisco is essentially on par with Juniper and Huawei. This

suggests Cisco’s incumbency rather than technology is the key differentiator for the

company. The real concern as we have discussed in much of this note is that SDN

threatens the very structure of the industry in which Cisco operates.

Exhibit 43: Cisco, HP, and Juniper Expected to Be the Most Disrupted On a scale of 1-5 (1=not vulnerable, 5=very vulnerable) which of the following vendors do you think is the most vulnerable to SDN?

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

NEC Brocade Dell AlcatelLucent

Ericsson Fujitsu ZTE Huawei Juniper HewlettPackard

Cisco

Perc

enta

ge o

f R

esp

on

den

ts

Source: Credit Suisse IT Survey, August 2013.

18 September 2013


Juniper fading gradually? With a 14% share in routing, Juniper has been a clear and

established number two player. Juniper ranks second overall on the scorecard and ranks

highly on its roadmap. However, it is interesting that the company is perceived to be only

narrowly ahead of some of its peers, especially in the routing market. For example on the

key metric of price and performance it is actually ranked lower than Huawei. It is also

disconcerting that over time its market share in its core segment of routing has been

slipping with Alcatel Lucent and Huawei simultaneously gaining in edge routing.

Huawei is a worth disruptor. Interestingly, Huawei scores well, coming in third, despite

having a very small share currently in switching and an emerging position in routing. While

the company comes first in the category of price, technology, and performance levels,

which is itself not surprising, it scores poorly on installed base. Over time we believe its

market share will build further in networking even if we see limited traction in the United

States. We believe that on a global basis, it can gain in almost all segments.

Alcatel Lucent—Moving from the edge to the core. Despite Alcatel Lucent’s troubles as a

company in recent years, one area where execution has been robust has been in the

routing market. On our scorecard the company ranks seven, partly due to the company’s

historic strength in service provider routing, where we do believe the company has the

potential to continue gaining in the routing market. More specifically, we believe early

traction for its new core router is encouraging. Adoption of its SDN strategy from Nuage

Networks should be helped by Alcatel's large service provider install base. Additionally, it

could continue to take share in edge routing.

HP shows lots of promise, but questioning potential. HP ranks fifth on our scorecard,

despite having the number two position in switching. The company scores well on overall

IT portfolio and services offering, but trails on roadmap and management performance.

With an 11% share, we do believe that HP has the potential to gain share in this market,

given its strategic status as a core enterprise player. However, we are concerned by the

stagnation in market share over the past two years in traditional enterprise networking.

That said, we believe its early development of an SDN strategy creates optionality.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

28

Exhibit 44: Credit Suisse Networking Vendor Scorecard—Cisco Is Well Ahead of Peers across Metrics and on Switching and Routing Market Share

Note: All scores are based on a scale of 1 to 10, with 10 being the best, and 1 being the worst score Source: Credit Suisse Survey

Vendor scores and commentary:

1) Cisco (8.0/10 – Rank #1). Cisco has the unquestionably highest rank in 6 of the 7 key metrics on our scorecard (Roadmap, Salesforce, Services offering, IT Portfolio,

Installed base and SDN strategy) giving it an overall score of 8.0/10 on the scorecard in addition to #1 market share in both switching and routing. Interestingly Cisco

ranks behind Huawei and ties with Juniper on the key metric of Price/Technology Performance/Reliability.

2) Juniper (7.1/10 – Rank #2). Juniper ranks second in several key metrics on our scorecard (Roadmap, Price/Technology Performance/Reliability, Services offering and

Installed base) giving it an overall score of 7.1/10 on the scorecard in addition to #2 market share in both switching and routing.

3) Huawei (6.6/10 – Rank #3). Huawei ranks third overall on our scorecard largely because of its #1 rank on Price/Technology Performance/Reliability. Not surprisingly

Huawei has a mediocre score with regard to its Services offering and IT Portfolio giving it an overall score of 6.6/10 on the scorecard. While Huawei has low share in the

Enterprise switching and routing businesses, we believe it could be a disruptive force in international markets.

4) HP (6.4/10 – Rank #5). HP ranks fifth overall on our scorecard and has strong scores on the IT Portfolio, Salesforce/Distribution and Services Offering metrics. It is also

worth noting that HP has a 10.4% share in the switching market, second only to Cisco giving it an overall weighted score of 6.4/10.

5) Alcatel (5.7/10 – Rank #7). Alcatel Lucent rank seventh (5.7/10) on our scorecard which isn’t surprising given the company does not have an enterprise focus within

their networking business. However having introduced new Edge and Core routing products recently as well as having a defined SDN strategy through Nuage, we believe

this will result in share gains over time.

Cisco Juniper Huawei Brocade HP Dell Alcatel Fujitsu NEC ZTE Ericsson Weight

Weighted Rank 1 2 3 4 5 6 7 8 9 10 11

Weighted Score 8.0 7.1 6.6 6.6 6.4 6.4 5.7 5.5 5.3 5.1 5.1

Switching Market Share 61.4% 2.2% 1.1% 1.4% 10.4% 1.2% 1.5% 0.5% 0.4% 3.4% 0.0%

Switching Market Share Rank 1 4 8 6 2 7 5 9 10 3 11

Enterprise Switching Market Share 64.7% 2.6% 0.0% 1.5% 11.9% 1.4% 1.2% 0.0% 0.0% 0.0% 0.0%

Service Provider Switching Market Share 38.9% 0.0% 8.9% 0.7% 0.0% 0.0% 3.3% 4.1% 2.9% 26.3% 0.3%

Routing Market Share 45.2% 14.0% 13.8% 1.2% 1.5% 0.0% 13.5% 0.8% 1.2% 1.4% 2.1%

Routing Market Share Rank 1 2 3 8 6 11 4 10 8 7 5

Enterprise Routing Market Share 71.5% 5.4% 1.4% 0.0% 6.1% 0.0% 0.7% 0.0% 1.1% 0.0% 0.0%

Service Provider Routing Market Share 37.0% 16.7% 17.7% 1.6% 0.0% 0.0% 17.5% 1.1% 1.2% 1.8% 2.7%

Price/Technology Performance/Reliability 7.3 7.3 7.8 6.7 6.8 6.6 6.0 6.2 5.8 6.0 6.0 45%

Roadmap (historical/futures) 9.3 7.8 6.2 7.2 5.9 6.4 5.4 5.1 5.0 4.0 5.0 25%

Installed base 9.2 6.9 5.3 6.6 6.2 6.3 4.9 4.2 5.0 4.9 5.3 10%

SDN Strategy 6.0 5.0 5.0 5.0 5.0 4.0 5.5 5.3 4.0 4.0 0.0 8%

IT Porfolio (storage/servers etc. integration) 7.8 6.3 5.3 5.5 6.9 6.8 5.5 4.7 5.6 5.3 5.3 5%

Services offering (Maintenance/Support) 8.5 7.0 5.6 6.9 7.0 6.4 5.8 4.4 5.0 4.9 5.0 5%

Salesforce/Distribution 7.6 6.2 5.3 6.1 6.4 6.6 5.2 4.4 4.6 4.7 4.8 3%

Source: Credit Suisse IT and Networking Surveys, August 2013.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

29

Exhibit 45: Industry Share Breakdown—Cisco Dominant Across Both Switching and Routing Although Competition Emerging from Huawei

Switching and Routing Industry Share Breakdown 1) Cisco – dominant across networking. Cisco is well placed

on the enterprise switching side with 65% share overall and

67% share in the 10GbE segment; however the company

has 49% share in the 40GbE & 100GbE segment. Although

this remains small we would expect Cisco's footprint in the

space to grow. On the SP routing side Cisco’ share in the

faster growing Edge market has declined to 31% from

~50% in 2007, although the business has rebounded since

the lows of 2011.

2) Juniper – focused on SP routing. Juniper has sustained

~15% share on the Edge routing side since 2007. We

expect to see continuing competition from Huawei, Cisco

and Alcatel Lucent, and other potential headwinds for share

in the LT. In Core routing Juniper's share has declined from

~30% in 2007 to ~24% in 2012. However this business is

only expected to grow 3.5% through 2017.

3) Huawei – could be a disruptive force. Huawei has rapidly

increased share in its Service provider routing business over

the past few years growing from 5% in 2007 to 17% by

2012. The scale, size and R&D dynamics for Huawei show

that the company cannot be underestimated. The company

has a share of 20% in the Edge market as compared to 6%

in 2007, and this segment is expected to grow at a 6.3% clip

through 2017.

4) ALU – #2 player in Edge. Alcatel Lucent's share in Edge

routing has steadily increased from ~18% in 2008 to ~22%

in 2012. Alcatel has also recently launched a family of XRS

Core routers which could be accretive to its 1% share in

Core service provider routing.

5) HP – shares plateauing. Following the acquisition of 3COM

HP’s share grew from mid-single digit share to ~11% which

the company has largely been able to maintain. In the

absence of any major product refreshes or acquisitions we

believe HP's Enterprise switching share is likely to stay static

going forward.

2012 Market Share Cisco Juniper HP Alcatel Huawei Market (mlns) CAGR '12-'17

Switching

Unmanaged / Web managed 11.0% 0.0% 18.7% 0.0% 0.0% 790 4.4%

Total Layer 2 58.2% 0.0% 22.8% 0.4% 0.0% 2,726 -1.6%

Total Layer 3 64.7% 4.2% 9.3% 1.6% 0.0% 9,962 5.3%

Total Chassis 74.4% 1.4% 10.5% 1.0% 0.0% 6,292 2.3%

Total 1GbE & Below 64.1% 2.7% 12.1% 1.4% 0.0% 12,678 -7.3%

Total 10GbE 66.5% 2.4% 11.8% 0.8% 0.0% 6,865 14.5%

Total 40GbE & 100GbE 49.1% 4.0% 3.5% 1.1% 0.0% 227 40.8%

Enterprise Switching 64.7% 2.6% 11.9% 1.2% 0.0% 19,770 3.5%

SP Ethernet Switching 38.9% 0.0% 0.0% 3.3% 8.8% 2,901 5.1%

Total Switching 61.4% 2.2% 10.4% 1.5% 1.1% 22,671 3.7%

Routing

High-end 47.0% 26.2% 15.9% 4.1% 0.0% 566 3.6%

Mid-Range 82.4% 2.3% 5.1% 0.0% 2.4% 1,122 3.2%

Branch Office 78.6% 0.9% 3.0% 0.0% 1.0% 1,493 2.5%

Low-end / SOHO 44.0% 0.0% 7.1% 0.0% 2.0% 328 5.2%

Total Enterprise Routing 71.5% 5.4% 6.1% 0.7% 1.4% 3,509 3.2%

Core 57.0% 24.3% 0.0% 0.6% 10.4% 2,669 3.5%

Edge 30.8% 14.3% 0.0% 22.7% 20.0% 8,640 6.3%

SP Routing 37.0% 16.7% 0.0% 17.5% 17.7% 11,308 5.7%

Total Routing 45.2% 14.0% 1.5% 13.5% 13.8%

Total Enterprise 65.8% 3.0% 11.0% 1.1% 0.2% 23,279 3.4%

Total Service Provider 37.4% 13.3% 0.0% 14.6% 15.9% 14,209 5.5%

Total Networking Spend 55.0% 6.9% 6.8% 6.2% 6.2% 37,488 4.2%


18 September 2013


What Makes a Networking Vendor Successful?

We score each of the key vendors across these seven metrics and use the aggregate

scores for each vendor to develop a ranking of vendor positioning. As shown in Exhibit 46,

these categories include key characteristics such as technological performance, product

roadmap, services offering, datacenter portfolio, and replacement costs. In the networking

world, price, performance, and reliability are valued above all.

Exhibit 46: Networking Vendors Price/ Performance/ Reliability High on the Lost Question: What makes a competitive networking vendor? Please rank in order

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Per

cen

tage

of

resp

on

den

ts

Source: Credit Suisse Networking Survey, August 2013.

Most importantly, our scores are based upon the Credit Suisse Networking survey of 40

major buyers. Thus, we are able to see which vendors may be vulnerable to share loss

and which may be able to capture value. We arrive at seven main conclusions.

Cisco to remain dominant in the traditional networking world. Cisco has continued to out

execute its rivals consistently in the networking world over time, commanding 61% share

in switching and 45% share in the routing market at the end of 2012. While we have

concerns about how the industry will ultimately evolve given the onset of SDN, we do

believe that in the traditional market, the company will retain a very strong position and

possibly enhance share over time.

Cisco ranks very strongly across the Credit Suisse scorecard. From a buyer's perspective,

Cisco's lead remains strong, as the company ranked first in six of seven metrics. This level

of dominance, though, is arguably already reflected in its market share. Equally

interestingly, on the most important metrics of price and performance and reliability, Cisco

is essentially on par with Juniper and Huawei. This suggests Cisco’s incumbency rather

than technology is the key differentiator for the company.

18 September 2013


Exhibit 47: Cisco Ranks #1 in Six out of Seven Key Metrics on Our Scorecard

#1 #1

#1

#1#1

#3

#1

0

2

4

6

8

10

Roadmap Installed base Services offering(Maintenance/Support)

IT Porfolio(storage/servers etc.

integration)

Salesforce/Distribution Price/TechnologyPerformance/Reliability

SDN Strategy

Source: Credit Suisse IT and Networking Survey, August 2013.

Portfolio refresh helps near term. We believe a key driver around the recent market share

turnaround for Cisco, especially in the switching market, has been a fairly comprehensive

portfolio refresh. Indeed, as shown in Exhibit 49, going back over the past 18 months,

there has been a refresh across its modular and fixed switching portfolio that we believe

has been key to extending its competitive advantage.

Exhibit 48: Cisco Dominates Across All Areas of Switching and Routing

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Cis

co M

ark

et S

hare

Enterprise switching Enterprise routing Service provider switching

Service provider routing Service provider edge routing Service provider core routing

Source: Infonetics.

Datacenter portfolio. For its datacenter portfolio, the company now has a range of switches

aimed at optimizing performance for the datacenter, centered around the Nexus portfolio

of switches that span from virtual switching to fixed and modular formats. Cisco Nexus

7000 Series Switches offer Cisco's highest switching capacity; up to 1.3 terabits per slot,

83+ terabits per chassis, and provides 1,10, 40, and 100 Gigabit Ethernet scalability.

Backbone and access portfolio. Aimed more at the Enterprise and the campus market, the

company has introduced the Cisco Catalyst 6800 Series Switches that are programmable

campus backbone switches optimized for 10/40/100 Gigabit Ethernet services. These

switches offer converged wired, wireless, and VPN security, as well as exceptional

investment protection with their Catalyst 6500. Whereas the catalyst 2960 is aimed at

branch offices or small- and medium-sized businesses, we would note such entry-level

enterprise-class Fast Ethernet switches support basic services.

http://www.cisco.com/en/US/products/ps9402/index.html


http://www.cisco.com/en/US/products/hw/switches/ps708/index.html

18 September 2013


Exhibit 49: Cisco Has Refreshed the Switching Portfolio Meaningfully of Late

2012 2013

UCS Central

September

Cisco Catalyst

3850

January

MDS 9710 Multilayer Director,

UCS, and NVIDIA GPU

May

Cisco Catalyst

6800

June

Nexus 7700

June

Nexus 3548

November

Nexus 6001, 6004

and 1000V

February

Cisco Catalyst

2960-X

June

Supervisor 8E for

Cisco Catalyst 4500-E

June

Source: Company data.

Forward looking at the Insieme spin in for SDN. Cisco’s initial response to the

opportunity/risk presented by SDN appears to be in the form of Insieme Networks, a

captive private company started by three former key Cisco enterprise switching executives

that Cisco has the right to acquire or “spin-in” for $750 million. While initial product strategy

and details remain forthcoming, we believe Cisco has essentially outsourced to Insieme

the development of its SDN response. We believe the reason for this is to avoid the

standard innovators dilemma, as the focus of Insieme will be to become Cisco’s next next-

generation switching platform, with a focus on 100Gbps Ethernet datacenter switching that

goes beyond traditional enterprise switching to add SDN and storage to Cisco’s Unified

Computing System converged datacenter architecture. We believe the strategy here

shows management's flexibility and realistic assessment that, in order to create an

industry leading SDN platform, it needs to be kept separate from the rest of the company.

While there is always a risk that this remains an expensive decision, we believe this

approach was successful with Nuova Systems, which brought to Cisco the Nexus product

family. We believe one potential negative on Insieme could be that it will continue to use

custom Cisco ASICs, and as a result, a network administrator will still need to purchase a

bundled solution from Cisco.

Dominant in the traditional world. Given the low growth prospects of about 4% in the

longer term within the Networking sector, we believe operating performance of individual

vendors will depend on having the right exposure across key segments within switching

and routing. Cisco's share in the Enterprise switching market peaked at ~70% in 2010,

and has fallen to about 65% in recent years. Overall, we believe Cisco is well positioned

on the enterprise switching side with 65% share overall and 67% share in the faster

growing 10GbE segments. However, the company only has 60% share in the 40GbE and

100GbE segment. Although this remains a nascent market, we would expect Cisco's

footprint in the space to grow over time.

On the service provider routing front, Cisco has lost share from over 50% in 2007-08 to

~37% today. While the company has largely maintained share in the Core business with

~50% share, this segment is only expected to grow ~3.5% through 2017. Additionally, we

highlight that Cisco has a smaller share within the faster growing edge routing market with

about 30% share, which has fallen from ~50% share in 2007, although the business has

rebounded slightly since the lows of 2011. Cisco's edge routing share could fall further

given that the edge routing business has comparatively lower barriers to entry compared

to the core, and as a result, will likely see increased competition from Huawei, Juniper,

and Alcatel Lucent.

18 September 2013


Wireless LAN—Secular Growth As a result of technological improvement, the WLAN market has been one of the fastest

growth markets within the networking sector. Based on Infonetics data, we think the

current WLAN market is $3.9 billion and is expected to grow at a five year CAGR of 12%,

reaching $6.7 billion by 2017. On a regional basis, we think emerging markets in Asia

Pacific and CALA will outgrow the overall market. North America will still maintain more

than half of the total market, as we expect to see the adoption of 802.11ac starting first in

developed markets in 2014.

Figure 50: WLAN Is a $4 Billion Market Growing at 12% CAGR 2013-17 in millions, unless otherwise stated

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

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

Total sales yoy growth

13-17 CAGR:11.9%


We believe in aggregate this growth is driven by several factors

BYOD—The main driver of Wireless LAN adoption. Based on Gartner, we estimate that

there could be over 500 million Wi-Fi enabled devices (mostly tablets and phones) brought

onto enterprise networks by 2015, driven by the quick adoption of the BYOD trend (bring

your own device). With an increasing amount of people using smartphones and tablets,

there is strong demand for having these devices connected to the Internet at the

workplace. To address this demand, IT managers have to build out networks that have

capacity to allocate bandwidth efficiently, offer secure corporate IT access, and enforce

corporate compliance policy. The BYOD adoption trend has created a major opportunity

for the WLAN market. To expand wireless coverage, enterprises need to install more

wireless access points for employees and guests. As the deployment of wireless APs

continues, IT administrators need a centralized platform, managed by a WLAN controller,

to manage the whole network.

Figure 51: Emerging Markets Outgrows the Overall Market In US$ millions, unless otherwise stated

Revenue by Region 2011 2012 2013E 2014E 2015E 2016E 2017E 13-17 CAGR

North America 1,550.9 1,983.8 2,181.7 2,431.9 2,788.6 3,116.7 3,411.8 11.8%

EMEA 858.0 1,061.1 1,186.4 1,246.3 1,420.1 1,635.6 1,812.7 11.2%

Asia Pacific 596.1 715.9 763.1 816.5 945.7 1,106.3 1,235.0 12.8%

CALA 110.4 129.0 135.5 149.0 186.8 214.4 238.4 15.2%

Total sales 3,115.3 3,889.8 4,266.7 4,643.7 5,341.1 6,073.0 6,698.0 11.9%


18 September 2013


Figure 52: More Employees at Credit Suisse Are

Connecting Their Personal Phones to Work

Figure 53: What Percentage of Your Mobile Workforce Will

Have Tablets?

0

5,000

10,000

15,000

20,000

25,000

30,000

MAY11

JULY11

SEP11

NOV11

JAN12

MAR12

MAY12

JUL12

SEP12

NOV12

JAN13

MAR13

MAY13

JUL13

SEP13

NOV13

BlackBerry MyMobile

0%

5%

10%

15%

20%

25%

30%

35%

40%

Currently 1 year from now 2 years from now 5 years from nowSurvey CS Forecast

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

Rising data speeds. The quality of the WLAN experience has improved as wireless speeds

rise. In 2009, the 802.11n standard was added to 802.11, which significantly improved

network throughput. It can operate in both 2.4Ghz and 5Ghz bands and support MIMO

(multiple-input, multiple-output). The maximum data transfer speed can reach 600Mbit/s.

Now the industry is moving towards 802.11ac, a standard which will provide throughput in

the 5GHz band. The maximum speed of 802.11ac can achieve is 6,900Mbit/s. (See Figure

55.)

Modernization of infrastructure. The networking functionalities for which enterprises

historically depended on a wired infrastructure can be increasingly performed through

wireless networking. As a direct result, enterprises are seeing the value of being able to

forego Ethernet cables and connect wirelessly, especially with mobile devices proliferating

in the enterprise. Additionally, with its high-bandwidth capabilities, WLAN can increasingly

improve the experience with multimedia applications.

Exhibit 54: Infonetics Expects 6 Million Wifi Public Hotspots by 2015 in millions, unless otherwise stated

0.0

1.0

2.0

3.0

4.0

5.0

6.0

2009 2010 2011 2012 2013 2014 2015

Pu

bli

c h

ots

po

ts (

mil

l.)

Source: Infonetics.

Business/vertical-specific functionalities. The obvious use case has been that many

enterprises find it beneficial to offer WiFi access to their end users and guests and visitors.

However, we are now seeing an increasing majority of enterprise WLAN being deployed

for business-critical applications, such as managing inventory in a warehouse, accessing

medical records while on the move, or accepting payments in a retail store. Enterprises

18 September 2013


are finding that wireless networking allows them to run their businesses more effectively.

For example, customers of enterprises such as hotels and restaurants choose these

businesses based on the availability of wireless Internet for business and pleasure.

A Brief Review of WLAN Industry—Driven by Ever Faster Speeds

A wireless local area network links multiple devices by connecting them to the internet

through access points within the network. Initially, the WLAN hardware deployment cost

was very high and thus was only used as an alternative to cabled LAN in rural places

where it was challenging to deploy cable. By the end of late 1990s, industry standards

were set up mainly based on various versions of IEEE 802.11. In 2009, the 802.11n

standard was added to 802.11, which significantly improved network throughput. It can

operate in both 2.4Ghz and 5Ghz bands and support MIMO (multiple-input,

multiple-output). The maximum data transfer speed can reach 600Mbit/s. Now the industry

is moving towards 802.11ac, a standard which will provide throughput in the 5GHz band.

The maximum speed of 802.11ac can achieve 6,900Mbit/s. (See Figure 55.)

Figure 55: The Evolution of WLAN Standards

Source: Cisco.

Competitive Dynamics—Cisco Dominates

Given the superior growth characteristics of the Enterprise Grade WLAN market, it is not

surprising that it remains highly competitive. We believe that several important attributes

are necessary to succeed in this market.

18 September 2013


Figure 56: Ubiquiti Has 1% Market Share of WLAN Market

Cisco, 51%

Other, 20%

Aruba, 11%

HP, 6%

Ruckus, 5%

Motorola, 5%

Meru, 2%

Ubiquiti, 1%


Global scale, consistency, and support. At the enterprise level, distribution remains

paramount. Indeed WLAN vendors that can serve and are easily accessible to

multinational and global enterprises across geographies will attract clients that want to

deploy consistent IT infrastructures worldwide with consistent customer service to service

them across their infrastructure. In this respect, Cisco’s installed base will inherently be at

an advantage, which will take some time for smaller vendors to overcome.

Wired infrastructure integration capabilities. Not to state the obvious but wireless LAN

cannot exist without a wired infrastructure. However, there are variations in how WLAN

infrastructures interact with the wired network. Good WLAN vendors will use

standards-based integration technology to promote good integration without major

modifications to either the wired or wireless infrastructure. Of course, incumbents in the

wired networking segment have the opportunity to integrate the management platforms as

well. The scoring criterion most heavily influenced by "wired infrastructure integration

capabilities" is "integration capabilities/future integration strategy."

Exhibit 57: Aruba Partners Very Effectively for the WLAN Market

Wired Switches Wireless Aps VPN Managed UnManaged Endpoint

Aruba Aruba Aruba Windows iPhone/ipad NAP,NAC

Cisco Meru Cisco Mac Android A/V,A/S

HP Meraki Juniper Linux Printers Firewall

Enterasys Cisco SonicWall VoIP Phones Discovery

Extreme Enterasys F5 Medical Devices Patch

Juniper Motorola Check Point MFG. Devices Management

Foundry Xirrus and More Security Vulnerability &

(Brocade) and More Equipment Port Scans

and More (cameras, etc.)


Partnerships. WLAN is one part of networking and IT infrastructure, and IT departments

and networking configurations will need to evolve and work with all vendors over time.

These vendors will display knowledge of networking needs in their sales, presales,

engineering, and marketing efforts and be able to provide effective first-level support. In

addition to the networking component, a successful channel partner will have experience

and proficiency with RF infrastructures. Additionally, players need a breadth of ecosystem

and integration, as Aruba has done. (See Exhibit 57). This is especially important for

18 September 2013


vendors targeting specific verticals. Similarly, a smart vendor strategy can enter into

strategic interoperability and sales and distribution partnerships with ubiquitous technology

providers to increase penetration and brand value.

Application development platform with the ability to add network intelligence services.

Owing to vertical- or business-specific needs, clients may be interested in network

applications that a WLAN vendor may not offer. To help solve this issue, the vendor may

be able to add third-party solutions that allow a given vendor to add incremental

functionality without having to develop it internally.

Controller-less or cloud-based Wi-Fi. In place of deploying a Wi-Fi network with a physical

controller and access points on premise, a new architecture has emerged that is

essentially a cloud-based WiFi solution, which leverages the cloud to create an

environment with a virtual, cloud-based controller, as opposed to having a physical box on

premise. Indeed Cisco’s acquisition of Meraki can be seen in this light.

Cisco to Dominate, Aruba a Worthy Disruptor

Cisco dominating WLAN. We believe Cisco will maintain its strong position in the WLAN

market, as enterprises with large budget dollars prefer to go with end-to-end hardware and

services support from a few concentrated vendors. Given this dynamic, we believe Cisco

is poised to benefit. With its key advantage of a large installed base in the enterprise

switch and router markets, Cisco is able to cross sell its WLAN routers. After the

acquisition of Meraki, Cisco is capable of hosting a network controller on the company’s

own servers and lets customers access the centralized management platform in the cloud.

Exhibit 58: Cisco Has Dominated the Market but Aruba Has Grown

41.6%47.4% 46.3% 49.7% 49.1% 50.6%

5.7%6.4% 8.6%

8.9% 10.2% 11.0%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2007 2008 2009 2010 2011 2012

Rev

enu

e S

har

e

Cisco Aruba Alcatel-Lucent D-Link HP Juniper Meru Motorola Other Ruckus

Source: Infonetics.

Aruba a strong number 2. Aruba is the number two player in the WLAN space with 11%

market share. While the company is a pure WLAN player and lacks the ability to bundle

sales with other enterprise products, it has managed to gain share since 2007. Aruba has

strong software solutions integrated with hardware products, especially on security and

network management. We believe this offers product differentiation and the newly

launched ClearPass system will give the company more leverage in catching the BYOD

opportunity. Aruba has been focusing on selling managed services deployments to service

providers rather than competing directly in the service provider Wi-Fi market.

HP a distant third place. HP is the third largest player with market share of 5.6%. The

company entered into the WLAN market through the acquisitions of Colubris and 3Com.

18 September 2013


Similar to Cisco, the company has a large installed base of enterprise products and is

aiming to integrate its wired and wireless product solutions within the networking space.

MSI losing share. Motorola Solutions has market share of 4.7% and has been struggling

and losing share in this market. The company mainly targets the retail vertical and

considers WLAN as part of its overall suite of enterprise products rather than an

independent product. As a result of the verticals, it targets growing much slower and,

having misexecuted the transition from a product-only sale to a managed services,

Motorola Solutions saw its market share drop to 4.7% from 6.4% in 2011.

Ruckus rising share. Ruckus is a newly emerging player in the WLAN market with about

5% market share. The company mainly targets small- and medium-sized businesses and

has about two-thirds of revenue coming from the segment. Ruckus competes directly with

Ubiquiti on this space and is also considered a low-cost vendor.

Ubiquiti trying to disrupt another market. Ubiquiti’s product portfolio mainly targets the

low-end of WLAN market, where customers tend to be small businesses and are more

price sensitive. While Ubiquiti just launched its WLAN product Unifi at the end of 2011, we

think the product is getting very good traction and by the end of 2012 Ubiquiti obtained

about 1% revenue share and 4% units share according to Gartner. Given Ubiquiti’s

disruptive pricing structure, we think the company will continue its momentum and keep

gaining share in the WLAN market. However, at the same time, we also want to point out

that a large part of the WLAN market is for large enterprises and it would be difficult for

Ubiquiti to penetrate this market. The Unifi portfolio is a scalable platform targeting the

Wireless LAN market. The platform consists of hardware that is easy to deploy indoor, and

unlike other enterprise Wi-Fi systems that use hardware switch, Unifi is deployed with a

virtual controller that allows for remote management from any PC within the network.

Similar to its other products, Ubiquiti is very aggressive on pricing, making the product

significantly cheaper than alternative products from industry peers. The main competitors

are Ruckus Wireless and Aruba.

Figure 59: UniFi Portfolio

UniFi AP UniFi AP-LR UniFi AP-PRO UniFi AP-AC

2.4GHz Speed 330 Mbps 300 Mbps 450 Mbps 450 Mbps

5GHz Speed 300 Mbps 1300 Mbps

Range 122m (400ft) 183m (600ft) 122m (400ft) 122m (400ft)

Wi-Fi Standards 802.11 b/g/n 802.11 b/g/n 802.11 a/b/g/n 802.11 a/b/g/n/ac

Price $73 $85 $225 $308 Source: Company data, Credit Suisse research.

18 September 2013


SDN—Reorienting Networking Catalyzed by hyperscale adoption, VMware’s acquisition of Nicira, and the promise of

lower opex/capex and choice, SDN or Software-defined Networking, has become the

latest industry buzzword, with management presentations striving to address this topic

meaningfully and numerous roadmaps rolled out. Indeed, there are host of technology

start-ups striving to address this area. At the heart of the move towards SDN, is the

explosion of networking endpoints and the desire for more efficient management. With

this, the changing nature of datacenter traffic requires more flexible networking and is a

departure from the infrastructures of the past 20 years. Consequently, legacy networking

architectures, in which hardware and software were bundled for a static, siloed solution,

are being displaced by systems that can effectively separate hardware and multiple planes

of management. Our analysis of SDN arrives at three important conclusions.

SDN—There are real drivers for its demand. At its most basic definition, SDN involves the

separation of hardware and software, rather than a traditional bundled networking sale.

The reason for this is not disruption to the industry structure for its own sake, but it is to

actually solve underlying challenges facing datacenter deployments. The ever increasing

demand on datacenters from changing traffic and the need for analyzing data, means than

there is a need to create dynamic infrastructure layers that perform the task needed, as

opposed to static traditionally engineered systems. As a result, by divorcing the software

control layer from the hardware, software defined networks can separate away from

individual hardware devices and manage systems at an overarching level.

Three segments for SDN adoption. The adoption in SDN will vary by segment and

anecdotal evidence suggests that this is occurring. From our perspective, we see three

segments, hyperscale datacenter vendors such as Google, Amazon, and Facebook,

carriers such as AT&T, BT, and Verizon, and traditional enterprises.

■ Traditional enterprise. The key benefit here is the ability to create an ease of

management and move towards a virtualized datacenter. We are seeing several

enterprises, mainly financial institutions such as Goldman Sachs and Credit Suisse,

initially test SDN. In such cases, typically, the deployments are more in greenfield

deployments. Outside this narrow vertical, adoption seems to be stem from the devops

movement and agile application development, where more dynamic systems are

favored. This aside, for the broader enterprise market, adoption will likely take time

and anecdotal evidence points to some adoption in the latter half of 2014 and 2015.

■ Carriers. Carriers are looking to embrace SDN and network function virtualization to

extend new services, improve efficiency, and lower capex. As part of this effort, 20

leading carriers formed the Network Functions Virtualization (NFV) working group to

promote a standards-based approach to SDN and NFV, with the goal of leveraging

volume and standardized x86 and switches to extend services on premise and in

branch offices. While the deployment of NFV-type functionality in appliance format has

long been a thorn in the side of carriers, NFV maturity portends a more prolonged

cycle of technology adoption.

■ Hyperscale datacenter. Hyperscale customers have adopted the technology first, with

Amazon and Google adopting versions of OpenFlow technology. Outside of this,

Facebook and the Open Compute are working to develop a specification for an open

networking switch. This design will be agnostic to the software running on it and

consequently, further SDN efforts on separating hardware and software. The Open

Networking and Open Daylight Foundations are involved in the effort along with

Broadcom, Intel, Vmware, and privately-held Cumulus Networks. Outside of the

massive hyperscale datacenters, SaaS vendors are adopting the technology to

streamline their datacenters and improve opex.

18 September 2013


Strategies from all segments emerge. Given the rise of SDN over the long term, it is not

surprising that many vendors see this disruption as an opportunity. Traditional networking

companies such as Cisco and Juniper are developing and pushing their own strategies to

enhance their software and services profiles. Meanwhile, IT and telecom equipment

companies such as HP and Huawei are less concerned around their legacy installed base

and see SDN as an avenue for disrupting the market and capturing share. Software

companies such as VMware see the opportunity to absorb functionality and new emerging

companies such as Big Switch, Plexxi, Cumulus, PLUMgrid, and Embrane have stated

that they believe they can gain from added functionality and flexibility to the network. What

is clear is the move to SDN has already resulted in strategies around it being formed.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

41

Exhibit 60: SDN Overview—Competition from the Old and New

Data plane Controller/SW Apps

Earlier this year, Cisco announced its One Platform Kit or onePK,

similar to other programmatic interfaces for SDN. Cisco acquired

Cloupia for orchestration and has a tap application.

Similar to other SDN vendors, the company has four component SDN

strategy: i) forwarding; ii) control; iii) services; and iv) management. The

Contrail controller can use commodity hardware.

IBM’s SDN strategy has been focused on near to server networking in

virtual environments. IBM has a nascent orchestration offering in

SmartCloud, based on OpenStack.

With it’s telco heritage, the company is rolling out a solid SDN offering,

integrated with VMware and emerging orchestration platforms.

Nicira SDN is added functionality to VMware, allowing creation of virtual

switches, routers, firewalls, etc. and ease configuration. VMware Dynamic

Ops serves as an orchestration layer and Nicira has some L4-7 functionality.

While other vendors focus on controllers, PLUMgrid is geared toward

automation, provisioning and enhancing operations. The company

partners with OpenStack for extended orchestration.

Cumulus is a prototypical next generation SDN vendor in focusing

exclusively on SDN software.

Plexxi is unique SDN vendor in that it focus on performance in providing

an optical switch and leveraging the “affinity networking model” .

Orchestration

HP has long been a proponent of SDN and began collaborating on the

technology in 2007, with the predecessor to OpenFlow, Ethane. The

company has nascent orchestration and app offerings.

While not strictly defined SDN, Arista focuses on similar concepts in

automation and programmability and is market ready.

Similar to PLUMgrid, Embrane does not focus on core networking

elements, like SDNs but rather, the establishment of value-added apps.

While currently focused on the network visibility/network packet brokering

space through appliances, Gigamon’s strategy is to layer on SDN.

Big Switch is among the lead proponents of open controller architectures.,

and looks to penetrate the market with apps like Big Tap.

Legacy Commodity HWProduct

Insieme/ONE

Contrail

HP SDN

IBM SDN for

Virtual

Environments

Virtualized

Services

Platform

Software

Defined

Cloud

Networking

Network

Virtualization

Platform

Virtual Network

Infrastructure

Cumulus

Linux

Big Virtual

Switch/Big

Tap/Big Network

Controller

heleos

Affinity

Networking

GigaVUE-CV

Applet


18 September 2013


SDN—Gradual but real adoption The danger of analyzing the impact of something like the onset of SDN that has the

potential to be so disruptive to the industry structure, revenues, and margins of the

networking market, is that it can simply be overhyped. Indeed, the best example of this is

the opportunity discussed and hoped for with the mobile internet. While various

technologies such as WAP, club Nokia, and GPRS arrived in the late 1990s, it took until

arguably the entrance of the iPhone and mass coverage of 3G networks for the actual

potential to be unleashed. While we firmly believe the SDN threat is real, it may take

multiple years for this to have an impact. The risk this creates is that if indeed, as Gartner

notes, the technology is two to five years away or perhaps longer, it allows time for

incumbents to reposition. In order to arrive at a pulse of what the actual industry believes,

we recently, as part of the Credit Suisse Networking survey, aimed to get a sense of the

adoption dynamic and arrived at four main conclusions.

Exhibit 61: The Networking Hype Cycle SDN at Its Peak

Source: Gartner.

Awareness building, mainly in a PoC stage. In Exhibit 62, we note that 35% of

respondents either have or are currently evaluating SDN use cases currently. Looked at

alternatively, as shown in Exhibit 63, 64% of respondents are in proof-of-concept or pilot

stage.

18 September 2013


Exhibit 62: SDN Being Slowly Evaluated

Have you evaluated or you evaluating SDN?

Exhibit 63: SDN Still at a POC Stage

At what stage are you with respect to SDN deployments?

Yes35%

No65%

PoC50%

I don't plan to14%

Deploy in Develop/Test

21%

Pilot14%

Source: Credit Suisse survey. Source: Credit Suisse survey.

Deployments start in 18 months. Some 79% of respondents expect to deploy SDN within

24 months. Even with this adoption, spending on SDN services will only account for 13%

of spending.

Exhibit 64: SDN Deployed 16 Months Out

When do you expect to deploy SDN?

Exhibit 65: SDN Will Be 13% of Spend Three Years Out

What percent of your spend will be SDN based in 3 years?

-1

4

9

14

19

24

0%

5%

10%

15%

20%

25%

30%

CurrentlyDeployed

Next 6Months

Next 7-12Months

Next 13-18Months

Next 19-24Months

Over 24Months

NotApplicable

Mon

ths

Perc

enta

ge o

f res

pond

ents

11.5%

13.3%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

1 Year 3 Years

Source: Credit Suisse survey. Source: Credit Suisse survey.

Seen as disruptor from a wary industry. Perhaps the best reflection on the outlook from

SDN is that while it is seen as a disruptor to the market, we believe that there is a mixed

view on how SDN will impact networking spend over the next 3 years.

Mixed views on SDN adoption's effect on spend. Some 43% of respondents expect the

adoption of SDN to cause their overall networking spend to go down (will save them

money) and 36% to go up (will not save them money).

Exhibit 66: Adoption of SDN on Networking Spend

Adoption of SDN will cause overall networking spend to:

Spending will increase, 36%

Spending will decrease, 43%

Spending will not change, 21%

Source: Credit Suisse survey.

18 September 2013


We have included what we believe are some interesting quotes from respondents:

“SDN will have the impact on computing networking as virtualization was to

servers. It will centralize much of the networking control, thus reducing the overall cost. It

will usher in a major change in the way "IT" network is performed and it will add new

dimensions to other innovative changes in IT, such as Cloud.”

“I believe it is a generational shift in networking, albeit still in its infancy. There really

is no SDN market today, due to the general immaturity and lack of adoption of the

products. In 2-3 years I expect to see winners and losers emerge. My guess today is HP

as a winner and Cisco as a loser. Juniper may win, or at least break even, although even

they could have serious issues. The best of the startups today will likely get acquired by

today's major vendors.”

“This is a generational shift in networking. The market will have to mature for

widespread adoption. The disruption in network support organizations will be similar to the

disruption in development organizations when object oriented programming was first

introduced.”

“SDN will and is revolutionizing present thinking about network architecture. As we

integrate boots on the ground (sales, marketing, etc.) with production, SDN plays an

increasingly important role in satisfying the need for an elastic cloud architecture and on

demand network availability.”

“There is still more hype than content. The infrastructure aspect of networking is/was

intensive. The short answer to cloud it, just means the intensity has shifted, not gone

away. Newcomers will and have front ended traditional providers, much like the cellular

networking experienced 12 years ago. This will happen with SDN. With that said, the

marketing or buzz & hype IS a forerunner to the provisional technology shift. It will happen

after a settle out in the market occurs and the actual infrastructure shifts happen.

Customer interaction has to be smooth and transparent and that has yet to happen.

Basically, there has to be a shift from a capex/high opex mentality to a consumerization

mentality for it to really be a common and available tool."

“Too early to call. But history of this type of technology suggests companies should wait

for it to mature before jumping too quickly. Maybe as long as 3-4 years.”

“Once a standard has been fully identified I think this will be an open market for new

vendors to compete.”

“Big Switch, Extreme Networks, Cisco, Juniper will all be big players. SDN seems to

be getting more traction than previously expected, so no, I don't think it is [overly hyped].”

"Will wait to see if it gets traction in the enterprise before committing major funds to it."

"Arista seems to be a vendor that could make an impact in the SDN arena."

“I'm looking towards Cisco to play catch-up with some of the smaller players that have

entered the market. It's still early and not sure if it's more hype than reality. If the benefits

are true it could shake up the field if Cisco doesn't acknowledge and enter the race as

stated earlier.”

“1) SDN is for large datacenters and clouds. 2) SDN will evolve from "dumb" hardware

"smart" software concept to automation and adding services without changing hardware

concept. 3) I believe PlumGrid's approach is the most perspective approach.”

18 September 2013


Software Disrupting Networking In the previous section, we outlined how very powerful forces exist for the gradual uptake

of software defined networking. However what matters for investors are the real economic

implications over the medium and long term. We understand that any analysis looking at

the economics of SDN is in some way flawed, given its nascence if nothing else. However,

we do believe that our analysis is instructive in at least one way: it is unlikely that

networking vendors will be able to extract current levels of economic profit. Our extensive

analysis of networking profitability suggests there are four main consequences.

Industry structure—Is networking so special? When comparing the networking industry to

other areas of IT spend, such as servers, storage, and even markets like wireless

infrastructure, we see that the networking market is the least competitive, has the most

vertically integrated structure, and the highest profitability. In other words, it is closer in an

economic sense to an oligopolistic structure. We believe there are several reasons for this.

There is some merit to the argument that switching costs are very high (creating barriers to

entry), as networks are a distributed system, creating interdependencies across a myriad

of devices. This distributed and performance sensitive architecture creates a need for

custom ASICs and vertical integration. Ultimately, it is more of an important component of

the IT infrastructure. However, equally important, and perhaps a contrary argument, the

onset of SDN in theory allows the flattening of the network, increasing choice at the

hardware, controller API, and orchestration levels. This ultimately creates more

competition and in the end industry structure more akin to servers, and/or storage, where

gross margins can be between 15 and 30 percentage points lower.

Exhibit 67: Despite Margin Premiums, Networking Is Not the Most Prioritized Area Respondents asked y/y growth for IT spend among the following categories

-1.5%

-1.0%

-0.5%

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

Printing PC Server Services Software Networking Storage Security Overall IT MobileDevices and

Services

YoY

IT s

pen

d g

row

th f

or

20

13

Source: Credit Suisse IT survey, August 2013.

Lessons from Sun. At its basic level, the onset and uptake of SDN involves the breakdown

of the composition of basic networking. Rather than proprietary bundled hardware and

software architecture, it allows network administrators to build networking separately to

more efficiently address the needs of growing and complex data traffic. The decoupling of

hardware and software of course has occurred before. We believe the experience of Sun

Microsystems is analogous to the onset of SDN, in terms of analyzing new industry

dynamics that networking vendors will face. First the failure to adapt quickly for the

incumbent can be severe in terms of performance. Indeed, Sun's revenue share of the

server industry eroded from 17% in 2000 to 7% in 2009 given its failure to gain traction in

the x86 segment as this segment rose. Next, over this period, Sun's GM fell by 1,300bps

from 53% to 40%, driven by scale, partially a reflection of commoditization of its end

18 September 2013


market. The concern is even if Cisco and Juniper do adapt and embrace SDN, today's

economics may ultimately look increasingly like the 'good times' of networking. Indeed, we

note whether it is margins or revenues, some 43% of our respondents believe it will cause

revenues in networking to shrink.

Exhibit 68: 43% of Respondents Say SDN Will Reduce Their Total Networking Spend Respondents asked what impact SDN will have on their total networking spend

Spending will increase, 36%

Spending will decrease, 43%

Spending will not change, 21%

Source: Credit Suisse Networking survey, August 2013.

Margins over time. While our view of declining gross margin profitability may look alarmist

to some, we would note that even adjusting Cisco and Juniper for major acquisitions, the

direction of gross margins over time has been downward. In fact, we find an R squared of

78% between networking margins and time based upon Cisco and Juniper’s experience,

with an average annual decline of 0.5% per year being the normal trend over the past

seven to eight years. There are margin cycles around this, but the downward pressure is

not insignificant.

Shrinking GP dollars will become a major headache. SDN promises huge change,

disruption, and opportunities. However, among all the hyperbole, we examine the savings

SDN offers and the economic consequences. We have benchmarked various costs to give

an order of magnitude the cost that an SDN-based solution may offer. We have compared

the total cost of ownership (ex-labor, running costs) for basic Cisco switches for the

datacenter (i.e., Nexus 3064), along with the SMARTnet services and software contracts

that network administrators would usually sell. The alternative in a SDN-based

environment would be to buy the Cumulus Linux software subscription and top of rack

bare metal switch. We conclude that the TCO is dramatically lower at approximately 80%

and that in turn the gross profit dollars available to the vendors are 75% lower. This

suggests to us that even if traditional networking vendors create successful SDN

environments, the economics will be drastically different.

Networking—Is It So Special? SDN May Change That

Cisco dominates the networking landscape; over any prolonged period of time, Cisco has

sustained market shares of between 30% (service provider routing) and 77% (enterprise

routing), with Juniper typically taking second place in routing markets (although we

acknowledge that Huawei has increasingly gained share in service provider routing to

equal Juniper's share essentially). While there have been all forms of competition over the

years, as shown in Exhibit 69, the one constant has been that combined both vendors

control approximately 60% of industry revenues for switching and routing.

18 September 2013


Exhibit 69: Cisco Has Seen Many Competitors, but Seems to Have Been the Constant

1990-1995 1996-2000 2001-2007 2008-Today 2015

ACC Bay Networks Alcatel HP ????????

WellFleet Ascend Ericsson Juniper

Proteon 3Com Siemens Microsoft

SynOptics Newbridge Foundry Checkpoint

DEC Fore Extreme Arista

3Com Xylan NEC Aruba

IBM Cabletron Nortel Riverbed

Riverstone Avaya

Lucent F5

Juniper Shortel

Redback Huawei

Dell

Fortinet

Brocade

Polycom


An interesting point to note is that when viewed from a market share perspective, we note

that networking is perhaps one of the least competitive markets:

■ In storage the top vendor—EMC—only has a 31% market share, and the top three

vendors have 55%;

■ In x86 servers, the top vendor has a 33% share, and the top three vendors have 70%;

■ In wireless infrastructure, the top vendor has a 40% share, and the top three vendors

have 72%; and

■ In enterprise switching, we note that the number one vendor Cisco has a 65% share,

and the top three account for approximately 80%.

The networking market and specifically the wider networking market is possibly the least

competitive and most oligopolistic in nature of all the typical hardware segments. This

could be explained by several factors:

■ Networking is a distributed system, meaning that costs of switching infrastructure are

much higher. Originally, networking was designed in such a manner that if one area of

the network failed, all others would remain intact, meaning it served as a distributed

system. Given the intelligence is placed in so many nodes, this increases the need for

high-performance software and highly reliable hardware in terms of silicon. This

means that typically a given enterprise would select one vendor and remain with them,

especially as the protocols and hardware were and remain all proprietary. This

indicates that historically networking appliances, such as Cisco switches, have been

tested at scale, in turn creating a virtuous-circle and making it very hard for new

vendors to break in.

■ The switching costs are very high. Cisco in particular has invested significantly in

certified administrators for networks (Cisco Certified Entry Networking Technician—

CCENT, Cisco Certified Network Associate—CCNA, Cisco Certified Network

Professional—CCNP, Cisco Certified Internetwork Expert—CCIE). This creates a

significant inertia in moving vendors and platforms, as its network administrators not

only face the cost of replacing networking infrastructure, but these administrators will

in essence need to be retrained.

■ Networks affect the rest of the IT infrastructure. If a network fails, it impacts the

compute and storage IT as well, meaning the need for reliability and taking less risk

remains high.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

48

Exhibit 70: —Less Competitive, with Premium Margins and Average Innovation

Cisco

CiscoEricsson

IBM EMC

HP

Juniper

NSN HP

IBM

ZTE

HuaweiHuawei

Dell

NetApp

Juniper

Alcatel

Lucent

HP

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Switching Routing WirelessInfrastructure

All servers Storage

Share by IT segment …networking appears far less competitive

Source: Company data, Credit Suisse estimates

Compared with any other major area of IT or service provider spending

Networking is far less competitive, with fewer serious alternatives to the

number one competitor, creating an oligopolistic structure Q. Which segment has shown the most innovation in the past 5 years

and which do you think will in the next 5?

Source: Credit Suisse Networking Survey, August 2013

Networking delivering less innovation

0%

5%

10%

15%

20%

25%

30%

35%

Printing Servers PC Networking Software Services Storage Security

Perc

ent

of

Vo

te

Past 5 years Next 5 years

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

Switching Routing Storage All servers Wireless

Infrastructure

Ble

nd

ed

GM

%


…leading to significant premium in margins


18 September 2013


Additionally, it could simply be that there is a level of scale advantage and consistent

innovation that Cisco, for example, has produced, resulting in a fundamentally different

market structure to other segments. We believe that there is some truth to the fact that

Cisco has consistently over the years simply out-executed peers.

Alternatively, it could be that there has simply not been the architectural shift in networking

that say an opportunity such as SDN offered. Interestingly enough, based upon our IT

survey, we note that networking is perceived as delivering the least innovation of other IT

segments over the past five years, which is normally associated with markets that lack a

competitive structure. The difference is now we see scope for a potentially much more

competitive and different market structure.

More competition. With the altered industry structure, there will be more competition, as

firms go head to head to absorb increasing levels of functionality into their platforms and

increase the attractiveness of APIs. Certainly, as we show, traditional networking vendors

will play a part in this market, yet the primary question is whether they can replicate

current economies due to today's oligopolistic landscape in networking segments.

Expanded opportunity to add value. Not dissimilar from prior platform evolutions, we

expect that there will be an increasing opportunity to add value in conjunction with the

networking infrastructure. Here, API-enabled applications will bring value to network

optimization, security, and enhanced application performance while orchestration layers

will mitigate ever increasingly complex network architectures.

A market structure over time resembling other parts of IT spend could mean that

economics could change. The movement we believe over time could suggest a radical

change in industry structure, to not one where the top two vendors control over 60% of

revenues and 65% of profits (for switching and routing), but possibly more like the storage

market or even servers. We note in Exhibit 70 that the margins extracted for such

segments tends to be much lower indeed, whereas storage GM are 55% and servers are

lower still at 30%. Fundamentally however, we believe that SDN will over time result in a

competitive industry structure. We believe a quote from Frank Frankovsky, VP of hardware

engineering at Facebook, is relevant.

“We should be able to treat a switch like a server in the rack….We should be able to load a

Linux-based operating system, and that server just happens to have a lot of I/O ports on it."

Hardware and Software Decouple—Lessons from

Sun

At its basic level, the onset and uptake of SDN involves the breakdown of the composure

of basic networking. Rather than an architecture with proprietary bundled hardware and

software, it allows network administrators to build networking separately to more efficiently

address the needs of growing and complex data traffic. The decoupling of hardware and

software of course has occurred before with various generations of the PC industry, and

even within servers. In fact, the experience of Sun Microsystems is analogous to the onset

of SDN for several reasons.

■ Like servers, networking is seen as a core aspect of IT infrastructure, where reliability

is cherished, security is essential, and failure of the appliance will not be tolerated.

■ Sun operated in the server market very much like Cisco and Juniper do today. In the

UNIX market, it designed its own processors and ASICs, appliances, and bundled the

sale with the software operating system Solaris. As x86 servers delivered more

powerful compute; it ultimately became an alternative over time, whereby a commodity

x86 server could be purchased from Dell, HP, or a wide range of alternatives, and the

software was decoupled from suppliers such as Microsoft or Linux vendors. This to us

sounds very similar to the SDN alternative to today’s switching architecture.

18 September 2013


■ There was an underlying need as the rise of PC computing demanded that enterprises

deliver computing power at a lower price. This is similar to the building pressure on

datacenters today, with growing data traffic, mobility, and analytics all requiring more

networking fabric.

Of course, like all analogies, this is not perfect, and Sun Microsystems did have ample

opportunities to execute and change course. However, what is interesting is the end

impact on the company.

Failure to adapt meant the ongoing shift to x86 caused material revenue pressure. We see

in Exhibit 72 that Sun's revenues share of the server industry eroded from 17% in 2000 to

7% in 2009 given its failure to gain traction in the x86 segment, as the x86 server market

rose to become 60% of industry revenues from 40% between 2000 and 2009. Revenues

for Sun collapsed by 70% to $3 billion in 2009.

A rapid change in industry margins. Between 1998 and 2009, Sun's GM fell by 1,400bps

from 54% to 40%, partially a reflection of scale and partially a reflection of commoditization

of its end market. Indeed, we believe today that HP, Dell, and IBM's gross margins just for

hardware are somewhere between 27% and 35%, significantly below networking levels.

Exhibit 71: Between 1998 and 2009 Sun's GM's Fell by 1,400bps

-5%

5%

15%

25%

35%

45%

55%

-1,000

-500

0

500

1,000

1,500

2,000

2,500

3,000

3,500

19

98

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

Sun Operating Profit Sun Operating Margin Sun Gross Margin


Industry pricing dramatically changed. The server market has essentially failed to grow

over the past 16 years. Although the end market for servers has grown, and x86 servers

have taken share, the end server market has actually in revenue terms has stayed flat.We

also note that the mix shift to x86 meant an acceleration in pricing pressure. From 1996 to

1998, prices came down 40%, and in the subsequent 14 years, they came down at a 7%

CAGR.

Of course the above cannot be simply overlayed to Cisco, Juniper, or the networking

industry directly, but it should be considered the worst case of the failure to adapt. What is

clear, however, is that if this is applicable, the net effect would be:

■ Limited growth in traditional networking over time and possibly a long-term decline;

■ A fundamentally different GM structure not in the 60-70% level, but rather in the

mid- 50's; and

■ A very different industry structure in market share terms.

The concern is that even if Cisco and Juniper do adapt and embrace SDN, today's

economics may, in the end, look increasingly like the good times of networking.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

51

Exhibit 72: Lessons from Sun—When Hardware and Software Decouple...It Gets Ugly

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

0%

10%

20%

30%

40%

50%

60%

70%

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Sun s

erv

er re

venue s

har

e

x86 S

har

e

x86 share of total server revenue Sun Server Revenue Share

-

10,000

20,000

30,000

40,000

50,000

60,000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Serv

er

Reve

nues

x86 Other RISC IPF Unix

x86 share up, Sun-Micro server share down

… industry ASP’s saw pressure …server market has been ex growth

Gross Profit and GM’s under pressure

Source: Company data, InfoneticsSource: Company data, Infonetics

Source: Company data, Infonetics Source: Company data, Infonetics

Since the adoption of x86 servers, over any prolonged period industry revenues have peaked off at ~$50bn

With x86 servers increasingly in the mix server ASPs have declined at ~9% CAGR 1996 through 2012

1400bp compression in GMs from 1998-2009

30%

35%

40%

45%

50%

55%

60%

$0

$2,000

$4,000

$6,000

$8,000

$10,000

$12,000

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Gro

ss P

rofi

t

Sun Product Gross Profit Sun Product GM

0%

10%

20%

30%

40%

50%

60%

70%

80%

0

5,000

10,000

15,000

20,000

25,000

30,000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

To

tal S

erv

er

AS

Ps

Total Server ASPs x86 mix

Steep initial decline of 40%+ over 2 years

9% ASP decline over the last 5 years

Source: Company data, Credit Suisse research, Infonetics

18 September 2013


SDN Drives a Major Headwind on Profit Dollars

While Cisco and Juniper have successfully dealt with various forms of competition over at

least the past ten years, the competitive response of the company cannot be

underestimated. As we previously discussed, the onset of SDN does present an

opportunity and a need for a generational shift in how networks within datacenters are

constructed. The issue is to what extent this evolution can disrupt Cisco's operating

structure. We believe the net impact over the long term would result in a gradual

commoditization of its business, even if the timing and extent becomes the "very" long

term.

The example of Cumulus. We believe there are a host of Silicon Valley start-ups that are

eager to change the economics of the networking industry with the onset of SDN and

Cumulus Networks is an interesting example.

Cumulus plans to replicate for the networking industry what Linux did for the server

industry, when it helped break up the stranglehold that vertically integrated computing

companies such as Sun Microsystems had before x86–based servers became

mainstream. Cumulus is supporting Linux for multiple reasons. First and foremost it has a

featured networking stack, with many networking companies either already using it for

implementation or even before this by building it into their proprietary platforms. Second

there is a material variety of vendor-agnostic tools for managing and operating Linux

platforms available. Coupled proliferating controllers from a variety of vendor, Cumulus

Linux allows a feature rich commodity hardware-based switch.

The Cumulus model. At its basic level, the company aims to alter networking so that

network devices run the same way that servers do, meaning using the same kind of tools

and operational platforms for deployment and management. The idea is that this will be

particularly attractive to companies whose datacenters are constantly expanding.

Supply chain and ecosystem. The company has built a supply chain of original design

manufacturers (ODMs), distributors, value-added resellers (VARs), and system integrators

to build and deliver bare-metal switches with Cumulus' software on-board, including Acton

and Quanta and silicon from Broadcom.

Exhibit 73: Cumulus Ecosystem

Technology Partners ODM Silicon/Components Distributors Customers

Open Compute Project Accton Broadcom Penguin Computing Dreamhost

AGEMA Quanta Finisar Synnex Fastly

CF Engine UNIXSurplus

Cloudscaling

Metacloud

Opscode

Piston Cloud

PLUMgrid

Puppet Labs

Quanta

10Gtek

VMware


A change in cost structure. Given the nascence of software defined networking as

technology, while the debate will grow around the strength of the offering or the approach

of one vendor versus another, it is possible that Cisco and/or Juniper will embrace SDN

successfully. The question for investors is whether they will be able to extract the same

level of economics from the networking budgets. In Exhibit 116 we show a simplistic

analysis of the impact of SDN in economic terms. We have compared the total cost of

ownership (ex-labor, operating costs) for the basic Cisco switches for the datacenter (i.e.,

Nexus 3064), along with the SMARTnet services and software contracts that network

18 September 2013


administrators would usually sell. The alternative in a SDN-based environment would be to

buy a Cumulus Linux software subscription and top of rack bare metal switch. Our analysis

as shown in Exhibit 116 highlights the following main issues.

The TCO is dramatically lower. Even assuming that the Cumulus approach has a shorter

switch life of three years on a TCO basis, the cost is some 88% lower. The saving comes

from effectively unbundling the solution between hardware and software.

The gross profit dollar difference is 74% lower. An additional issue is that, whereas Cisco

will deliver a gross margin of 60% and 67% on its products and services in fiscal 2014, in

the Cumulus scenario, we believe the gross margin is lower in aggregate at 56%. Possibly

more importantly, the gross profits dollars are 74% lower. The point here is that even if

today’s networking vendors were to embrace an open SDN strategy, the gross margin

percentage and gross profit dollars would be significantly lower.

Can Cisco do the upsell? Seems unlikely. We believe Cisco and others understand this

risk, even if they continue to highlight the alternative advantages of integrated ASICs,

software, and services. Indeed, we believe the push into services or new software is

partially to offset this decline. Nevertheless, in order for Cisco to be able to be gross-profit

neutral versus today's economics, the customer would either need to spend an additional

$8,000 or 3.3 times versus the alternative from the Cumulus option or this much value in

Cisco’s wider offering and pay a premium. While we do not dispute the value of Cisco’s

installed base, trained administrators, new services, and new software it offers, this level

of difference would certainly lead us to conclude that the headwind on gross profit dollars

is significant. Moreover, Cisco’s push into expanded software offerings comes at a time

when Linux-based platforms are set to growth their application, developer, and services

ecosystems.

The above analysis is simplistic in many ways. We acknowledge that the above analysis is

simplistic in many ways as it excludes the costs of labor, power, and cooling, all which

have an impact on the relative TCO. Additionally, while hard to place a numerical value on

the risk of using a new supplier, the inertia of Cisco network administrators cannot be

underestimated. We do believe, however, that the relative magnitude in the economics of

today versus tomorrow present a headwind.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

54

Exhibit 74: SDN—Threatening the Last Bastion of Proprietary IT Profits

Cisco Proposal USD

Cisco Nexus 3064 list price 48,000

Cisco Nexus 3064 realised price 20,000

License 4500

Maintence 6,000

Average Switch Life years 5

TCO 54,500

TCO per year 10,900

Cisco Switching & Services GM 70%

Cisco GP 7,630

Cumulus Proposal

Top of rack bare metal switch 6,000

Cumulus Annual Subscription 1,500


TCO 10,500

TCO per year 3,500

Cost Saving -68%

Hardware GM 30%

Hardware GP 600

Software GM 90%

Software GP 1,350

Total GP 1,950

Blended GM 56%

Difference -74%

Cisco Value add needed to be GP neutral 5,680

Cisco Upsell to be GP neutral 8,114

Multiple of Cumulus Propasal 3.32

1) A look at the economics of SDN. Perhaps the most significant issue for the networking

segment is what happens to economics with the onset of SDN. Even if traditional

networking vendors shift their portfolio, strategies and software to a SDN defined

approach can today’s economics can be replicated? Our analysis compares on a

simplistic TCO basis the relative costs savings to the customer and gross profit headwind

to networking vendors is around 70%

2) Cisco basis Nexus 3064 DC switch. We base our analysis on a lower end Cisco Nexus

3064 datacenter switch with a list price of $48,000. However we believe the price to most

enterprises would be more than 50% lower. Additionally Cisco will license the OS and

charge maintenance, assuming a standard 5 years switch life the TCO is $54.5k or

$10.9k per annum.

3) Cumulus Proposal, 70% cheaper per year. With Cumulus the customer buys a basic

commodity switch and they charge an annual subscription of $1,500, bringing the TCO to

$10,500, $3,500 per annum. Even assuming a lower life for switch the TCO is 80% lower.

4) Gross Profit Dollars at risk. In the Cisco proposal the GP dollars available (assuming a

70% GM) are $7.6k, in the cumulus proposal (assuming standard hardware commodity

GMs) the GP is $1950m some 74% lower. In other words even if networking vendors

offer this, it significant shrinks the profits available.

5) Can Cisco achieve the upsell? We believe that Cisco is aware to some degree of this

and will attempt to upsell services, or argue for higher value add (sales-force, support,

legacy, reliability), however the needed upsell needs to worth 3.3x the Cumulus proposal,

this seems like a stretch.

6) A simplistic analysis, but…This analysis ignores other costs like labor, cooling, power,

and does not take into account the value of support, alternative features or

platform/portfolio but does highlight the disruptive economics of SDN, and the economic

surplus being earned in networking. Moreover, it does not take into account the full

solution sale. For example, with Nexus 3000-series switches sold, Cisco would bundle a

Nexus 7000 router, which has carries more upgrades and support over time, which would

support margin and allow Cisco to get more aggressive on the Nexus 3000 switches.

Even with this discounting, SDN solutions would be less expensive with a Nexus 7500

selling for $200-250k versus $100-150k for an SDN alternative.


18 September 2013


Mean Reversion on Margins

In the neoclassical theory of the firm, excess economic returns in competitive markets will

be competed away. In other words, absent barriers to entry, economic profits will be

competed away. As previously mentioned there are reasons (reliability, innovation and

technology) why networking margins in segments such as switching and routing have

sustained a material premium to margins in other areas of IT and even communication

equipment. We would note, however, that preliminary inspection would suggest that gross

margin in networking at least for Juniper and Cisco have been on a long-term decline for

some time.

Cisco core gross margins have been sliding. A glance at Exhibit 112 shows Cisco's gross

margins over time. Here, we present the corporate gross margin and the adjusted GM, the

later having been adjusted for the impact of major acquisitions (such as Scientific Atlanta,

Tandberg, and the impact of UCS revenue growth that we believe is meaningfully GM

dilutive as discussed previously). This is a simplistic analysis and we acknowledge that we

have not adjusted for every acquisition, which given the software bias, would actually

imply that underlying margin declines have been more severe. What is clear is that over

any prolonged period of time, GMs have been under pressure. We note that whereas once

Cisco gross margins were 70%, today they are 63%.

Exhibit 75: Cisco GMs Have Been Declining over the Long Term in millions, unless otherwise stated

50.0%

55.0%

60.0%

65.0%

70.0%

75.0%

FY00 FY01 FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 FY13

GM % GM Cisco (adjusted)

Apart from the GM recovery FY01 to 03, Cisco GM have been on a

slow and steady GM trajectory on an adjusted basis (ex impact of

major dilutive events) down by 800 bps in 10 years


A decent correlation with time. In Exhibit 112 and Exhibit 113 we show the gross margins

for Cisco and Juniper over time. While each only include approximately ten data points,

what is interesting is that R squared is decent, in other words over time networking gross

margins have been shown to erode at a rate of ~50bps per year.

18 September 2013


Exhibit 76: A high correlation between Cisco GM and time in millions, unless otherwise stated

Exhibit 77: ...and for Juniper

FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 FY13

y = -0.0052x + 0.7236

R² = 0.7233

60.0%

62.0%

64.0%

66.0%

68.0%

70.0%

72.0%

74.0%

76.0%

GM reported GM Cisco (adjusted)


y = -0.0056x + 0.7049

R² = 0.8522

60.0%

62.0%

64.0%

66.0%

68.0%

70.0%

72.0%

GM Juniper

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

Cisco seems to be guiding for this. At the company’s analyst meeting in 2012, Cisco

published not only its revenue and gross profit breakdown, but also its targeted levels for

the next three to five years. What is interesting is, based upon our estimates, margins in

switching and routing may come down in the long term. (See Exhibit 114)

Exhibit 78: Cisco Has Implied that GMs in Switching and Routing May Decline in millions, unless otherwise stated

Gross Margin

FY16 Target

FY12 FY13E Low Mid High

Switching 71.5% 70.8% 66.4% 67.0% 67.5%

Routing 61.2% 61.3% 60.0% 60.5% 61.0%

Service Provider Video 30.0% 42.0% 40.7% 41.0% 41.3%

Datacenter 26.2% 31.0% 27.9% 28.1% 28.3%

Wireless 69.4% 71.0% 64.8% 65.3% 65.9%

Security 71.3% 71.3% 77.1% 77.7% 78.3%

Collaboration 61.1% 61.1% 63.0% 63.5% 64.0%

Other Products 54.9% 28.0% 54.2% 54.7% 55.1%

Services 66.7% 67.5% 66.6% 67.2% 67.7%

Total Revenue 62.4% 62.5% 61.0% 61.5% 62.0%


Note we have estimated these based upon revenue and GP from analyst day 2012.

18 September 2013


UCS—Gaining Share in Servers One of Cisco’s key growth initiatives in recent years has been the growth in its datacenter

business, which has now grown successfully to a $2-billion business in fiscal 2013. The

principal driver has been the company's Unified Computing Systems (UCS) server

platform that now commands a 6% share in x86 servers. (See Exhibit 88.) We believe that

this level of share gain can continue, and we see this business growing to $2.9 billion by

fiscal 2015, based upon several factors.

x86 servers some pockets of strength. We see the x86 server market driving 3% long-term

revenue growth to $37 billion. However, we note very different dynamics by price point and

type of server.

■ 2-way/below $2,000: The growth side of a bifurcating segment. Strong unit demand

should propel unit demand for sub-$2,000 2-way servers. These servers are more

stripped down relative to x86 servers, geared toward virtualization but satisfy the

demands of the most rapidly growing workloads, web-centric, and HPC. However,

given commoditization, we see ASPs facing stagnation. In aggregate, as shown in

Exhibit 79, we expect revenues to grow at a 9% CAGR from 2010-15.

■ 2-way/above $2,000: As virtualization increasingly pressures units, we expect that 2-

way servers geared toward virtualization will see increasing unit pressure, despite

workload growth. Machines that support multiple virtual machines benefit from a

higher memory footprint and management functionality. Memory remains a key

constrain in virtualization, and as virtualization adoption and virtualization density

continue to increase, configuration should improve.

■ 4-way and above: Here we see unit decline accelerating. Similar to the virtualization

geared machines in the 2-way, $2,000 and above category, 4-way and above faces

pressures from virtualization, but to a greater extent, since more of these machines

are virtualized and can hold more virtual machines. As a result, units should see

significant pressure, something rising ASPs will be unable to offset.

Blade market seems to slowing. It is important to note that for UCS some 71% of units and

76% of revenues are based upon blade servers, according to Gartner. For blades, we see

this as a microcosm of the trends in the high-end x86 market due to the price points. While

blades help the management paradigm within the enterprise, with reduced cabling and

management overhead, the economics have become less favorable for vendors with the

growth of cloud architectures. As a result, blade revenue growth has ebbed. Given the

relative premium for blades is about 56%, it is understandable that the unit share of the

aggregate x86 market continues to fade and ASPs only support steady revenue share.

With this, we expect blade market growth, given higher ASPs, to tread water.

Share heading to 30%? With 20% of the blade server market, Cisco has the opportunity to

continue to gain share; we would not be surprised to see Cisco grow its share to 30% of

the market. This is increasing likely with the questions surrounding the future of IBM’s x86

business. Moreover, HP has been the leader, but has seen its share continue to decrease,

as Cisco challenges the industry leader with strong technology and a unique,

network-centric, systems management approach. However, we would note a cap to how

far this may extend. Interestingly, Cisco is gaining share in the blade market with

higher-priced, more richly configured systems. The ASP for a Cisco blade is higher than

that of IBM and HP, while from an architectural perspective, the solution costs less due to

lower switch content. We believe that Cisco's servers, given their ability to manage

workloads more efficiently, are being configured for heavier workloads and thus have

higher ASPs.

18 September 2013


x86 Server Outlook—Pockets of Strength

Over time, x86 servers have become key building blocks of various types of IT

infrastructures, ranging from pedestrian printing to cloud and high-end appliances such as

Oracle’s Exadata. The segment comprises 71% and 98% of server industry revenue and

unit, respectively, and represents the only growing subsegment of the broader server

market. While total x86 revenue has grown at a CAGR of 5% between 2005-2010 as seen

in Exhibit 79, there are certain subsegments of the market, specifically, 2-way servers

under $2,000, that are enjoying very healthy revenue growth (19% CAGR 2005-2010 and

9% 2010-15). This growth is offset by a 4% decline in 1-way from 2010-15 Considering

these market dynamics, we forecast the server market along the following categories.

■ 1-way: This category represents generic infrastructure servers largely used for file and

print, networking, security, and systems management. Effectively, these units run

ancillary functions in datacenters and small departmental and remote office workloads.

■ 2-way under $2,000: The critical 2-way or 2-processor x86 category, which accounts

for over half of server units, is bifurcating based on applications and workloads. Less

expensive servers, or those under $2,000, are being leveraged for web and high-

performance computing workloads for which having inexpensive compute power is

critical.

■ 2-way over $2,000: This category represents the more richly configured 2-way servers

that are being used as server virtualization platforms. Specifically, these machines

tend to have a larger memory footprint to support higher numbers of virtual machines.

These servers are the mainstay of traditional enterprise datacenters.

■ 4-way and above: This category represents more powerful servers that are

increasingly used in virtualization environments in which IT managers look to increase

virtual machine densities. In addition, given the increasing mission-critical nature of

x86, these systems are increasingly being used to absorb database workloads housed

on UNIX.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

59

Exhibit 79: Credit Suisse Worldwide Server Forecast—x86 Market Summary

Source: Gartner Servers Quarterly Statistics Worldwide Database, August 2013. Credit Suisse estimates for 2013-2015.

1) 1-way: Market should see slower growth. The 1-way market should see headwinds to growth as they lack a growth impetus and are challenged relative to 2-way servers on price/performance

metrics. 1-way servers, unlike 2-way servers and web and HPC specific workloads, lack a substantive growth driver.

2) 2-way/below $2,000: The growth side of a bifurcating segment. Strong unit demand should propel unit demand for sub-$2,000 2-way servers. These servers are more stripped down

relative to x86 servers geared toward virtualization but satisfy the demands of the most rapidly growing workloads, web-centric and HPC.

3) 2-way/below $2,000: ASPs face stagnation. Unlike other segments in the x86 market, these servers are not richly configured to support increasing numbers of virtual machines and are

relatively more simple in design. As such, commodity pressures increasingly factor, although there are elements of increased configurations as commodity elements become more richly configured.

4) 2-way/above $2,000: Virtualization increasingly pressures units. We calculate that the 2-way servers geared toward virtualization will see increasing unit pressure, despite workload growth.

5) 2-way/above $2,000: ASPs should see a marked benefit from richer configurations. Machines that support multiple virtual machines benefit from a higher memory footprint and

management functionality. Memory remains a key constrain in virtualization and as virtualization adoption increases and virtualization density continues to increase, configuration should improve.

6) 4-way and above: Unit decline accelerates. Similar to the virtualization geared machines in the 2-way, $2,000 and above category, 4-way and above faced pressures from virtualization,

but to a greater extent since more of these machines are virtualized and can hold more virtual machines. As a result, units should see significant pressure, something rising ASPs can only

partially offset.

Revenue (in $mn) 2005 2006 2007 2008 2009 2010 2011E 2012E 2013E 2014E 2015E CAGR '00-'10 CAGR '05-'10 CAGR '10-'15

x86

x86 1-w ay 2,678 3,098 3,673 3,568 2,938 3,245 3,482 3,040 2,859 2,785 2,699 1.2% 3.9% -3.6%

x86 2-w ay 18,107 19,101 21,541 20,156 18,018 24,168 26,117 28,804 29,917 29,903 29,374 8.3% 5.9% 4.0%

x86 2-w ay <$2,000 864 1,168 1,342 1,986 1,838 2,058 2,569 1,937 2,255 2,645 3,103 26.4% 19.0% 8.6%

x86 2-w ay >$2,000 17,243 17,933 20,199 18,169 16,180 22,110 23,547 26,867 27,662 27,258 26,271 7.5% 5.1% 3.5%

x86 4-w ay and above 4,872 4,908 5,087 4,788 3,772 4,568 5,975 5,685 5,353 5,540 4,788 -3.8% -1.3% 0.9%

Total 25,657 27,106 30,301 28,512 24,728 31,981 35,574 37,529 38,130 38,228 36,861 4.5% 4.5% 2.9%

Seq. change (%) 10.8% 5.6% 11.8% -5.9% -13.3% 29.3% 11.2% 5.5% 1.6% 0.3% -3.6%

Units (in thousands) 2005 2006 2007 2008 2009 2010 2011E 2012E 2013E 2014E 2015E CAGR '00-'10 CAGR '05-'10 CAGR '10-'15

x86

x86 1-w ay 1,386 1,552 1,835 1,901 1,599 1,853 2,021 1,959 1,989 2,039 2,080 7.7% 6.0% 2.3%

x86 2-w ay 5,245 5,707 6,059 6,291 5,369 6,467 6,875 7,148 7,434 7,598 7,793 10.3% 4.3% 3.8%

x86 2-w ay <$2,000 548 767 886 1,332 1,372 1,566 1,972 1,478 1,676 1,927 2,216 28.6% 23.4% 7.2%

x86 2-w ay >$2,000 4,697 4,940 5,173 4,958 3,997 4,900 4,903 5,671 5,759 5,671 5,577 7.9% 0.8% 2.6%

x86 4-w ay and above 372 438 473 468 330 359 426 399 364 366 307 -0.6% -0.8% -3.0%

Total 7,004 7,697 8,366 8,660 7,299 8,678 9,323 9,507 9,788 10,003 10,180 8.9% 4.4% 3.2%

Seq. change (%) 14.1% 9.9% 8.7% 3.5% -15.7% 18.9% 7.4% 2.0% 3.0% 2.2% 1.8%

ASP (in $) 2005 2006 2007 2008 2009 2010 2011E 2012E 2013E 2014E 2015E CAGR '00-'10 CAGR '05-'10 CAGR '10-'15

x86

x86 1-w ay 1,932 1,997 2,002 1,877 1,837 1,751 1,723 1,551 1,438 1,366 1,298 -6.0% -1.9% -5.8%

x86 2-w ay 3,452 3,347 3,555 3,204 3,356 3,737 3,799 4,030 4,024 3,936 3,769 -1.9% 1.6% 0.2%

x86 2-w ay <$2,000 1,577 1,523 1,515 1,491 1,339 1,314 1,303 1,311 1,346 1,373 1,400 -1.7% -3.6% 1.3%

x86 2-w ay >$2,000 3,671 3,630 3,905 3,664 4,048 4,512 4,802 4,738 4,803 4,807 4,711 -0.3% 4.2% 0.9%

x86 4-w ay and above 13,084 11,200 10,765 10,234 11,437 12,739 14,023 14,233 14,688 15,130 15,584 -3.2% -0.5% 4.1%

Total 3,663 3,522 3,622 3,293 3,388 3,685 3,816 3,947 3,896 3,822 3,621 -4.0% 0.1% -0.4%

Seq. change (%) -2.9% -3.9% 2.8% -9.1% 2.9% 8.8% 3.5% 3.4% -1.3% -1.9% -5.3%

Source: Gartner Servers Quarterly Statistics Worldwide Database, August 2013. Credit Suisse estimates for 2013-15.

18 September 2013


x86 Server Market—Positioning Critical

We highlight the following key drivers of the x86 server market over the next several years,

outside of core compute demand. First, internet traffic (Web tier), Web application, and

high-performance computing (HPC) workloads are generating strong demand for low-end

volume servers. Second, compelling x86 price performance relative to UNIX, driven by a

strong roadmap and a steady cadence of chip releases from Intel, provides incremental

growth. Finally, we see strong and continued developer and enterprise customer support

for the x86 platform. This is accentuated by the rising adoption of x86 virtualization.

(1) x86 Volumes Being Driven by Web/High-Performance Computing Workloads

There are discernible trends within the low end of the x86 server market that are worth

highlighting. While we expect commodity x86 1-way servers to continue to enjoy

mid-single-digit type volume growth driven by the need of generic infrastructure workloads,

the 2-way under $2,000 market should enjoy robust volume growth. We expect revenues

and volumes in this subsegment of the x86 market to grow by a CAGR of 8% between

2010-15 owing to strong underlying demand in Web/HPC workloads for which these

servers are best suited. As seen in Exhibit 80, Gartner estimates Web application, front-

end, indexing, and HPC workload growth of 13%, 13%, 14%, and 12% between 2010-15,

respectively, or 13% in aggregate. This compares with the relatively anemic 11% growth

for other types of workloads, and this is likely overstated because these are more likely to

be virtualized, resulting in lower physical server counts.

Exhibit 80: Gartner—Proportion of Server Shipments by Workload

2007 2008 2009 2010 2011 2012 2013 2014 2015 CAGR '10-'15E

Web Applications 12% 12% 12% 13% 13% 14% 13% 13% 14% 13%

Infrastructure 14% 13% 10% 9% 8% 7% 7% 7% 7% 5%

OLTP DBMS 9% 9% 9% 9% 9% 10% 9% 9% 9% 12%

Email/Messaging 10% 11% 11% 11% 12% 13% 12% 12% 12% 14%

Front-End 11% 11% 11% 12% 12% 13% 13% 13% 13% 13%

Indexing 9% 10% 10% 10% 10% 12% 11% 11% 11% 14%

Streaming Media 5% 5% 6% 6% 6% 7% 6% 6% 6% 13%

HPC 7% 7% 8% 8% 8% 9% 8% 8% 8% 12%

Data Warehouse 4% 4% 5% 5% 5% 5% 5% 5% 5% 12%

Collaboration 5% 5% 6% 7% 7% 1% 7% 7% 7% 13%

Virtual Desktop 1% 1% 1% 3% 3% 4% 4% 4% 4% 17%

Other 14% 12% 10% 7% 6% 5% 5% 4% 4% 1%

Web/Front-End/HPC/Indexing

% of total w orkloads 39% 40% 41% 43% 43% 48% 45% 45% 46%

Seq. change (%) 15% -8% 35% 12% 16% 12% 12% 12% 13%

Other

% of total w orkloads 61% 60% 59% 57% 57% 52% 55% 55% 54%

Seq. change (%) 10% -12% 24% 10% -4% 26% 12% 12% 11%

1) Front-end (web-tier/ Internet traffic), web applications, index ing and high performance computing (HPC) workloads. These workloads are among the biggest and fastest

growing. In aggregate, Gartner expects these workloads to grow at an 13% CAGR, faster than most of the other areas. These fast growing workloads do not require highly

configured servers with a significant memory footprint.

2) These three workloads account for a good portion of x86 growth. Excluding these relatively high growing workloads, all others are expected to grow at a more anemic 11%.

CAGR 2010-15. We note that this 11% CAGR is overstated when viewed from a physical server perspective given that many of these are virtualized workloads.

2

1

1

1

1

2

Source: Gartner.

Looked at in another way, increasing time spent online is driving demand for servers

dedicated to serving up Internet traffic, content, and applications. Herein, we highlight the

persistent and relatively high-level growth in the amount of minutes users spend online,

which is likely causing strains on back-end servers and providing an impetus to build out

more capacity.

(2) High-end x86 Continues to See Pressure and Blades Are No Exception

We expect to see continued pressure on the enterprise x86 as virtualization adoption

continues. We estimate that this increasing adoption of virtualization will result in more

highly configured x86 servers growing at a mere 3% y/y, although ASP increases should

lift revenue growth to 4% for the category. More powerful x86 servers, which can support

more virtual machines, continue to see yet more pressure. Here, we forecast unit declines

18 September 2013


of 11% y/y, with ASP support mitigating revenue declines to 7% through 2015. In short, we

continue to see significant pressure on tier 1 OEM enterprise servers due to consolidation

and over time, white box and Open Compute initiatives. We highlight virtualization’s

negative calculus in Exhibit 81.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

62

Exhibit 81: Virtualization and The Impact on Server Units US$ in millions, unless otherwise stated

1) Installed base. We start with the installed base of x86 servers that

are likely to be used in virtualization environments. Then using a

replacement rate of 4.2 years, which we believe is reasonable

based on the results of the Credit Suisse IT Survey, we estimate

the number of units that effectively fall out of the installed base

every quarter and the amount of units that need to be shipped to

replace them.

2) Gross additions. We model growth in new server shipments

using historical shipment rates that are in excess of the

replacement rate. We conservatively assume these shipments are

at pre-recession highs, prior to the mass virtualization.

3) New servers virtualized. As a base level, we conservatively

assumed 29% of all new shipments are being virtualized with that

number stepping up in the future. IDC expected 22.3% of all

machines being virtualized by 2013 while our Survey respondents

expect 61% of their new machines will be virtualized over time.

4) VMs per server. We adjust this by the number of virtual machines

coming from legacy servers vs. newly created ones. Our survey

results pointed to 13 VMs per machine in 2011 and 21 in 2013.

Our data indicates that 33% of virtual machines are coming from

legacy physical servers. This is a critical finding since it points to

the fact that only 1/3 of virtual machines are a result of

consolidation with 2/3 for organic workload growth.

5) Physical servers needed to support virtual machines. We

calculate the number of physical servers necessary to support the

newly created virtual machines, replacing physical servers.

Because the consolidation effect is somewhat mitigated by organic

VM growth, compression is not as significant as it potentially as it

might be otherwise. Nevertheless, there is a significant number of

servers annual lost to virtualization, approximately 2.3 million this

year and running to 3.2 million units by next year.

2011 2012 2013E 2014E 2015E

2-way over $2,000 server units (in millions)

Servers needed to maintain installed base 4.6 4.5 4.3 4.1 4.9

Incremental units in excess of refreshing installed base 0.3 1.2 2.9 4.8 4.8

Total theortical server demand 4.9 5.7 7.2 8.9 9.8

% of servers virtualized 29% 34% 38% 42% 48%

Virtualized physical servers 1.4 1.9 2.7 3.7 4.7

Virtual machines per server 12.6 16.2 20.5 24.3 29.0

% of virtual servers coming from legacy physical servers 0.3 0.3 0.3 0.3 0.3

% of virtual servers coming from w orkload grow th 0.7 0.7 0.7 0.7 0.7

Virtual machines per server coming from legacy servers 4.2 5.4 6.8 8.0 9.6

Virtual machines/server from organic w orkload grow th 8.5 10.9 13.7 16.3 19.4

Physical virtualized servers meeting demand 0.3 0.4 0.4 0.5 0.5

Non-virtualized servers 3.5 3.8 4.5 5.2 5.1

Total servers 3.8 4.1 4.9 5.7 5.6

YoY growth 0.1% 15.6% 1.6% -1.5% -1.7%

Physical servers lost to virtualization 1.1 1.5 2.3 3.2 4.2


18 September 2013


For blades, we see this as a microcosm of the trends in high-end x86 due to the price

points. While blades help the management paradigm within the enterprise, with reduced

cabling and management overhead, the economics have become less favorable with the

growth of cloud architectures. (See Exhibit 82.) As a result, blade revenue growth has

ebbed. (See Exhibit 83.)

Exhibit 82: Blades: More Expensive Than Volume Servers 2Q13 ASP for x86 2-way

Exhibit 83: Revenue Growth for Blades Has Declined Worldwide x86 blade annual revenue growth

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

Rack mount Blade

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Source: Gartner. Source: Gartner.

Given the relative premium for blades, it is understandable that the unit share of the

aggregate x86 market continues to fade and ASPs only support steady revenue share.

(See Exhibit 84 and Exhibit 85.) With this, we expect blade market growth, given higher

ASPs, to tread water.

Exhibit 84: Blades: Units in a Share Downtrend Blade units as % of x86

Exhibit 85: Revenue Share Has Modesty Increased Blade revenue as % of x86

0%

5%

10%

15%

20%

25%

30%

0%

5%

10%

15%

20%

25%

30%


Cisco Gaining Share with UCS

In 2009, Cisco entered the server market with its Unified Computing System (UCS). The

offering, comprised of compute, networking, and management resources is geared toward

improving the efficiency of virtualized datacenters. Given Cisco’s networking heritage, the

product obviated the need for multiple switches in the compute infrastructure and offered a

centralized approach, easing management overhead and cost. Effectively, while

competitors have to use a switch at each chassis level, Cisco allows for one centralized

switch. This approach proves more simple to manage and less expensive as the one

switch is amortized across chassis.

18 September 2013


Exhibit 86: Cisco's Advanced Blade Design Streamlines Networking

Blade

Traditional blade architecture

Blade Blade Blade

Blade Blade Blade Blade

Switch



Switch



Switch

Switch

Blade

Cisco blade architecture



Switch eliminated





Switch

Blade

Switch eliminated

Switch eliminated

Source: Company Data, Credit Suisse research.

These advantages helped Cisco quickly gain a solid server footprint. (See Exhibit 87 and

Exhibit 88.) We note that higher ASPs associated with blades have helped Cisco gain a

disproportionate amount of server revenue share.

Exhibit 87: Cisco Server Unit Share % of worldwide x86 units

Exhibit 88: Cisco Server Revenue Share % of worldwide x86 revenue

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%


Blades Servers Dominate Cisco UCS

With white box servers increasingly a factor in the lower cost, higher volume server market

given the need for inexpensive compute and consolidation in traditional enterprise, it is not

surprising that blades dominate the UCS segment, both on a unit and revenue basis. (See

Exhibit 89 and Exhibit 90.)

18 September 2013


Exhibit 89: Blades Account for 70% of UCS units… Blades as % of Cisco units

Exhibit 90: ... and Nearly 80% of Cisco UCS Revenue Blades as % of Cisco UCS revenue

40%

50%

60%

70%

80%

90%

100%

40%

50%

60%

70%

80%

90%

100%


At a high level, Cisco blade revenue has grown extremely well. (See Exhibit 91.) Given

that blades are used to consolidate traditional enterprise infrastructures, rather than by

large hyperscale properties, we are concerned that growth could be limited.

Exhibit 91: Cisco Blade Revenue Has Grown Well Off a Small Base Y/Y % change in Cisco blade revenue

0%

50%

100%

150%

200%

250%

300%

350%

400%

450%

500%

1Q11 2Q11 3Q11 4Q11 1Q12 2Q12 3Q12 4Q12 1Q13 2Q13

Source: Gartner.

Cisco blade revenue has grown well at the expense of traditional server vendors IBM and

HP. Here, we note that Cisco has quickly taken share in the blade market and this has

allowed for growth. (See Exhibit 92.) Cisco is now the second leading blade vendor in the

like-for-like x86 category, recently topping IBM. We believe Cisco’s approach to server

management can lead to share gains in an ex-growth category but this has limited upside

potential over time.

18 September 2013


Exhibit 92: Cisco Has Gained Share In the Blade Market % of worldwide blade revenue

Exhibit 93: Cisco an Emerging Leader in Blades % of worldwide blade revenue

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

0%

10%

20%

30%

40%

50%

60%

HP IBM Cisco


With 20% of the blade server market, Cisco has the opportunity to continue to gain share.

We would not be surprised to see this grow to 30% over time, especially given the

uncertainty of IBM’s x86 business. Moreover, the leader, HP has seen its share continue

to decline, as Cisco challenges the industry leader with strong technology and a unique

management approach.

Interestingly, Cisco is gaining share in the blade market with higher priced, more richly

configured systems. The ASP for a Cisco blade is markedly higher than that of IBM and

HP, while from an architectural perspective, the solution costs less due to lower switch

content. We believe that Cisco servers, given their ability to manage workloads more

efficiently, are being configured for heavier workloads and thus have higher ASPs.

Exhibit 94: Cisco Blades Carry a Premium ASPs for 2-processor x86 blades

-

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

IBM HP Cisco

Source: Gartner.

Unfavorable Margin Dynamics

While servers are a rapidly growing product line for Cisco, the margins are below the

corporate average. Traditional x86 gross margins are in the 25-30% range and these

products compose the majority of the Cisco configuration. Meanwhile, Cisco interconnects,

which are higher at 75%, form the minority. Even with these generous margin

assumptions, we estimate the blended UCS margin to be in the low 30% range, well below

the corporate average. (See Exhibit 95.)

18 September 2013


Exhibit 95: Cisco UCS: High 30% Gross Margins as Servers Dilute Switches Revenues and gross margins

Cisco UCS 6248UP Switch

2 x $24,589

M3 blade server + chassis




8 x $7,853 + $2,850

8 x $7,853 + $2,850

8 x $7,853 + $2,850

8 x $7,853 + $2,850

70% gross margin

25% gross margin

25% gross margin

25% gross margin

25% gross margin

Sample Configuration Price

Blended margin 32% gross margin

Source: CDW, Credit Suisse research.

18 September 2013


Aiming for the IT Stack The first era of IT was arguably centered around mainframes and terminals. The second

era of IT was based around the PC, client server architectures, and smaller-scale, static

internet. We are now well within the third era, which is based around mobile devices,

richer content, more users that have driven the secular trends of cloud computing, broader

mobility, social networking, and Big Data. In this third era there are billions of users and

millions of applications that need support, causing IT to be fundamentally

re-architected. Driven by these needs and a less robust IT spending environment, we are

seeing companies that previously partnered now compete more intensely, whether it be

Oracle and HP, Cisco and EMC/VMware, or Cisco with HP. The opportunity is to become

a trusted end-to-end IT supplier, which in itself is attractive, given a $2 trillion TAM. After

our extensive review of the convergence strategies of the major enterprise vendors, we

arrived at three main conclusions.

The shift toward the integrated supplier will be slow. While there is an abundance of

literature addressing the $2-trillion market opportunity, our survey noted that 35% of

respondents do not even agree with the notion of converged infrastructure and 51% of

respondents prefer a best-of-breed approach when selecting vendors for different parts of

the network. This suggests quite emphatically to us that the movement to address new

segments as an integrated vendor will be at best a slow approach for the industry.

Additionally, we note our survey showed that 46% of respondents said that whether there

is a shift toward either the private or public, it will result in lower IT spending over time.

IT stack analysis: Everyone has their gaps. In Exhibit 100 we present the current offerings

from what would be described as the key IT players who have evolved their enterprise

infrastructure over time. It is interesting that even many strong vendors in their respective

segments have sizeable gaps. For example, Cisco is a strong networking player, has a

strong emerging server business, and has relationships for storage. However, it has

sizeable gaps in the areas of middleware, database software, and enterprise software. In

this respect, being able to become a more strategic player such as IBM could prove

challenging. We see similar issues for Juniper and NetApp.

The industry seems to appreciate the wider options from Cisco, IBM, and MSFT. Looking

at the strategies and evolution in a different way, we surveyed IT decision makers on the

key metrics that we believe determine success in the IT market, such as roadmap,

management, price/performance/reliability, management, sales, and cloud strategy. What

was clear is that Cisco is perceived to have the core competencies and ranked top three

across these metrics when benchmarked versus its peers, scoring well on price and

performance, management, and roadmap. We note IBM and Microsoft also seemed to do

well. Looking at cloud specifically, Cisco rated well along with IBM and EMC. We do

believe all these vendors have the capacity to leverage these competencies into new

segments and grow the TAM. In our view, the only question would be the rate at which

adoption occurs.

Chasing TAM Expansion Is the Norm

What is different in this third era of IT is that historically competition remained within

distinct IT categories such as servers, storage, software, PCs, and networking. Vendors

within these segments operated independently, and alliances were built where necessary.

The difference now is that there is a gradual convergence in all areas of IT vendors, which

means previous partners become competitors and all vendors are strategically evolving

into end-to-end IT partners. There are many examples of this building convergence, which

creates new areas of conflict.

■ Oracle making a push on its hardware platform through the acquisition of Sun

Microsystems and its Exadata and Exalogic systems. This has directly placed the

company into competition with former hardware partners, such as HP and IBM.

18 September 2013


■ In 2009, Cisco entered the server market, gaining decent headway with its UCS

offering and creating tension with HP and Dell.

■ Following the acquisition of 3Com, HP has made decent initial traction in the

enterprise switching market directly competing with Cisco.

■ VMware is pursuing a strategy around the software defined datacenter that in essence

threatens the wider IT market even over parts of EMC's own storage business. This in

turn strains the business partnership on the VCE alliance.

All of these moves are not based on the desire for more competition in each segment, but

are more about ensuring the long-term survival for each business and becoming a more

trusted IT partner. Furthermore, it opens up the opportunity for TAM expansion. For

example, rather than Cisco addressing just the $37 billion networking market, it may one

day aim for more expansion into servers, or even storage and enterprise software, to

attempt to expand its TAM significantly.

Exhibit 96: Chasing a Wider IT Wallet—A $2 Trillion Dollar Market

Servers, $56bn , 2% PCs/Smartphones/ Tablets, $174bn , 7%

Networking, $40bn , 1%

Software, $354bn , 13%

Storage, $28bn, 1%

Internal Services, $526bn , 20%

IT Services, $906bn ,

34%

Printers, $47bn , 2%

Telecom Services, $526bn , 20%


In this context, for the networking stocks, such as Cisco and Juniper, this evolution is not

necessarily about outright competition with each other, but is more about competing with

the wider IT industry veterans such as IBM, HP, and Oracle within a larger TAM. From our

analysis of the opportunities around this, we arrive at two main conclusions.

Bundling versus Best-of-Breed—Shift Will Be Gradual

Few would doubt the strength of Cisco or Oracle’s salesforce or distribution. The question

is whether bundled or best-of-breed solutions are preferred. While the debate around this

is not a new issue, we would highlight, as noted in Exhibit 97 and Exhibit 98, that at least

from a purchasing perspective, there appears to be a strong bias toward

best-of-breed products. In the Credit Suisse IT Survey, some 51% of IT decision makers

noted a preference for best-of-breed versus purchasing from one vendor. In other words,

while the strategies of Cisco, for example, may already be evolving to be the one IT

vendor of choice, the purchasing of today’s customer base suggests any shift will be

gradual. Indeed, an interesting quote from our survey highlights this point:

“I think the notion of a single vendor supplier is a very dangerous one. And the larger

suppliers on this list, particularly Oracle and SAP, should not be trusted with all of our

business. They abuse any power they have now.”

18 September 2013


Exhibit 97: 51% of Respondents Prefer Best-of-Breed Question: If you were to build (or currently have) an internal private cloud, you would:

Exhibit 98: 36% of Respondents Don't Want One Supplier Question: In 2014, if you could choose to purchase the majority of your IT hardware/software/services from one vendor, who would it be?

Choose a single vendor that could provide all of the elements – only if

it was cheaper31%

Choose a single vendor that could provide all of the

elements –regardless of costs

18%

Build a solution from several best

of breed hardware, software,

networking and services vendors

51%

I don’t agree with the “converged

infrastructure” or “one IT supplier”

notion36%

Hewlett Packard15%

IBM8%

Dell5%

Cisco13%

Microsoft3%

EMC/VMWare13%

Source: Credit Suisse IT Survey, August 2013. Source: Credit Suisse IT Survey, August 2013.

The Cloud May Not Be That Incremental

Nowadays, rarely a conference call, presentation, or trade journal article goes by without

some reference to the revolutionary benefits of cloud computing as extolled by the IT

industry. However, what matters in the context of this report is what cloud computing

means for the IT industry from the viewpoint of investors.

In theory, if all workloads shifted to the public cloud, then the purchaser of IT infrastructure

would shift from enterprise IT departments to service providers. For example, we note that

certain service providers, such as Google, actually manufacture their own servers, while

other service providers, such as Rackspace, buy commodity components from IT vendors,

such as Dell. The location and the ongoing operations of the underlying hardware and

software are essentially transferred. When viewed from this perspective, the shift toward

cloud computing will only be partially incremental. As service providers grow in size and

numbers, the efficiencies they gain will allow them to use the same amount of IT

equipment (and headcount) to service a larger number of customers. Indeed, some 61%

of respondents in the Credit Suisse IT Survey felt that, as a result of cloud computing, IT

spending would be flat to down. (See Exhibit 99.)

Exhibit 99: Cloud Computing Will Reduce IT Spend in the Coming Years Question: Adoption of cloud computing will cause your overall IT spend to:

To grow39%

To decline46%

Will not change our IT spend

15%


1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

71

Exhibit 100: The IT Stack: IBM Has the Fullest Portfolio

Oracle IBM HP EMC Cisco SAP Microsoft Juniper Dell NetApp

E, M, T-Series/EX,

QFX-Series/Qfabric

7%

Operating system: $29bn

Management: $73bn

Applications: $168bn

Application dev. & deployment: $82bn

Middleware: $19bn

Database: $28bn

EqualLogic/

Compellent

7%

FAS/E-Series

11%

Virtualization: $3bn

Servers: $53bn

Networking: $37bn

Storage: $53bn

StorageTek/Sun

Storage

1%

IBM

DS/XIV/Storw ize

13%

3PAR/EVA/MSA

9%

VMAX/VNX/Isilon

34%

Partner/VCE/FlexPod

Pow erEdge

15%FlexPod/Partner

Partner/

IBM System

Netw ork

1%

ProCurve/3Com

7%VCE

Nexus/Catalyst/

Insieme/ASR/CRS

55%

Force10/Partner

1%FlexPod/Partner

OracleVM

1%

Pow erVM

7%

VSE

2%

VMw are

85%Partner/VCE/FlexPod

x86/SPARC/

Exadata/Exalogic

5%

System z/p/x

30%

ProLiant/Integrity

27%

VCE/Lenovo/

Partner

UCS/VCE/FlexPod

3%

Window s

79%

Hyper-V

1%FlexPod/Partner

Solaris/Linux

1%

AIX/zOS

6%

HP-UX Non

Stop/OpenVMS

3%

VMw are

Oracle

DB/TimesTen

MySQL

45%

DB2

18%

Vertica/Partner

0%

Greenplum/

Partner

0%

Sybase

5%

SQL Server

20%

NetWeaver

1%

.NET

5%

Java development

tools

22%

Rational/ILOG

20%

Quality/

Performance

Center

1%

Pivotal/Spring

Source

0%

Pow er Builder

6%

Fusion

Middlew are

16%

WebSphere

33%

VMw are/Pivotal/

Partner

0%

Backup Advisor/

RecoverPoint/

Control Center

5%

Partner/VCE/FlexPod

Visual Studio

11%

System Center

35%

Quest

1%

On Command

2%

E-business/

Fusion applications

6%

Lotus

Domino/Notes

3%

Webex/Partner

1%

ERP/

SuccessFactors

7%

Microsoft

Exchange/Lync/

Dynamics

14%

Tivoli 10%

OpenView /

Insight Manager

5%


18 September 2013


IT Strategies in Evolution Oracle: For Oracle, covered by Credit Suisse analyst Phil Winslow, the focus is on

expanding its enterprise wallet share, leveraging its core premium, growing its database

footprint, extending this down the stack with Sun operating system/hardware and up in the

stack with strong horizontal on-premise application financials, HR, supply chain, and CRM.

As part of this effort, Oracle is looking to consolidate the former Sun hardware business

and turn it into an application platform, akin to the iPhone. Here the company contends

that the focus of IT managers is optimal application performance regardless of underlying

infrastructure, and as a result, optimizes Sun hardware to the server as the platform for

apps could result in TCO advantages. From our perspective, the core competitive

advantage of high performing hardware and on-premise applications will be increasingly

challenged by off-premise PaaS and SaaS models and the changing nature of

applications, which we discussed in our May 30, 2013 report, “IT Hardware: “Un” clouding

Storage – EMC Shines”.

IBM: IBM is positioned as a global technology infrastructure solutions provider that is

honed with its “Smarter” initiatives. In providing the full solution, with services, hardware,

and software, IBM is a one-stop solutions shop. This one-stop shop approach serves the

company well, as it targets the C-Suite with a services-led solutions model. By selling

solutions and services, IBM can attach the solution to its own software and proprietary

hardware, making an implementation 'sticky'. As technologies emerge, the company is

able to appropriate them on top of its stack and integrate them with the existing solution.

An example of this integration is Linux and Java workloads to mainframe. Similar to

Oracle, the company faces a challenge, as its solutions are relatively more expensive in

the context of the changing nature of applications and the espoused architecture is not

ideal for volume workloads. As a result, commodity infrastructures on premise are likely to

pose an increasing risk to the company. We detail our analysis in our August 6, 2013

downgrade report, titled, “IBM: Not so Easy Blue…”.

HP: In the enterprise, HP is targeting a converged strategy, bringing storage, servers, and

networking together. Given that HP is fairly unique in having a solid footprint across the

three product groups, this approach is rational. Surrounding these hardware assets is

management software that is bundled. HP continues to strive to improve its hardware

portfolio with initiatives such as Moon Shot, storage with software such as StoreOnce, and

networking with SDN. In addition, the company continues to move up the stack slowly with

Fortify code analytics and ArcSight security. Importantly, HP continues to make an

analytics push with the big data solution HAVEn (Hadoop, Autonomy, Vertica). In services,

the company has several broad offerings in infrastructure outsourcing and

application/business. While these services help the company garner a footprint, they are

not as C-Suite-centric as those of IBM. In our view, HP has several challenges. First, the

company plays in the lower elements of the stack where increasingly servers are seeing

commodity pressure, storage is seeing pure play consolidation, and networking is facing

entry from software-led vendors with cloud-managed offerings. Second, while the

company is attempting to increase offerings up the stack, the focus on infrastructure

leaves the company exposed to market pressures, such as lower cost for on-premise

infrastructures and off-premise workload migration.

EMC: EMC’s broad strategy is to take advantage of growth markets through continued

share gains, consolidation, and best-of-breed products. In the secularly growing storage

market, EMC continues to consolidate share, from 32% last year to 35% today, through

the acquisition of products at the leading edge of storage technology, SSDs, and volume

and service-level technology. While EMC storage is but one part of the IT stack, VMware,

which accounts for 20% of EMC revenue, is emerging as the datacenter operating system,

amalgamating server, and networking and storage functionality. While VMware leads in

consolidating traditional datacenter applications, Pivotal remains uniquely positioned to

capture next-generation application growth, which is expected to have a long-term CAGR

18 September 2013


of 70% or 4 times those of traditional applications. While Salesforce.com, Amazon, and

Microsoft PaaS offerings are solely off premise, Pivotal allows for the internal PaaS for

those that do not wish to go off premise or find the economics unappealing.

Cisco: Cisco dominates the networking market and is looking to fend off erosion of the

market by opening its platform with onePK. Similar to EMC, the company is looking to

extend out of networking with the UCS effort being one initiative. With UCS as part of

VCE, Cisco is increasingly brought into traditional enterprise datacenters. Nevertheless,

Cisco lacks critical in-house capability, notably across storage. Perhaps more importantly,

Cisco lacks any meaningful presence in further up the stack application, infrastructure, or

platform software. This may increasingly pressure revenues and margins as the core

networking business is exposed to competition.

SAP: SAP, covered by Credit Suisse analyst Phil Winslow, is dominant on the application

side, providing leading ERP solutions for customers. The company is extending its

application reach and importantly, how the data from the applications is used to improve

decisions. For this, the company created its HANA in-memory database appliance to

provide analytics to business users rapidly. Outside of these efforts, the company has

vertically integrated the Sybase database into its offering in an effort to provide a broader

solution. In our view, the company is challenged by off-premise SaaS firms, with natively

developed applications. To offset this, SAP has allowed its software to be used by

Amazon Web Services; however, the competitive dynamic is evolving.

Microsoft: Microsoft’s enterprise strategy revolves around several elements. First, the

company seeks to be the volume platform, with operating systems ranging from PCs to

datacenters to virtualization. Perhaps more importantly, the operating systems' dominance

allowed the company to build a loyal developer base. The company seeks to continue to

leverage this platform strategy with Azure, an off-premise PaaS, that eventually could be

brought on premise. In this way, the company hopes to compete with Amazon,

Salesforce.com, etc. Secondly, the company has a significant number of non-operating

systems, developer assets that it can bundle that are often 'good enough' for the needs of

SMB, and some enterprises, including ERP applications, SharePoint, SQL Server

databases, and other products. As a result of the sheer volume of these products,

Microsoft can be a top wallet share holder and a competitor to firms that have more

premium and category leading products. The company has several weaknesses, which

our interlinked. First, the off-premise dynamic introduces new competitors as new

applications are created. Second, as a part of this, there is a 're-platforming' occurring in

enterprises, from VMware as an OS to Amazon Web Services and Pivotal as platforms.

This could erode Microsoft’s core strength over time.

NetApp: NetApp’s focus is on providing storage and data management capabilities for

enterprises and service providers. Given the significant levels of data growth (40% for

external arrays by our estimates), this strategy is understandable. To extend its datacenter

presence, NetApp partnered with VMware and Cisco to create a VCE alternative.

Furthermore, to leverage its storage, NetApp is offering a Hadoop distribution.

Unfortunately, we are concerned that NetApp is seeing increased competitive pressure

from smaller private vendors, EMC, and alternatives for large-scale data storage. With few

assets outside storage, this may be an issue for the company over time.

The Industry Says…C sc , IBM, and MSFT

Assessing the quality of each companies' IT stack is challenging from an external

perspective. As such, we surveyed 40 main IT buyers on the portfolio. Based upon this

survey, customers believe that the vendors with the core competencies to address

increasing levels of the IT stack are Cisco, IBM, and MSFT, with Huawei, HP, and Juniper

ranking quite poorly.

Cisco ranked well on core competencies. Within our survey, we asked IT buyers not only

to rank the companies competing for this wider IT wallet and to score each vendor on a

18 September 2013


given metric ranging from product portfolio to roadmap, but also rank the quality of the

salesforce and cloud strategy. Our survey was aimed not at product-level network buyers,

but at IT decision makers. As such, we believe it gives a decent reflection of each player.

Interestingly, Cisco scored the highest, with a top-three position in every category. The

company scored notably well on product portfolio management and distribution. (See

Exhibit 101.) Cisco was followed closely by Microsoft and IBM, with the latter scoring well

on roadmap sales and management. Companies that scored poorly were no surprise: HP,

Dell, and Huawei.

Exhibit 101: Cisco, IBM, and MSFT Compete Well in Terms of Core Competencies Please rate the following IT vendors across the attributes below, using the following scale: 1 = Very poor performance 5 = Excellent performance

Source: Company data, Credit Suisse IT Survey, August 2013

Cisco, IBM, and EMC building cloud credentials. From a vendor perspective, we believe

that the components of cloud computing are similar to existing datacenter architectures.

Recently announced cloud offerings by all major vendors, including IBM, HP, and Dell, are

fundamentally similar to infrastructure outsourcing services in that they have provided their

service arms for several years. The real change is to ensure that their portfolios are

aligned with new purchasers—service providers, customers seeking private cloud

implementations, and customers seeking public cloud services directly from the vendors

themselves. The question becomes whether the shift towards cloud computing could allow

newer entrants for IT infrastructure to win share. We note that, according to the Credit

Suisse IT Survey, Cisco is heavily favored to win incremental cloud infrastructure business,

as seen in Exhibit 102, followed closely by IBM and EMC.

18 September 2013


Exhibit 102: Cisco Is Likely to Be a Key Beneficiary of Cloud Deployments On a scale of 1-5 (1= unlikely, 5= very likely), how likely are the following vendors to GAIN/LOSE significant revenue as cloud computing adoption increases?

0%

5%

10%

15%

20%

25%

30%

35%

Huawei SAP Juniper NetApp Dell Oracle HewlettPackard

Microsoft IBM EMC Cisco

Perc

enta

ge o

f R

esp

on

den

ts

Company will gain share with transtion to the cloud Company will lose share with transtion to the cloud


While no one vendor has yet mastered the complete stack of hardware, software, and

services (as shown in Exhibit 100), with many of their products and strategies still

evolving, we would highlight the following in order to determine the impact on the

technology industry.

Who is showing the most innovation? While the shift to the cloud will be gradual, perhaps

the best indicator in the IT environment is companies that are expected to show true

innovation and ones that are perhaps not. According to our survey, over the past five

years, Cisco is appreciated as showing the most innovation; however, on a

forward-looking basis, it drops to fifth behind NetApp, Microsoft, IBM, and SAP.

Exhibit 103: Cisco Has the Biggest Drop in Innovation Perceptions Which company has shown the most innovation in the past 5 years and which will in the next 5?

0%

5%

10%

15%

20%

25%

30%

35%

Dell Huawei Juniper HewlettPackard

EMC Oracle Cisco IBM Microsoft NetApp SAP

Past 5 years Next 5 years


18 September 2013


Major Headwinds on Margins Perhaps one the hardest aspects of predicting Cisco’s operating outlook comes to the

dynamic on margins, simply because there are so many different drivers of profitability—

business mix (UCS, services, software upselling), competitive pressure, acquisitions as

well basic cost management. Also complicating matters is that Cisco only disclose gross

margins at the product and services level. We have looked at the outlook for margins

through five different approaches, and the end conclusion that we believe is that the

pressure on profitability will remain more to the downside. As shown in Exhibit 105, we

project 26% LT OM, given several factors.

Segmentation analysis—Negative UCS mix shift, all depends on the software and services

upsell. Shown in Exhibit 106, our proprietary segmentation analysis for Cisco’s gross

margins by business segment reveals several major conclusions. First and not

surprisingly, UCS is materially GM dilutive; we believe margins are around 30%, half of

corporate levels, given that we believe growth will increasingly come from share gains

expansion here and on an even higher scale that seems unlikely. Second, approximately

43% of incremental gross profit or 43% that Cisco targets in the medium term comes from

the ability to mix-up from services (and software). Here we believe that the company has

shown decent growth of late and we acknowledge that a strong installed base should

mean that Cisco increasingly upsells software and services. The only doubt we have with

this part of this strategy is that it involves upselling software, restructuring contracts, as

well competing with large IT companies such Microsoft and IBM, compared to which we

believe Cisco's IT portfolio still has gaps. Third, we believe that there is a combination of

disruption to networking GMs over time from SDN. On balance, we believe that GMs over

time will erode.

Industry structure and SDN Gross profit dollar headwind. As we discuss extensively in our

sector note and in the section "SDN an opportunity to shake up margins," while the impact

may be over a five-year view, we believe SDN will alter the dynamics of the networking

market, whereby at its most basic level, hardware, software, and services may be

delivered in a far less integrated manner. It remains to be seen how the traditional

networking vendors adapt to this change; however, what we believe is that based upon

our analysis, SDN will present a major change and headwind in terms of gross profit

dollars.

GM over time, the long trend is downwards. Stepping back from the segmentation

analysis, we note that adjusting Cisco over time for major acquisitions, underlying gross

margins have continued to come down, whether it is because they are high versus other

areas of IT spend or competitive pressures, we believe this is something to note.

Interestingly enough, Cisco's own guidance suggests a continued drop in the switching

GM’s over time.

SDN a change to industry margins structure beckons, no matter who wins. We have

compared the cost and total cost of ownership (ex-labor, running costs) for the basic Cisco

switch for the datacenter the Nexus 3064, along with the Smartnet services and software

contract that network administrator would usually sell. The alternative in an SDN world to

buy Cumulus Linux software and top-of-line rack bare metal switch, we demonstrate that

the Cumulus Option is some 70% cheaper, but also will result in gross profits of hardware

and software combined which is 74% lower. In other words, even if Cisco were to embrace

SDN potentially the GP dollars available significantly shrink. While this is simplistic in

many ways, we believe that it does highlight the headwind facing the traditional margin

structure of Cisco and something that we believe the software and service upsell cannot

offset easily.

18 September 2013


Exhibit 104: Cisco, HP, and Juniper Expected to Be the Most Disrupted On a scale of 1-5 (1=not vulnerable, 5=very vulnerable) which of the following vendors do you think is the most vulnerable to SDN?

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

NEC Brocade Dell AlcatelLucent

Ericsson Fujitsu ZTE Huawei Juniper HewlettPackard

Cisco

Perc

enta

ge o

f R

esp

on

den

ts


Huawei’s impact will be felt internationally. Huawei has already made an impact in the

service provider market, capturing a 16% share of this segment over time. Further, we

believe that it has the potential gain to share in the enterprise segment given its focus on

this segment and re-organization (away from telecom infrastructure), significant R&D base

(70,000 employees), and potential to be disruptive on pricing. While the company may

struggle for traction in the United States, we believe the remaining 80% and 90% of

service provider routing and enterprise switching markets are ripe for traction.

Cost management has been decent, danger of a perennial restructure. One aspect of

Cisco’s margin management that we believe has been solid is simply the focus on the cost

base. Here we note that the recent 4,000 employee workforce rebalancing shows this and

the need to consistently pursue operational excellence. We assume a long-term opex to

sales ratio of ~34% is sustainable given this. Equally however, we would note from a

revenue-per-head basis, or compared with other major IT companies, Cisco’s opex base

seems relatively lean.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

78

Exhibit 105: Cisco Gross Margin Segmentation—Long Term, We Expect GMs Around 60%

1) Proprietary segmentation analysis for GM. Based upon disclosure at the time

of the FY11/12 analyst meetings we have estimated GP and GM by business

line, based upon this we believe there is a secular underlying pressure and GM.

While the company targets 61-62%, we believe the risk is that over time they

undershoot this.

2) Switching and routing GM’s long term pressure. Interestingly Cisco’s own

guidance suggest GM’s declining. We see switching dropping to 66% and

routing to 59% given several factors

1) Increased competitive pressure in routing (mix shift to edge)

2) The gradual impact of SDN

3) Competitive factors in markets ex the US (Huawei)

3) UCS meaningfully dilutive. We estimate GM’s here at 31% roughly half the

corporate average. Given slowing growth in the blade server market, incremental

growth through share gains become challenging given it already has an 20%

share now, we essentially assume this rises to 30% over time.

4) WLAN risk of gradual commoditization. Based upon the FY12 analyst

meeting we believe Cisco’s GM’s in the wireless business to be ~71%, at a

material premium to the corporate avg. We question the sustainability of these

margins given that Enterprise WLAN (Aruba, Ruckus) is highly competitive.

5) Services and software the savior? Over the medium term Cisco targets some

43% to come from services growth. We accept that the company has a decent

installed base, and can achieve some of this, however equally we believe that

without material acquisition the needed level of expansion would need for the

company to become a strategic supplier in IT where it currently has material gaps

ex networking.

6) OPEX rising to $19bn…$2.13 EPS LT. An important focus of Cisco’s margin

management has been the solid focus on the cost base. The recent 4,000

employee restructuring shows this, as well as a general focus on operational

excellence. We assume a LT OPEX to sales of ~34% is sustainable and on a

revenue/head basis Cisco’s OPEX base seems lean compared to other major IT

companies.Source: Company data, Credit Suisse estimates

Revenue

In $ millions FY14 FY15 FY16 Target 13-16 CAGR CS FY16 13-16 CAGR

Switching 15,239 15,734 16,242 3.3% 16,049 2.9%

Routing 8,522 9,136 9,970 6.6% 9,501 4.9%

SP Video 4,836 4,806 5,066 1.4% 4,857 0.0%

DC 2,516 2,893 2,814 10.7% 3,182 15.4%

Wireless 2,708 3,077 2,573 5.9% 3,385 16.0%

Security 1,374 1,429 1,528 4.3% 1,472 3.0%

Collaboration 3,963 4,083 4,985 8.0% 4,207 2.1%

Other 398 359 724 2.9% 323 -21.4%

Services 11,267 12,139 13,990 9.8% 12,989 7.1%

Revenue 50,823 53,656 57,892 6.0% 55,965 4.8%

Gross Margin

FY14 FY15 FY16 Target Growth bps CS FY16 Growth bps

Switching 69.0% 67.2% 67.0% (382) 65.4% (5)

Routing 60.6% 59.1% 60.5% (74) 57.6% (4)

SP Video 42.0% 42.0% 41.0% (100) 42.0% -

DC 30.0% 29.0% 28.1% (289) 28.0% (3)

Wireless 70.0% 67.5% 65.3% (566) 65.0% (6)

Security 71.8% 72.3% 77.7% 635 72.8% 2

Collaboration 61.6% 62.1% 63.5% 235 62.6% 2

Other 28.0% 28.0% 54.7% 2,667 28.0% -

Services 68.0% 68.5% 67.2% (34) 69.0% 2

Gross Margin 62.1% 61.3% 61.5% (104) 60.5% (2)

Total Opex 17,132 18,213 19,394 5.1% 18,973 4.3%

OPEX % Sales 33.7% 33.9% 33.5% -0.9% 33.9%

In $ millions FY14 FY15 FY16 Target 13-16 CAGR CS FY16 13-16 CAGR

Operating Income 14,427 14,676 16,210 5.8% 14,727 2.5%

Operating Margin 28.4% 27.4% 28.0% -0.2% 26.3%

Net Income 11,485 11,714 13,297 7.0% 11,815 2.8%

Share Count 5,462 5,502 5,542 1.1% 5,542 1.1%

EPS 2.10 2.13 2.40 5.8% 2.13 1.7%


18 September 2013


Segmentation Analysis—UCS Dilutive

We believe that any analysis of Cisco margins needs to look at the margins by business

line. The challenge is that Cisco does not disclose them. Here we have relied upon charts

initially disclosed at the Cisco FY12 analyst meeting that gave a breakdown of gross

profits by business line, as well as Cisco’s targets. We show our estimates on a full-year

basis by business segment in Exhibit 106 and would highlight three main conclusions.

UCS materially gross margin dilutive… According to our estimates, UCS delivers a GM of

25-30%, some 30% lower than the corporate average or at half the level. This puts Cisco

in-line with most server peers. For example, we believe Dell operates at close to 27%.

While the company has a much higher ASP than Dell, we note that it has far lower scale in

terms of purchasing x86 chips, so we believe on balance that our estimate makes sense.

While the company has made no secret of the impact of server growth, the magnitude of

the growth is somewhat surprising and from a mix perspective, the robust growth seen at

Cisco indicates that the company will face a headwind from this on a long-term basis.

Exhibit 106: Cisco GM by Business Line Significant Dilution from UCS

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

Switching Security Wireless Services Corporate

average

Routing Collaboration Other SP Video Data Center

FY12 FY13


…growth and GM expansion for UCS? We note that UCS today stands at a revenue run

rate of over $2.0bn per year and this has been a clear share gainer in the server market.

However, equally we would note that in the segments in which Cisco operates, i.e., the

blade server market, the market is ex-growth long term. We believe that if Cisco is to

continue to grow revenues, either it will need to become truly dominant or it will need to

embrace lower price points, especially given its blade server share stands at 20%. Our

estimate of $3.3bn assumes it raises market share to 30% effectively over time.

Wireless Networking, performing well, but GM headwinds? Based upon the FY12 analyst

meeting, we believe that Cisco delivers GMs in the wireless business of around 65%, at a

material premium to the corporate average. This itself is not surprising given that Cisco

competes with peers reporting similar gross margins at Aruba. This segment is expected

to grow at a rate of 15% and as such this provides a GM tailwind. The only concern that

we hold is that, given the highly competitive nature of Enterprise WLAN, remaining

accretive GM’s will actually decline. (See Exhibit 107.)

18 September 2013


Exhibit 107: WLAN a Competitive Market Although Cisco Dominates with over 50% Share

Cisco, 51%

Other, 20%

Aruba, 11%

HP, 6%

Ruckus, 5%

Motorola, 5%

Meru, 2%

Ubiquiti, 1%


Building Software and Services Some Challenges

Looking at the above analysis another way in terms of incremental gross profit dollars, we

note that over the medium term, Cisco targets some 44% to come from services growth.

We note that to date the company has been successful in transitioning the overall revenue

mix towards services. We note that services now represent some 22% of overall Cisco

revenues, up from 19% in 2010. Additionally, within services Cisco is executing relatively

well and shifting the focus towards smart services over time, where GM’s are higher than

today’s services average of 67%. We note that in terms of incremental GP dollars, it

seems that some 44% comes from services. Put another way, for Cisco to deliver on its

long-term vision and offset competitive pressures, the execution of higher software and

services attach is critical.

Exhibit 108: Cisco Is Targeting Some 44% of GP Dollars to Come from Services

$25

$26

$27

$28

$29

$30

$31

$32

$33

$34

$35

$36

$37

$ B

n

$30.40

FY16 GPFY13 GP

$35.60

Switching Routing ServicesOthers Wireless DC SP videoSecurity Collaboration

1%

44%

19%

14%

7%5%

4%3%3%


This transition towards software and services, if it occurs, could be very powerful given

several factors.

■ It has the potential to change Cisco’s value add and annuity stream as an IT supplier

giving less volatility to revenues. Today this is 22% of revenues and Cisco aims to

expand this by 25-30% over time.

18 September 2013


■ It can elevate the company’s margin structure and deal with several competitive

elements and other mix factors.

The question is, can Cisco transform and perform a migration such as IBM performed

beginning in the late 1990s. We believe that the arguments for and against this means that

it will not necessarily be straightforward and assume a lower ramp. (See Exhibit 108.)

What is Cisco trying to sell? A key target that Cisco has emphasized over the past 12

months is to increase its recurring revenues through a combination of software and

services expansion. On the software side, we believe through the various embedded

software such as the IOS, the Nexus OS Cisco has always developed software. However,

now the company targets raising its software-related revenues from $6bn to $12bn, this

we believe is being driven by several factors.

Charging for software separately. Even though Cisco remains adamant that integrated

ASIC, software, and product strategy is a differentiator, the company will target to charge

for software separately. Indeed as noted by CEO John Chambers in late 2012:

“When you think about where we're going in software, remember software is across the

board for us. But we've got to learn how do we charge for it differently. How do we charge

in terms of recurring capability? How do we not get caught in bidding hardware and

software bundled together, against somebody who just provides hardware for a given price

point.”

It seems to us that Cisco aims to operate in a very similar manner to the past, the

difference being that it intends to charge for software more clearly. While this is plausible

and will no doubt increase focus, we question whether a change to market will actually

work easily. Of course as a dominant switching and routing supplier and the switching

costs of moving means that renegotiations of such sort are plausible, however, they can

equally backfire especially as large-scale customers begin to assess SDN strategies.

Upselling new software. One area where Cisco could gain traction on the area of adding

additional value through software is in the area of networking management, as

virtualization takes hold. This could present new opportunities for the company; however,

given Cisco’s desire to integrate with its own hardware could mean that its dominance in

today’s networking is unlikely to be replicated.

An installed base, and trusted position will help… One aspect of Cisco's strategy to grow

its software and services stream is simply its installed base. Today this amounts to over

$180bn of Cisco's equipment that is installed, with a highly trained base of network

administrators, we believe this in itself cannot be underestimated. It both allows the

company an initial entry into discussion for new software platforms and services, and

equally retains an inertia to moving away.

…but can Cisco really replicate IBM on software and services? A key delta between IBM

and Cisco was the ability of IBM to simultaneously dispose of commodity hardware (PCs,

retail solutions) and purchase software assets, while meaningfully cutting costs. The issue

we believe is ultimately that all of the company’s current software sales link to an actual

product sale, for example, OS will link with switching sales. The question is can Cisco

move from being a focused networking company to where the company could assemble a

software portfolio that makes it a more relevant software vendor. We note that Cisco has

over the past five years diverted significantly more of its attention to software acquisitions;

however, these still seem to relate largely to networking. In other words, we believe that a

major acquisition and/or internal move of R&D and focus may be needed. As shown in

Exhibit 110, although Cisco is aiming to become a more significant IT player, it lacks

breadth across the IT stack.

…but ranked well on core competencies. Within our survey, we posed the question to the

IT buyers to rank the companies competing for this wider IT wallet and score each vendor

on a given metric ranging from product portfolio, to roadmap, but also the quality of the

18 September 2013


salesforce and the cloud strategy. Note that our survey was aimed not at product-level

network buyers, but more at IT decision makers; as such, it gives a decent reflection of as

to how each player is seen more broadly. Interestingly, Cisco scored highly and achieved

the most #1 ranked responses and held a top-three position in every category. The

company scored notably well on product portfolio management and distribution. (See

Exhibit 109.) Cisco was followed closely by Microsoft and IBM, with the later scoring well

on roadmap sales and management. Companies that scored poorly here not surprisingly

were HP, Dell, and Huawei.

Exhibit 109: Cisco, IBM, and MSFT Compete Well in Terms of Core Competencies Please rate the following IT vendors across the attributes below, using the following scale: 1 = Very poor performance 5 = Excellent performance


1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

83

Exhibit 110: Can Cisco Compete in the IT Stack? Many Gaps versus an IBM and Oracle in millions, unless otherwise stated

Oracle IBM HP EMC Cisco SAP Microsoft Juniper Dell NetApp

E, M, T-Series/EX,

QFX-Series/Qfabric

7%

Operating system: $29bn

Management: $73bn

Applications: $168bn

Application dev. & deployment: $82bn

Middleware: $19bn

Database: $28bn

EqualLogic/

Compellent

7%

FAS/E-Series

11%

Virtualization: $3bn

Servers: $53bn

Networking: $37bn

Storage: $53bn

StorageTek/Sun

Storage

1%

IBM

DS/XIV/Storw ize

13%

3PAR/EVA/MSA

9%

VMAX/VNX/Isilon

34%

Partner/VCE/FlexPod

Pow erEdge

15%FlexPod/Partner

Partner/

IBM System

Netw ork

1%

ProCurve/3Com

7%VCE

Nexus/Catalyst/

Insieme/ASR/CRS

55%

Force10/Partner

1%FlexPod/Partner

OracleVM

1%

Pow erVM

7%

VSE

2%

VMw are

85%Partner/VCE/FlexPod

x86/SPARC/

Exadata/Exalogic

5%

System z/p/x

30%

ProLiant/Integrity

27%

VCE/Lenovo/

Partner

UCS/VCE/FlexPod

3%

Window s

79%

Hyper-V

1%FlexPod/Partner

Solaris/Linux

1%

AIX/zOS

6%

HP-UX Non

Stop/OpenVMS

3%

VMw are

Oracle

DB/TimesTen

MySQL

45%

DB2

18%

Vertica/Partner

0%

Greenplum/

Partner

0%

Sybase

5%

SQL Server

20%

NetWeaver

1%

.NET

5%

Java development

tools

22%

Rational/ILOG

20%

Quality/

Performance

Center

1%

Pivotal/Spring

Source

0%

Pow er Builder

6%

Fusion

Middlew are

16%

WebSphere

33%

VMw are/Pivotal/

Partner

0%

Backup Advisor/

RecoverPoint/

Control Center

5%

Partner/VCE/FlexPod

Visual Studio

11%

System Center

35%

Quest

1%

On Command

2%

E-business/

Fusion applications

6%

Lotus

Domino/Notes

3%

Webex/Partner

1%

ERP/

SuccessFactors

7%

Microsoft

Exchange/Lync/

Dynamics

14%

Tivoli 10%

OpenView /

Insight Manager

5%


18 September 2013


Mean Reversion on Margins

In the neoclassical theory of the firm excess economic returns in competitive markets will

be competed away. As previously mentioned there are reasons (reliability, innovation and

technology) why networking margins in segments such as switching and routing have

sustained a material premium to margins in other areas of IT and even communication

equipment. We would note however preliminary inspection would suggest that gross

margin in networking at least for Juniper and Cisco have been on a long term decline for

some time.

Cisco core GM has been sliding. A glance at Exhibit 112, shows Cisco gross margins over

time. Here, we present the corporate gross margin and the adjusted GM, the later has

been adjusted for the impact of major acquisitions (such as Scientific Atlanta, Tandberg

and the impact of UCS revenue growth that we believe is meaningfully GM dilutive as

discussed previously. This is a simplistic analysis and we acknowledge that we have not

adjusted for every acquisition, which given the software bias would actually be implying

that underlying margin declines have been more severe. What is clear that over any

prolonged period of time GM’s have been under pressure. Indeed we note that whereas

once Cisco gross margins were ~74%, today they are 63%.

Exhibit 111: Cisco GM's Have Been Declining over the Long Term in millions, unless otherwise stated

50.0%

55.0%

60.0%

65.0%

70.0%

75.0%

FY00 FY01 FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 FY13

GM % GM Cisco (adjusted)

Apart from the GM recovery FY01 to 03, Cisco GM have been on a

slow and steady GM trajectory on an adjusted basis (ex impact of

major dilutive events) down by 800 bps in 10 years


A decent correlation with time. In Exhibit 112 and Exhibit 113 we show the gross margins

for Cisco and Juniper over time, while each only includes approximately ten data points,

what is interesting, is that the R-squared is decent, in other words over time networking

gross margins have been shown to erode at a rate of about 20bps per year.

18 September 2013


Exhibit 112: A Tight and Declining Relationship Between

Cisco GM and Time… in millions, unless otherwise stated

Exhibit 113: ...and for Juniper


y = -0.0052x + 0.7236

R² = 0.7233

60.0%

62.0%

64.0%

66.0%

68.0%

70.0%

72.0%

74.0%

76.0%

GM reported GM Cisco (adjusted)


y = -0.0056x + 0.7049

R² = 0.8522

60.0%

62.0%

64.0%

66.0%

68.0%

70.0%

72.0%

GM Juniper

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

Cisco seems to be guiding for this. At the company’s analyst meeting in 2012, Cisco

published not only its revenue and gross profit breakdown, but also its targeted levels for

the next three to five years. What is interesting is that from this we can measure the

company's expectation on gross margins by product line and based upon our estimates,

margins in switching and routing may come down even further from current levels. (See

Exhibit 114.)

Exhibit 114: Cisco Has Implied that GM's in Switching and Routing May Decline in millions, unless otherwise stated

Gross Margin

FY16 Target

FY12 FY13 Low Mid High

Switching 71.5% 70.8% 66.4% 67.0% 67.5%

Routing 61.2% 61.3% 60.0% 60.5% 61.0%

Service Provider Video 30.0% 42.0% 40.7% 41.0% 41.3%

Datacenter 26.2% 31.0% 27.9% 28.1% 28.3%

Wireless 69.4% 71.0% 64.8% 65.3% 65.9%

Security 71.3% 71.3% 77.1% 77.7% 78.3%

Collaboration 61.1% 61.1% 63.0% 63.5% 64.0%

Other Products 54.9% 28.0% 54.2% 54.7% 55.1%

Services 66.7% 67.5% 66.6% 67.2% 67.7%

Total Revenue 62.4% 62.5% 61.0% 61.5% 62.0%


Note we have estimated these based upon revenue and GP from analyst day 2012.

SDN Drives a Major Headwind on Profit Dollars

While Cisco and Juniper have successfully dealt with various forms of competition over at

least the past 10 years, the competitive response of the company cannot be

underestimated. As we previously discussed extensively, the onset of SDN does present

an opportunity and a need for a generational shift in how networks within datacenters are

constructed. The issue is to what extent this evolution can disrupt Cisco's operating

structure. We believe the net impact over the long term would result in a gradual

commoditization of their business, even if the timing and extent becomes the "very" long

term:

The example of Cumulus. We believe there are host of Silicon Valley start-ups that are

eager to change the economics of the networking industry with the onset of SDN, including

Cumulus Networks.

18 September 2013


Cumulus plans to replicate for the networking industry what Linux did for the server

industry when it helped break up the stranglehold that vertically integrated computing

companies like Sun Microsystems had before x86–based servers became mainstream.

Cumulus is supporting Linux for multiple reasons. First and foremost it has a featured

networking stack, with many networking companies either already using it for

implementation, or even before this, by building it into their proprietary platforms. Second

there is a material variety of vendor-agnostic tools for managing and operating Linux

platforms available. By combining controllers from a variety of vendors, Cumulus'

operating system allows a feature rich commodity hardware-based switch.

The Cumulus model. At its basic level the company aims to alter networking so that

network devices run the same way that servers do, meaning using the same kind of tools

and operational platforms for deployment and management. The idea is that this will be

particularly attractive to companies whose data centers are constantly expanding.

Supply chain and ecosystem. The company has built a supply chain of original design

manufacturers (ODMs), distributors, value-added resellers (VARs) and system integrators

to build and deliver bare-metal switches with Cumulus' software on-board, including Acton

and Quanta and silicon from Broadcom.

Exhibit 115: Cumulus Ecosystem

Technology Partners ODM Silicon/Components Distributors Customers

Open Compute Project Accton Broadcom Penguin Computing Dreamhost

AGEMA Quanta Finisar Synnex Fastly

CF Engine UNIXSurplus

Cloudscaling

Metacloud

Opscode

Piston Cloud

PLUMgrid

Puppet Labs

Quanta

10Gtek

VMware


A change in cost structure. Given the nascence of software defined networking as

technology, while the debate will grow around the strength of one vendor's offering versus

another's, it is possible that Cisco and/or Juniper will embrace SDN successfully. The

question for investors is whether they will be able to extract the same level of economics

from the networking budgets. In Exhibit 116 we show a simplistic analysis of the impact of

SDN in economic terms. We have compared the total cost of ownership (ex-labor,

operating costs) for basic Cisco switches for the datacenter (i.e. Nexus 3064), along with

the SMARTnet services and software contracts that network administrators would usually

sell. The alternative in a “SDN” based environment would be to buy Cumulus Linux

software subscription and top of rack bare metal switch. Our analysis as shown in Exhibit

116 highlights these main issues:

The TCO is dramatically lower. Even assuming that the Cumulus approach has a shorter

switch life of three years on a TCO basis, the cost is some 88% lower. The saving comes

from effectively unbundling the solution between hardware and software.

The gross profit dollar difference is 74% lower. An additional issue becomes whereas

Cisco will deliver a gross margins of 60% and 67% on its products and services in fiscal

2014, in the Cumulus scenario we believe the gross margin is lower in aggregate at 56%.

Possibly more importantly the gross profits dollars are 74% lower. The point here is that

even if today’s networking vendor was to embrace an open SDN strategy, the gross

margin percentage and gross profit dollars will be significantly lower.

18 September 2013


Can Cisco do the upsell? Seems unlikely. We believe Cisco and others understand this

risk, even if they continue to highlight the alternative advantages of integrated ASICs,

software and services. Indeed we believe the push into services or new software is

partially to offset this decline. Nevertheless, in order Cisco to be able to be gross profit

neutral versus today's economics the customer would either need to spend additional

$8,000 or 3.3x versus the alternative from the Cumulus option or this much value in

Cisco’s wider offering and pay a premium. While we do not dispute the value of Cisco’s

installed base, trained administrators, new services, and new software they offer, this level

of difference would certainly lead us to conclude that the headwind on gross profit dollars

is significant. Moreover, Cisco’s push into expanded software offerings comes at a time

when Linux-based platforms are set to growth their application, developer and services

ecosystems.

The above analysis is simplistic in many ways. We acknowledge that the above analysis is

simplistic in many ways as it excludes the costs of labor, power, cooling, that all have an

impact on the relative TCO. Additionally while hard to place a numerical value on the risk

of using a new supplier, the inertia of Cisco network administrators cannot be

underestimated. We do believe however that the relative magnitude in the economics of

today versus tomorrow present a headwind.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

88

Exhibit 116: SDN—Threatening the Last Bastion of Proprietary IT Profits

Cisco Proposal USD

Cisco Nexus 3064 list price 48,000

Cisco Nexus 3064 realised price 20,000

License 4500

Maintence 6,000


TCO 54,500

TCO per year 10,900

Cisco Switching & Services GM 70%

Cisco GP 7,630

Cumulus Proposal

Top of rack bare metal switch 6,000

Cumulus Annual Subscription 1,500


TCO 10,500

TCO per year 3,500

Cost Saving -68%

Hardware GM 30%

Hardware GP 600

Software GM 90%

Software GP 1,350

Total GP 1,950

Blended GM 56%

Difference -74%

Cisco Value add needed to be GP neutral 5,680

Cisco Upsell to be GP neutral 8,114

Multiple of Cumulus Propasal 3.32

1) A look at the economics of SDN. Perhaps the most significant issue for the networking

segment is what happens to economics with the onset of SDN. Even if traditional

networking vendors shift their portfolio, strategies and software to a SDN defined

approach can today’s economics can be replicated? Our analysis compares on a

simplistic TCO basis the relative costs savings to the customer and gross profit headwind

to networking vendors is around 70%

2) Cisco basis Nexus 3064 DC switch. We base our analysis on a lower end Cisco Nexus

3064 datacenter switch with a list price of $48,000. However we believe the price to most

enterprises would be more than 50% lower. Additionally Cisco will license the OS and

charge maintenance, assuming a standard 5 years switch life the TCO is $54.5k or

$10.9k per annum.

3) Cumulus Proposal, 70% cheaper per year. With Cumulus the customer buys a basic

commodity switch and they charge an annual subscription of $1,500, bringing the TCO to

$10,500, $3,500 per annum. Even assuming a lower life for switch the TCO is 80% lower.

4) Gross Profit Dollars at risk. In the Cisco proposal the GP dollars available (assuming a

70% GM) are $7.6k, in the cumulus proposal (assuming standard hardware commodity

GMs) the GP is $1950m some 74% lower. In other words even if networking vendors

offer this, it significant shrinks the profits available.

5) Can Cisco achieve the upsell? We believe that Cisco is aware to some degree of this

and will attempt to upsell services, or argue for higher value add (sales-force, support,

legacy, reliability), however the needed upsell needs to worth 3.3x the Cumulus proposal,

this seems like a stretch.

6) A simplistic analysis, but…This analysis ignores other costs like labor, cooling, power,

and does not take into account the value of support, alternative features or

platform/portfolio but does highlight the disruptive economics of SDN, and the economic

surplus being earned in networking. Moreover, it does not take into account the full

solution sale. For example, with Nexus 3000-series switches sold, Cisco would bundle a

Nexus 7000 router, which has carries more upgrades and support over time, which would

support margin and allow Cisco to get more aggressive on the Nexus 3000 switches.

Even with this discounting, SDN solutions would be less expensive with a Nexus 7500

selling for $200-250k versus $100-150k for an SDN alternative.


18 September 2013


ASIC Strategy a Differentiator in an SDN World?

Cisco leverages a two silicon strategy that by and large can offer some offset on its margin

structure over time. The company leverages its custom ASICs with the aim to deliver

strategic and sustainable differentiation, schedule autonomy for product releases, and

investment protection. The company uses commodity components where needed to

facilitate time to market and opportunistic products. Here, we believe Cisco’s scale largely

comes in to its own, as it is able to engage a wide ecosystem of partners for its ASIC

strategy, in a manner that many of its rival peers cannot.

Exhibit 117: Cisco Silicon Ecosystem Is Comprehensive

Services

offered:

Industry

Leaders:

Foundries ASIC partners EDA suppliers IP suppliersMerchant Si

suppliers

ASIC and silicon

manufacturer

Cisco’s partners for

custom Silicon

development;

source IP & supply

design services

Manage foundry

relationships &

subcontractors

Provide front end

design tools (rtl

design simulation) and

back end design tools

(physical design

timing) for custom

ASIC development

Supply strategic

IP to Cisco (when

not developed

internally)

Cisco licenses

directly or via ASIC

partners

Cisco’s suppliers

for commercial,

off-the-shelf

silicon


Pricing pressure and competitive risks. By our measurement, switching accounts for 34%

of gross profits and routing accounts for 17% of gross profits; as such, while mix shifts can

have some impact, disproportionately the most significant is whether the GM levels in such

segments are sustainable. Indeed we note that, as shown in Exhibit 118, according the

Cisco’s own disclosures, by far the largest mix effect that seems to weigh on gross

margins has been at the product level, i.e. the pressure from “sales, discounts, rebates,

and product pricing”. Looking at the average of the past five years, this seems to be an

ongoing level of pressure at the rate of 2.50% that Cisco seeks to offset from other levers.

Exhibit 118: GM Drivers—Product Mix and Pricing, Offset by Manufacturing Cost Downs in millions, unless otherwise stated

PRODUCT GROSS MARGIN BRIDGE FY09 FY10 FY11 FY12

Previous Fiscal Year (GAAP) 64.8% 64.0% 64.2% 60.5%

Discounts, rebates, and product pricing -2.1% -2.2% -2.9% -2.8%

Mix of products sold -0.4% 0.1% -1.6% -0.6%

Restructuring and other charges 0.0% 0.0% -0.4% 0.5%

Amor. of purchased intang. and SBC 0.0% 0.0% -0.6% 0.3%

Acquisition fair value adjustment to inv. NA NA NA NA

Shipment volume, net of certain costs -0.4% 0.6% 0.4% 0.5%

Overall manufacturing costs 2.1% 1.7% 1.4% 1.7%

Current Fiscal Year (GAAP) 64.0% 64.2% 60.5% 60.1%


ARM-based networking a threat long term? Cisco in our view heavily utilizes its own

ASICs and the company’s dominant share has shown that this strategy in turn has proved

effective. However, we would highlight a building threat that, we believe, can come from

18 September 2013


ARM-based Networking Chips that in turn could create a new alternative of networking

suppliers based on lower power, lower cost chips. Interestingly, ARM targets a 20% share

in routing and networking, which we believe in aggregate places some long-term risk on

Cisco.

Exhibit 119: ARM Targeting a 20% Share in Switching and Routing LT in millions, unless otherwise stated

2017 Enterprise Networking Chip TAM (m) Chip Value ($bn) Target Penetration

Base station equipment 65 $3.5 60%

Carrier Infrastructure 75 $3.0 <5%

Enterprise Access Points 270 $2.0 50%

L2/L3 Switching 150 $3.0 20%

Routing 80 $2.5 20%

Other 100 $2.5 20%

Total 700 $16.5


Huawei a Threat Not to Be Underestimated

Although Huawei’s Enterprise Business group was only set up in 2011, it achieved $1.5bn

in revenues purely from opportunistic projects, for example as an extension of its existing

carrier and consumer businesses. In 2012, revenues grew some 26% to $1.9bn. While the

initial targets of some $15bn by 2017 have proved quite lofty, we equally believe it is a

threat and not be underestimated for several reasons.

A shift in organization and focus. Over the past two years, Huawei has emphasized the

growing importance of Enterprise, supported by the belief that Carrier and Consumer

markets will eventually cease to be distinct; and together with Enterprise it is expected to

form an ecosystem in which complex problems are addressed through an integrated

solution. For the Enterprise segment, although the company has not provided any specific

targets for 2012, Huawei is hoping for sales here to grow faster in 2012 compared to the

57% growth seen in 2011. In addition, in the long term, it expects the Enterprise segment

to account for as high as $10bn of revenue by 2017, with the specific goal being to take

share from traditional IT hardware companies.

18 September 2013


Figure 120: Huawei Offers a Range of Products and Services Focused at Enterprise Aimed at Select Verticals Brief description of Huawei’s Enterprise product and services portfolio


R&D a major force to be reckoned with. Both the IT Hardware and Telecom Equipment

sectors remain R&D intensive businesses, which is crucial for ongoing cost reductions as

well as maintaining a competitive advantage. Interestingly, Huawei’s capabilities here

remain strong, with the company continuing to add more resources to maintain its

positioning with the industry.

Figure 121: Huawei Keeping Up with Innovation vs. Competitors with Rising R&D Spend US$ in millions, unless otherwise stated

R&D (US$ mn) 2007 2008 2009 2010 2011 2012

Comm. Equipment companies

Cisco 4,487 4,739 4,363 5,090 5,184 5,364

Ericsson 4,266 4,689 3,570 4,147 4,939 4,722

Huawei 992 1,506 1,953 2,611 3,671 4,774

Alcatel-Lucent 3,675 3,543 3,299 3,314 3,293 2,997

Nokia Siemens 2,841 3,328 2,867 2,591 2,954 2,604

ZTE 422 575 846 1,049 1,316 1,401

Juniper 584 684 682 837 926 992

Tellabs 343 305 269 300 324 239

Sub-total 17,611 19,369 17,849 19,939 22,606 23,092

IT Hardware companies

IBM 6,266 6,426 5,943 6,152 6,345 6,322

Hewlett-Packard 3,611 3,543 2,819 2,959 3,254 3,399

EMC 1,408 1,546 1,400 1,606 1,834 2,225

Dell 610 663 617 653 856 1,072

NetApp 382 442 485 568 643 717

Sub-total 12,276 12,621 11,264 11,938 12,932 13,735


Also a significant number of staff to be reckoned with. As shown in Exhibit 122, at the end

of 2012, Huawei employed some 140K employees, of which ~70K or ~50% were in R&D.

This is significantly higher than 34% at ZTE (c22K R&D staff), 27% at Nokia (27K R&D

18 September 2013


engineers including Nokia Siemens), and 21% at Ericsson (c24K R&D employees). In

addition, the R&D cost per R&D employee at Huawei is $61K versus Ericsson at $196K

and Nokia at $205K, as shown in Exhibit 123, with only ZTE operating at a lower cost (at

around $47K). In addition to the higher level of R&D headcount, we believe this lower cost

per employee in R&D continues to give Huawei a material cost advantage versus peers.

This is aided by the company having a well distributed R&D base. Although in the past,

Huawei had a local R&D bias, now the company has R&D operations throughout the world

with centers in India, the United States, Germany, Russia, and Sweden.

Exhibit 122: With 70K R&D Staff, Huawei Remains a

Force…

Exhibit 123: …Especially with the Cost Advantage in R&D

0

10

20

30

40

50

60

70

80

Tellabs NetApp Juniper EMC Cisco Ericsson Alcatel-Lucent

ZTE Nokia (inclNSN)

Huawei

R&

D e

mp

loye

e c

ou

nt

('0

00

)

2007 2008 2009 2010 2011 2012

47

68

140159

170

196205

227 232243

0

50

100

150

200

250

300

ZTE Huawei Alcatel-Lucent EMC NetApp Ericsson Nokia Tellabs Cisco Juniper

R&

D c

ost

pe

r R

&D

em

plo

ye

e (

US

$ '0

00

) in

20

12

Source: Company data, Credit Suisse research. Source: Company data, Credit Suisse research.

Leveraging Huawei’s broader strengths for growth within the Enterprise. To support its

growth, the Enterprise business is also looking to tap into the company’s broader portfolio

of nearly 60K patent applications (including patent applications in China, patents under

patent cooperation treaties and patent applications overseas), which Huawei continues to

focus on expanding. With regard to geographic presence, the bulk of its Enterprise

Business today comes from Asia – specifically within China, somewhere in the range of

30-50%. The remainder of revenue comes from Africa and Europe, with a small portion

from Latin America and almost zero revenue from North America. Within these markets,

the primary verticals are government, education, and infrastructure (transportation).

Interestingly, for its Enterprise business, Huawei has adopted the same “outside in”

strategy as the broader company, focusing first on penetrating relatively untapped markets

(rural China, Eastern Europe, and Canada).

Margin structure conducive to Huawei’s strategy on Enterprise side. Compared to some of

the major players in the infrastructure space, Huawei commands gross margins in the

range of 37-40% and operating margins of around 10% (as per Huawei’s 2012 annual

report). This compares to Cisco and IBM’s gross/operating margins, which range around

62%/27% and 50%/22%, respectively. While Huawei’s margins are considerably lower

than those of its IT/Enterprise competitors, it is worth noting that despite a lower gross

margin profile, Huawei continues to be an operationally well-run organization. A close

parallel has been drawn between Huawei’s approach to enterprise and its history of

margin deflation tactics. Just as Huawei has built products at lower cost in the wireless

infrastructure and smartphone markets, it intends to compete in enterprise by building

channel programs that, from the outset, allow for better margins – a threatening strategy to

competitors with higher margin structures firmly in place. The company in the past has

noted that there has been significant consolidation and only a handful of new entrants into

the IT market over the last decade. It increasingly believes that by offering a broad

portfolio it can significantly shake up the competitive dynamics.

18 September 2013


Exhibit 124: Gross Margins Conducive to Huawei

Strategy...

Exhibit 125: ... as Are Operating Margins

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

Source: Company data - 2012 gross margins Source: Company data - 2012 operating margins

What impact could Huawei have? Huawei has rapidly increased share in its Service

provider routing business over the past few years growing from 5% in 2007 to 15% by

2012. The experience of Huawei in the wireless infrastructure market and the above scale,

size, R&D dynamics means that the company cannot be underestimated. On the other

hand the incumbency within enterprise switching and installed base means there will be

some challenges. Not surprisingly the company has seen traction in the service provider

segment and especially in the area of routing, where it has a share of 18% in the EDGE

market as compared to 6% in 2007. Importantly, this segment is expected to grow at a 5%

CAGR through 2017. Huawei also has a 10% share in Core routing versus 3% in 2007,

although this business is only expected to grow 1% through 2017.

We do believe the company has the potential to be disruptive given a lack of installed base

even within switching. We note the recent launch of the Agile switch which is designed

with its own Ethernet network processor. The company claims its product will offer better

performance over rival switches built with application-specific integrated circuit (ASIC)

chips, but at a still affordable low price. We equally acknowledge that there is a cap to how

far their share will go in the medium term, given security pressures especially within the

United States.

There will be some cap as to how far Huawei will go. While we fundamentally believe that

Huawei has the potential to disrupt elements of the networking market starting with its

traditional strengths in service provider segments, we believe there is a cap as to how far it

will go within enterprise. We believe indeed that one difference is simply that North

America represents some 33% of the service provider segment, 42% of the enterprise

segment, and 25% of telecom capex. Here we acknowledge that political concerns could

be an issue over time.

18 September 2013


Exhibit 126: North America is a Sizable Proportion of Networking and Telco Capex US$ in millions, unless otherwise stated

4,006 8,313 14,372

81,737

2,666

5,987

10,239

86,280

3,936 4,687 11,020

118,789

698 782 1,852 30,705

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Service Provider

Routing

Enterprise Switching Networking Overall Telco Capex

Perc

enta

ge S

hare

by

regio

n

North America EMEA Asia Pacific CALA

Source: Company data, Credit Suisse estimates, Infonetics – 2012 Spending.

Cost Management Has Been Strong of Late

Whether it be due to the potential business mix shifts or secular pressure on gross

margins, what is clear is that Cisco is keeping a vigilant eye on its cost base. Indeed

based upon our analysis, we see scope for opex to remain within check and for this to

perhaps even prove a moderate offset versus gross margin pressures described.

Headcount managing down again. Since Cisco’s more focused strategy presented in 2H

FY12, Cisco management's approach has been somewhat more disciplined than the past.

There have been two main restructurings in the past five years, with a major employee cut

of 8,000 as shown implemented in late FY11. The company has recently decided to cut by

a further 4,000 employees, around 5% of its work force. It does appear that restructuring

and continuous cost monitoring will be a key part of the Cisco business model going

forward, to make room for movement of resources and the incremental investments that

need to be made.

18 September 2013


Exhibit 127: Headcount Post Increases (Due to Acquisition) Looking to Cut by Another

4,000

50,000

55,000

60,000

65,000

70,000

75,000

80,000

Head

co

un

t ('

000)


Revenue per head. A glance at Exhibit 128 shows that the company is keen on expanding

revenue per head and this seems to be a metric to which the business is almost

benchmarked against, there was a step down to $650,000 per employee which seems to

have triggered the recent cut. We believe this has been more about the impact of the NDS

acquisition and now across the organization management are targeting new cost savings.

Exhibit 128: Revenue Per Head, Keeping It in a Tight Range

500

550

600

650

700

750

Re

ve

nu

e p

er

em

plo

ye

e (

'000

US

D)


Opex to sales discipline. Cisco targets an operating margin of 28% and an opex to sales

ratio of ~34%. We would note that the recent execution around this has been strong.

18 September 2013


Exhibit 129: Opex to Sales Already at Long-Term Targets

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

OP

EX

to

sale

s (

trailin

g)


Is there more to come…we think probably unlikely. It is of course possible that Cisco could

offset certain gross margin and competitive pressures with incremental restructuring and

efficiency gains. We believe that, while this cannot be ruled out and the company is

focused on managing costs closely, it is subject to competitive factors and needs to

maintain a certain amount of resources and headcount. Here we note that on a revenue-

per-employee perspective, Cisco operates at close to $650-700k per employee, which is

already 90% above its peer group. While its superior scale and position in networking

justifies this, it is equally unlikely that the opex to sales base can move down sizably.

Exhibit 130: Cisco Revenue Per Employee Is Already 90% above Peers

-

100,000

200,000

300,000

400,000

500,000

600,000

700,000

800,000

ZTE

Hu

awe

i

IBM

Alc

ate

l-Lu

cen

t

NSN

Eric

sso

n

EMC

Hew

lett

-Pac

kard

Tella

bs

Jun

ipe

r

Net

Ap

p

De

ll

Cis

co S

yste

ms

20

12

Re

ven

ue

pe

r e

mp

loye

e $


18 September 2013


Portfolio Refresh Helping Share Switching Portfolio Refresh

We believe a key driver around the recent market share turnaround for Cisco, especially in

the switching market, has been a fairly comprehensive portfolio refresh. (See Exhibit 132.)

Going back over the past 18 months, there has been a refresh across its modular and

fixed switching portfolio that we believe has been key to extending its competitive

advantage:

Exhibit 131: Cisco Market Share—Steady within Enterprise Switching and Routing

20%

30%

40%

50%

60%

70%

80%

90%

Cis

co

Mark

et

Sh

are

Enterprise switching Enterprise routing Service provider switching

Service provider routing Service provider edge routing Service provider core routing


Datacenter portfolio. The company now has a range of switches aimed at optimizing

performance for the datacenter built around the Nexus portfolio of switches that span from

virtual switching to fixed and modular formats. The Cisco Nexus 7000 Series Switches

offers Cisco's highest switching capacity; up to 1.3 terabits per slot, 83+ terabits per

chassis and provides 1,10, 40, and 100 Gigabit Ethernet scalability.

Backbone and access portfolio. Aimed more at the Enterprise and campus markets the

company has introduced the Cisco Catalyst 6800 Series Switches are programmable

campus backbone switches optimized for 10/40/100 Gigabit Ethernet services. These

switches offer converged wired, wireless, and VPN security, and exceptional investment

protection with their Catalyst 6500 DNA. Whereas the catalyst 2960 is aimed at branch

offices or small- and medium-sized businesses, we would note such entry-level enterprise-

class Fast Ethernet switches support basic services.


http://www.cisco.com/en/US/products/hw/switches/ps708/index.html

18 September 2013


Exhibit 132: Cisco Has Refreshed the Switching Portfolio Meaningfully of Late

2012 2013

UCS Central

September

Cisco Catalyst

3850

January

MDS 9710 Multilayer Director,

UCS, and NVIDIA GPU

May

Cisco Catalyst

6800

June

Nexus 7700

June

Nexus 3548

November

Nexus 6001, 6004

and 1000V

February

Cisco Catalyst

2960-X

June

Supervisor 8E for

Cisco Catalyst 4500-E

June


Forward looking at the Insieme spin-in for SDN. Cisco’s initial response to the

opportunity/risk presented by Software Defined Networking (SDN) appears to be in the

form of Insieme Networks, a captive private company started by three former key Cisco

enterprise switching executives that Cisco has the right to acquire or 'spin-in' for $750

million. While initial product strategy and details remain forthcoming, we believe Cisco has

essentially outsourced to Insieme the development of its SDN response. We believe the

reason for this is to avoid the standard innovator's dilemma as the focus of Insieme will be

to become Cisco’s next-next generation switching platform, with a focus on 100Gbps

Ethernet datacenter switching, that goes beyond traditional enterprise switching to add

SDN and storage to Cisco’s Unified Computing System converged datacenter architecture.

We believe the strategy here shows flexibility on management's part and realistic

assessment that the approach to SDN for the impact it needs to kept separate from the

company. While there is always a risk that this remains an expensive decision, we believe

the approach with Nuova Systems, which brought to Cisco the Nexus product family is an

expensive decision; however, at 5–10x of what we estimate it would cost Cisco to develop

a solution internally.

Routing. Cisco holds a 45% share in the $15bn routing market, followed by Juniper,

Alcatel-Lucent, and Huawei. Given increased levels of competition, the company’s share

continues to erode gradually from 64% in 2005. We highlight the underlying portfolio.

Service provider. The majority of Cisco’s routing business is geared toward the service

provider business and within that, edge routing.

■ Edge: Cisco’s edge service provider routers include the XR 12000, 7300/7600/10000

series and ASR 1000/9000. The ASR 1000 has an overlap with high-end enterprise

routing. The ASR 1000 is targeted at providing secure, high performance

software-enabled services while the 9000 is geared for more intensive video and

carrier residential, mobile and business services. Meanwhile, the Cisco XR 12000

provides for secure virtualization and is able to prioritize voice, video, and services.

■ Core: The main line of Cisco core routing is the Carrier Routing System or CRS. While

the XR 12000 can be used in the core element, the CRS-1 and CRS-3 form the

backbone of Cisco’s offering. The product family is focused on high performance and

scale with a distinct focus on service providers, unlike smaller routing offerings.

Smaller variants of the portfolio offer up to one terabit of switching capacity but when

aggregated in a shelf system, this scales to over 900 terabits. The products enable

service providers to upgrade to 100GbE and blend optics for improved efficiency. CRS

18 September 2013


enables carriers to extend additional services through Layer 4-7 enablement. Cisco

believes that carriers can see a 24% lower TCO through more efficient demand

optimization and up to 50% lower TCO from enabling nLight, which improves optical

and capacity utilization.

■ Enterprise: On the enterprise side, Cisco’s strength is at the branch office and

mid-range, with less penetration in the low-end SOHO market and the high end.

Fortunately, these mid-segments comprise three quarters of the enterprise routing

business. Cisco’s mainline routers are the ASR 1000 and the Cisco 7200/7600. As

discussed above, these products are overlapping with lower-end service providers. In

the mid-range, the Cisco 3900 is an example product for the branch. This product can

deliver media and services, such as Telepresence. The 1900/2900 at the branch office

as well, focused on cost effective service delivery, although with lower speed

performance than the 3900. On the low end and SOHO market, Cisco participates with

the Cisco Small Business Router and the 800 series. The 800 series comes with

802.11n access point functionality. For the branch offerings, advanced encryption,

intrusion prevention, content filtering, and Cisco Unified Communications integration

are standard.

18 September 2013


Capital Distributions Could Move Higher A key positive over the past few years at Cisco for investors has been the company's

willingness to engage more significantly on cash distribution. After all the company

generates consistent FCF which totaled $11.7 billion in FY13 and holds a robust balance

sheet of $50 billion. The 'religion' of returning cash to shareholders is not necessarily a

new theme within large-cap technology; however, the question going forward is whether

we believe that such cash payouts are conservative and how they compare versus peers.

We would make several points:

■ FCF generation strong; $11.7bn in 2012. We note that compared with large-cap tech

peers, Cisco generates healthy levels of annual FCF. Over time, FCF margins have

expanded to 24% in FY13 from 21% in FY11 with conversion of 118% in FY13.

■ Dividend yield exceeds peers. Cisco's dividend yield is 2.8%, above large-cap tech

peers at 2.5%.

■ Cisco pay out levels second only to Apple over three years. Given the trend of late for

enterprise companies to begin paying out cash flow to investors, we note that Cisco

has committed over the next three years to paying out some 15% of the market cap in

buybacks and dividends, which puts it just behind Apple and narrowly ahead of an

Intel.

Overall our analysis implies that Cisco's current levels of cash return are relatively in-line

with peers.

Could Cisco Do More?

One interesting issue is, that based on our estimates, Cisco’s cash pile will continue to

swell absent any sizeable acquisitions. Indeed, even allowing for the onshore versus the

offshore cash dynamics, we see a situation whereby balancing with debt issuance the

company has high levels of excess cash. We note that the company has a net cash to

market cap ratio well above peers at 26%.

Exhibit 133: Net Cash to Market Cap for Large-Cap Tech—Cisco Still Hoards

-30%

-20%

-10%

0%

10%

20%

30%

40%

Net

Ap

p

Ap

ple

EMC

Cis

co

Qu

alco

mm

Mic

roso

ft

Del

l

Jun

iper

Acc

entu

re

Eric

sso

n

Ora

cle

Lexm

ark

Inte

l

IBM HP TI

Xer

ox

Net

cas

h a

s a

% o

f m

arke

t ca

p

Source: Company data, Factset

18 September 2013


It would be too simplistic to purely look at Cisco's $31bn of net cash and $51bn of gross

cash and reach the conclusion that the company holds an excessive amount of cash. A

more thorough analysis follows in which we look at the onshore/offshore cash generation

as well as the company's ability to access the debt market.

Current Estimates Allow for Cisco to Pay Out 50%+ of Ongoing Cash Generated: Per

current estimates, Cisco pays out ~50% of free cash flow generated in the form of

dividends and buybacks. With this cash distribution program, we believe Cisco could end

FY15 with $62.6bn in gross cash, with $59.8bn being offshore and $2.8bn onshore and a

net cash position of $43.6bn. Looking at these estimates, we believe that $59.8bn is an

excessive amount of cash to have offshore, while $2.8bn onshore might be somewhat

light. On account of the strong balance sheet, we believe the company has a significant

ability to have an expanded buyback program by taking on additional debt.

Debt Issuance/Buyback Scenario: Over the next two years, we assume Cisco has the

capacity to issue up to $25bn in with $12.5bn in each of FY14/FY15, which would bring

Cisco's leverage ratio to ~2.5x exiting FY15. In addition, we assume $4.6bn in cash

remains onshore in order to fund operations. In this scenario, Cisco is left with ~$22bn in

excess cash in FY14-15 to fund additional buybacks. Allowing for a conservative interest

rate (4.8%), we would have EPS estimates of $2.19 and $2.39 for FY14 and FY15, 4%

and 12% higher, respectively, than our current estimates. Based off the company’s two-

year forward PE multiple over the past five years, the target price range is from $21-25.

Significant Liquidity after Delivering Another $25bn: Even after levering up and

increasing cash distribution through buybacks, the company is still left with a pristine

balance sheet, with gross cash of $64.4bn ($4.6bn onshore/$59.8bn offshore) and net

cash of $23.2bn exiting FY15. We believe that this is ample liquidity for whichever

scenario results strategically.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

10

2

Exhibit 134: Cisco Screens Modestly above Peers on FCF Generation and Pay Outs. Could It Do More?

-5%

0%

5%

10%

15%

20%

25%

30%

35%

40%O

racl

e

Mic

roso

ft

Qua

lcom

m

EMC TI

Cisc

o

App

le

Net

App

IBM

(IB

M b

asis

)

Inte

l

Juni

per

IBM

(CS

basi

s)

Acc

entu

re

Xero

x

HP

Eric

sson

Lexm

ark

Del

l

FCF

Mar

gin

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

App

le

Net

App H

P

Ora

cle

Inte

l

Cisc

o

Mic

roso

ft TI

EMC

IBM

Xero

x

Juni

per

Lexm

ark

Eric

sson Del

l

Acc

entu

re

Qua

lcom

m

% o

f mar

ket

cap

0.0x

5.0x

10.0x

15.0x

20.0x

25.0x

Net

App

Jun

iper

IBM

(C

S b

asis

)

Eric

sso

n

Qu

alco

mm

Inte

l TI

Acc

entu

re

IBM

(IB

M b

asis

)

Cis

co

Ora

cle

Lexm

ark

Mic

roso

ft

App

le

EMC

Xero

x

HP

Pric

e to

FC

F

0.0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

4.0%

4.5%

Inte

l

Lexm

ark

Eric

sson TI

Mic

roso

ft

Cis

co

App

le HP

Xero

x

Del

l

Acc

entu

re

Qua

lco

mm

IBM

EMC

Ora

cle

Net

App

Jun

iper

Div

iden

d Yi

eld

CSCO above on FCF margins …a slightly better dividend yield…

… trades in line Market Cap/FCF …returning cash at a higher rate than peers

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

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

Source: Company data, Credit Suisse estimates, FactSet 2014 FCF estimates, Credit Suisse research estimates

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

10

3

Exhibit 135: Cisco Has the Capacity to Return Another $22bn

Increased debt issuance/buyback scenario1. Current estimates allow for Cisco to pay out 50% of ongoing

cash generated. Per current estimates Cisco pays out ~50% of free

cash flow generated in the form of dividends and buybacks. With this

cash distribution program, we believe Cisco would end FY15 with

$62.6bn in gross cash with $59.8bn being offshore and $2.8bn

onshore and a net cash position of $43.6bn.

i. Looking at these estimates we believe $59.8bn is an excessive

amount of cash to have offshore while $2.8bn onshore might be

somewhat light. On account of the strong balance sheet we

believe the company has significant ability to have an expanded

buyback program by taking on additional debt.

2. Debt issuance/buyback scenario

i. Over the next 2 years we assume Cisco has the capacity to issue

up to $25bn in debt, we assume $12.5bn in each of FY14/FY15,

which brings its leverage ratio to ~2.5x exiting FY15.

ii. In addition we assume $4.6bn in cash remains onshore in order

to fund operations.

iii. In this scenario Cisco is left with ~$22bn in excess cash in FY14-

15 to fund additional buybacks.

iv. Allowing for a conservative interest rate (4.8%) we have an EPS

of $2.19/$2.39 for FY14/FY15, 4%/12% higher. Based off the

company’s 2 year forward PE multiple over the past 5 years the

range of price targets is from $21-$25.

v. Even after levering up and increasing cash distribution through

buybacks, the company is still left with a pristine balance sheet,

with gross cash of $64.4bn ($4.6bn onshore/$59.8bn offshore)

and net cash of $23.2bn exiting FY15.

FY14E FY15E

Jul-14 Jul-15

Pre Debt Issuance

Gross Cash 56,750 62,598

On-Shore Cash 2,924 2,793

Off-Shore Cash 53,826 59,804

Total Debt 16,211 16,211

Net debt (cash) (37,615) (43,593)

EBITDA 16,290 16,323

Net Income 11,485 11,714

Diluted Shares 5,462 5,502

EPS 2.10 2.13

Post Debt Issuance

Debt issued 12,500 12,500

Cost of new debt 4.80% 4.80%

Additional Interest Payment 600 1,200

Total Debt 28,711 41,211

Leverage (Debt/EBITDA) 1.76x 2.52x

On Shore cash needed for operations 4,607 4,607

On Shore cash post Debt issuance 15,424 17,081

Excess On Shore Cash available for buybacks 10,217 11,274

On Shore Cash post issuance and accelerated buyback 4,607 4,607

Offshore Cash 53,826 59,804

Total Debt 28,711 41,211

Net Debt (Cash) (29,722) (23,200)

Shares repurchased @ $24.00 426 470

Shares outstanding post repurchase 5,036 4,531

Net Income 11,011 10,766

EPS FY15 post buyback 2.19 2.38

% upside due to buyback 4% 12%

Implied stock price Multiple

Downside scenario | -1 Stdv multiple 8.6x 20.54

Base scenario | Average multiple 9.7x 23.02

Upside scenario | +1 Stdv multiple 10.7x 25.49


18 September 2013


Valuation—FV of $21 Per Share We focus our valuation on a three-pronged approach: (1) P/E multiple valuation

(comparing P/E with peers); (2) discounted cash flow (DCF) analysis; and (3) Credit

Suisse HOLT analysis. Our approach suggests that the company’s equity value is $114

billion to $120 billion, or $21 to $22 per share, with a blended value of $21 per share.

Exhibit 136: Cisco Average Value of $21

Approach Equity value (in $mn) Valuation Value/Share ($)

P/E 113,696 9.8x CY14 Earnings $20.91

DCF 116,457 Term. growth of 1%, WACC of 11% $21.42

Credit Suisse HOLT 119,779 R&D life of 5 years $22.03

Blended Average 116,644 $21.45


Price/Earnings Multiple Suggests FV of $21 per Share

We apply a P/E multiple of 9.8 on Cisco’s 2014 Non-GAAP EPS estimate. Given that the

company is already very mature and has a stable operating structure, it is reasonable to

use standard forward P/E valuation.

Exhibit 137: Valuing Cisco at 9.8 times 2014 P/E in millions, unless otherwise stated

P/E Valuation

Current Price (in $, per share) 24.29

Diluted Share Count 5,437

Market Cap 130,899

Net Debt -34,399

Enterprise Value 96,500

Peer Group 12.5x

CSCO CY14 EPS $2.13

2Y Fwd. P/E Multiple 9.8x

Implied stock price $20.91

Implied equity value 113,696


Apply a 9.8 CY14 P/E multiple based on comps. To analyze a fair earnings multiple for

Cisco, we have looked at the current P/E multiples for the company along with the related

peer group. Given Cisco’s growth trajectory in the networking industry and the size of

company, we form a peer group of comparable networking companies and a group of

large cap technology firms, which includes EMC, IBM, Microsoft, Oracle, and Qualcomm.

Additionally, over the past two years, Cisco has traded at an average of 9.8x consensus

two year forward earnings.

Cisco trades at a modest discount to large cap peers on 2014 P/E. On average,

comparable large cap companies trade at a P/E multiple of 12.5 on 2014 EPS. Given the

fact that Cisco has lower growth profile and part of its growth is inorganic, Cisco is valued

at a discounted multiple to fast growing networking companies such as Riverbed, F5

Networks, and Aruba.

18 September 2013


Exhibit 138: Cisco Trades at a Discount to Both Large-Cap Tech and Networking Peers P/E (x) EV/FCF(x) Mkt cap/ FCF FCF conversion FCF Margin

2013E 2014E 2013E 2014E 2013E 2014E 2013E 2013E

Large cap tech

Apple 11.7 9.6 6.8 6.7 9.9 9.6 119% 25%

EMC 14.2 12.0 7.5 6.8 9.2 8.3 154% 26%

HP 6.2 5.6 4.8 4.4 5.0 4.6 124% 8%

IBM 11.8 10.5 15.4 11.8 15.5 11.9 76% 13%

Google 20.3 17.3 17.2 13.6 20.6 16.4 99% 24%

Intel 12.7 12.0 15.8 13.9 16.4 14.4 78% 14%

Microsoft 12.2 11.4 8.4 7.7 10.8 9.9 113% 32%

Oracle 11.9 10.9 9.9 9.4 10.9 10.3 109% 37%

Qualcomm 15.2 13.7 11.7 10.9 15.3 14.3 99% 30%

TI 21.0 18.2 18.6 14.4 17.8 13.8 118% 21%Accenture 17.8 16.4 13.5 12.4 15.1 13.9 118% 11%

Mean 14.1 12.5 11.8 10.2 13.3 11.6 110% 22%

Networking/ infrastructure

Adtran 39.5 28.2 26.6 21.5 29.3 23.6 135% 8%

Alcatel-Lucent NM NM NM NM NM NM 230% -7%

Ciena 37.4 23.2 NM 28.7 NM 22.5 7% 0%

Ericsson 17.2 14.9 16.0 13.3 18.3 15.3 94% 7%

F5 Networks 19.7 17.3 14.4 13.3 15.6 14.4 126% 30%

Finisar 20.3 15.1 33.6 17.7 37.6 19.8 54% 6%

Quanta Services 18.8 15.7 34.6 18.3 36.9 19.5 51% 2%

Aruba 27.2 23.2 16.5 16.6 20.7 20.8 131% 15%

Riverbed 16.4 13.9 16.6 12.1 15.9 11.6 103% 15%

Cisco 11.9 11.4 8.2 7.8 11.1 10.6 106% 24%

Juniper 17.8 16.0 18.4 15.4 22.0 18.4 81% 11%

Mean 22.6 17.9 20.5 16.5 23.1 17.7 102% 10%

Company


Over the past two years, Cisco has traded in the range of 8-13 times two year forward P/E.

Currently the stock is trading at 11 times our CY14 EPS estimate, which is slightly above

the two-year average. If we assume a historical average multiple of 9.8, the implied target

price for Cisco is $21.

Exhibit 139: Historical P/E Suggests Value of $21/Share

7.0

8.0

9.0

10.0

11.0

12.0

13.0

Aug-11 Oct-11 Dec-11 Feb-12 Apr-12 Jun-12 Aug-12 Oct-12 Dec-12 Feb-13 Apr-13 Jun-13 Aug-13

-1 SD PE FY2 +1 SD Avg

Source: Company data, Credit Suisse estimates, Factset.

FCF generation of $13 billion - Cheap on a FCF basis. We note that compared with large-

cap tech peers, Cisco generates healthy levels of annual free cash flow, which has been

consistently expanding over time. FCF margins are 24% and conversion is a respectable

108%. Currently, Cisco trades at 10 times our price to FCF multiple, which is below large-

cap tech peers.

18 September 2013


Exhibit 140: On a Price/FCF basis Cisco Remains Relatively Cheap to Large-Cap Tech Peers

0.0x

5.0x

10.0x

15.0x

20.0x

25.0x

Net

Ap

p

Jun

iper

IBM

(C

S b

asis

)

Eric

sso

n

Qu

alco

mm

Inte

l

TI

Acc

entu

re

IBM

(IB

M b

asis

)

Cis

co

Ora

cle

Lexm

ark

Mic

roso

ft

Ap

ple

EMC

Xer

ox

HP

Pri

ce t

o F

CF

Source: Company data, FactSet, Credit Suisse estimates.

DCF Analysis Yields an Equity Value of $116bn or $21 per Share for Cisco

Our discounted cash flow analysis for Cisco suggests a fair value of $21 per share. (See

Exhibit 141.) This is based on the following assumptions.

Revenue growth to moderate going forward. For 2014 and 2015, we expect Cisco to see

revenue growth of 5.6% and 4.7%, respectively. Beyond 2016, we expect revenue growth

to moderate around to 2% before stabilizing at 1% in the long term.

EBIT margins to decline to 21% in the long term. Driven by its dominant share position, we

believe Cisco’s gross margin will decline to 62% and 61% in 2014 and 2015. This, in turn,

will keep operating margins at 28% and 27% in the next two years. In the long term, we

expect gross margins to meaningfully decline with EBIT margins eventually reaching 21%,

as the company faces increasing competition within traditional networking as well as a

growing SDN threat.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

10

7

Exhibit 141: DCF Yields a Fair Value of $21 In US$ millions, unless otherwise stated

2013E 2014E 2015E 2016E 2017E 2018E 2019E 2020E 2021E 2022E 2023E

Revenue 49,520 52,293 54,768 56,411 57,539 58,690 59,863 60,462 61,067 61,677 62,294

% change 4.8% 5.6% 4.7% 3.0% 2.0% 2.0% 2.0% 1.0% 1.0% 1.0% 1.0%

Gross Profit 30,836 32,379 33,270 33,788 34,004 34,214 34,420 34,280 34,134 33,982 33,824

Gross Margin (%) 62.3% 61.9% 60.7% 59.9% 59.1% 58.3% 57.5% 56.7% 55.9% 55.1% 54.3%

R&D 5,637 6,040 6,476 6,670 6,804 6,940 7,079 7,149 7,221 7,293 7,366

% of sales 11% 12% 12% 12% 12% 12% 12% 12% 12% 12% 12%

S&M 9,086 9,596 9,926 10,224 10,428 10,637 10,850 10,958 11,068 11,178 11,290

% of sales 18% 18% 18.1% 18% 18% 18% 18% 18% 18% 18% 18%

G&A 2,157 2,078 2,081 2,143 2,186 2,230 2,275 2,297 2,320 2,344 2,367

% of sales 4% 4% 4% 4% 4% 4% 4% 4% 4% 4% 4%

EBIT 13,957 14,665 14,786 14,751 14,585 14,407 14,217 13,875 13,525 13,167 12,801

% of sales 28% 28% 27.0% 26.1% 25.3% 24.5% 23.7% 22.9% 22.1% 21.3% 20.5%

Add: Depreciation & Amortization 2,051 1,755 1,614 1,662 1,695 1,729 ,764 1,782 1,799 1,817 1,836

% of sales 4.1% 3.4% 2.9% 2.9% 2.9% 2.9% 2.9% 2.9% 2.9% 2.9% 2.9%

EBITDA 16,008 16,420 16,400 16,413 16,281 16,137 15,981 15,657 15,325 14,985 14,636

EBITDA Margin 32.3% 31.4% 29.9% 29.1% 28.3% 27.5% 26.7% 25.9% 25.1% 24.3% 23.5%

Less: CAPEX 1,251 1,317 1,400 1,442 1,471 1,500 1,530 1,546 1,561 1,577 1,592

% sales 2.5% 2.5% 2.6% 2.6% 2.6% 2.6% 2.6% 2.6% 2.6% 2.6% 2.6%

Add/Less: Working Capital change 1,322 984 833 469 322 328 335 171 173 174 176

Less: Net Other Assets/Liabilities change (305) (150) (85) 73 76 79 82 85 88 92 95

Add: Change in deferred revenue 269 534 426 310 213 217 222 113 114 115 116

Total Cash Flow before Tax 13,399 14,003 14,008 14,385 14,276 14,104 13,919 13,638 13,294 12,941 12,580

Cash Taxes 2,920 3,109 3,147 4,037 3,991 3,943 3,890 3,797 3,701 3,603 3,503

Cash Tax Rate 28.4% 28.3% 27.4% 27.4% 27.4% 27.4% 27.4% 27.4% 27.4% 27.4% 27.4%

Free Cash Flow 10,479 10,893 10,860 10,348 10,285 10,161 10,028 9,841 9,592 9,338 9,077


18 September 2013


Exhibit 142: DCF Summary and Weighted Average Cost of Capital (WACC) Calculation for

Cisco in millions, unless otherwise stated

Summary Assumptions

PV of Free Cash Flow 61,163 WACC 10.6% Debt to Cap

Terminal Value 95,744 Beta 1.13 10.9%

PV of Terminal Value 35,038 Risk free rate 4.0%

Enterprise Value 96,201 Equity Risk Premium 6.5%

Plus: Cash & ST Investments 50,610 Cost of Equity 11.3%

On Shore 10,200 Avg cost of debt (pretax) 6.0%

Of Shore fully taxed 26,267 Avg Cost of debt (post tax) 4.3%

Tax rate 28.4%

Less: Debt 16,211

Equity Value 116,457

12-month share price target $21.42

Diluted share count 5,437


Terminal Growth Rate of 1.0% and WACC of 10.6%. For the DCF analysis, we have

assumed a terminal growth rate of 1.0% for Cisco. We estimate the weighted average cost

of capital (WACC) to be 9.8%.

Credit Suisse HOLT Implies a Fair Value of $22 per Share

CFROI® ~13% in the long term. To take a long-term perspective on valuation, we utilize

our forecasts for the next five years until 2020 (from our discounted cash flow analysis) for

the Credit Suisse HOLT valuation methodology. For 2020, we project a Credit Suisse

HOLT CFROI of 13%. (See Exhibit 144.)

Exhibit 143: HOLT Implies a FV of $22 for Cisco in millions, except per share values

Valuation Results

PV of Existing Assets 125,479

NPV of Future Investments 2,476

+ Market Value of Investments 153

Total Economic Value 128,109

+ Share Issuance 316

- Debt & Equivalents 20,457

- Minority Interests 31

Warranted Market Cap. 107,936

Shares Outstanding 5,437

Warranted Share Price (USD) 22.03


EBITDA margins to drop to 24% long term. As top-line growth for Cisco moderates toward

1%, we believe EBITDA margins will decline gradually, eventually reaching 24% in 2020.

18 September 2013


Exhibit 144: Cisco Long-Term HOLT Analysis

CFROI Results 2012 2013 2014 2015 2016 2017 2018 2019 2020

CFROI 13.57% 16.37% 15.17% 15.48% 14.92% 14.10% 13.54% 13.03% 12.61%

Transaction CFROI 10.71% 12.33% 11.56% 11.87% 11.54% 11.00% 10.65% 10.32% 10.04%

Normalized Real Growth Rate 6.84% 5.18% 4.01% 4.69% 4.16% 4.32% 3.68% 3.34% 2.82%

Real Growth Rate 4.48% -0.15% 2.77% 2.12% 1.10% 0.96% 0.54% 0.41% 0.08%

Sales Growth 6.58% 5.53% 4.56% 5.57% 3.00% 2.00% 2.00% 2.00% 1.00%

EBITDA Margins 26.68% 27.87% 29.55% 28.29% 27.44% 26.64% 25.84% 25.04% 24.24%

Asset Turns 0.51 0.53 0.53 0.54 0.54 0.54 0.54 0.53 0.53

Gross Cash Flow 16,840 19,351 19,200 20,188 20,317 20,340 20,360 20,374 20,381

Non Depreciating Assets 47,519 47,605 50,474 52,112 53,289 54,033 54,793 55,568 56,358

Gross Investment 90,197 91,452 95,742 99,246 102,135 104,958 107,413 109,777 111,831

Life 6.4 6.4 6.4 6.4 6.4 6.3 6.3 6.3 6.3

Country Specific Discount Rate 4.90%

Size Differential 0.70%

Leverage Differential 0.40%

Company Discount Rate 6.00%

Discount Rate Used 6.00%


18 September 2013


Risks: Where Could We Be Wrong? Given our negative stance on Cisco, in this section, we review the several potential risks of

our Underperform rating.

Macro spending recovery? Cisco remains, in part, a call on the macroeconomic

environment. We tend to resist recommendations based upon such views, largely because

our ratings are relative to our overall sector coverage. In isolation, in Error! Reference

source not found., we looked at past economic cycles and looked at the relative level of

revenue growth and gross margin expansion in Cisco’s core networking business. We

arrive at several major conclusions:

If Cisco can reach EPS of $2.30 in FY14… We assume that Cisco is able to see

networking revenue growth of 9% in FY14, driven by a pick-up in IT spending in general.

This is in-line with how the company has managed to recover in previous spending cycles.

Further, we assume gross margins expand to 63% driven by continued strength in

services and switching. Additionally, we believe that the company hits its opex targets.

…the stock could head to $26 versus $21. This would result in EPS of $2.30, which is

some 10% above current consensus expectations. Applying the forward P/E one standard

deviation above the previous two-year average would suggest upside potential to $26,

which is about 24% above our current target price of $21.

Prefer EMC. We do acknowledge, however, in a broad IT spending recovery that many of

the names in our universe would benefit. However, we believe that in this scenario, our

relative preference would be to own EMC, given superior revenue, growth, a cheaper

valuation, and a higher-quality position in the industry. Indeed, as shown in Error!

Reference source not found., in a recovering IT spending environment, EMC screens

better than both IBM and Cisco.

SDN development is much later than expected. A core part of our thesis is that the onset

of SDN will create a more competitive structure and that will erode the premium level of

gross margins that networking vendors currently enjoy. This will result in an aggregate

lower level of profitability for the sector. Our current view is that pilots will be launched

over the next 18 months, with full-scale production over a 2 to 5 year period. We believe

that there are good business, technology, and performance reasons for this. However, as

with any new technology, we believe that this shift can be overstated in the near term. A

longer-time horizon may allow Cisco to reposition its business and to capitalize on its

opportunity; alternatively, Cisco could pursue new start-up vendors. In the end, we remain

confident that this would shrink gross profit dollars for the company.

Balance sheet receives little credit currently, but it could be increasingly important. Cisco

has a net cash pile of $34 billion. Given the relatively stable level of free cash flow, we

believe the company could take on board debt and use this as a form of accelerated

buybacks or capital distribution. While we believe that this strategy is unlikely to happen

overnight, the company may need to reposition itself in coming years. IndeedError!

Reference source not found., we note that by having a multiple of ~2x EBITDA in terms

of leverage could drive FY15 EPS towards $2.38, which drives a fair value on the stock of

$21-25 based on a range of historical multiples.

1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

11

1

Exhibit 145: Cisco Bull-Case Scenario Gives Upside Potential to $26

Cisco FY14 Base case yoy % FY14 Bull case yoy %

Switching 15,239 3.4% 16,510 12.0%

Routing 8,522 3.6% 8,888 8.0%

SP Video 4,836 -0.3% 4,836 -0.3%

DC 2,516 21.4% 2,695 30.0%

Wireless 2,708 25.0% 2,708 25.0%

Security 1,374 2.0% 1,374 2.0%

Collaboration 3,963 0.2% 3,963 0.2%

Other 398 -40.0% 398 -40.0%

Services 11,267 6.5% 11,267 6.5%

Revenue 50,823 4.6% 52,639 8.3%

GP 31,563 3.8% 33,373 9.8%

GM % 62.1% 63.4%

OPEX 17,132 2.5% 17,634 5.5%

OPEX as % sales 33.7% 33.5%

OP 14,431 2.6% 15,739 9.1%

OPM % 28.4% 29.9%

Other income 106 106

Tax rate % 21.0% 21.0%

Net income 11,485 5.7% 12,518 15.2%

Share count 5,462 5,462

EPS $2.10 3.9% $2.29 13.3%

Multiple

-1 Stdv 9.3x 9.3x

Average 10.4x 10.4x

+1 Stdv 11.5x 11.5x

Implied stock price

Downside scenario $19.64 $21.41

Base scenario $21.87 $23.84

Upside scenario $24.11 $26.28

The upside on the cycle. Understanding that Cisco is cyclical, herewe present a bull case in which the networking segment matchprevious upturns seen in 2003 and 2009 with:

12% growth in switching

8% in routing

30% growth in data centers

GM expands to 63%: better margin mix driven by continued strengthin services (+9% yoy) and switching (+12% yoy) with higher thancorporate average GMs.

Leading to an EPS. In this scenario the EPS comes out at $2.29versus our base case of $2.10, 8% above current consensus

Upside back to $26?: Based on our bull case revenue build andmargin structures as well as a forward P/E one standard deviationabove the previous two year average would suggest upside to $26which is about 24% above our current price target of $21.

CSCO’s switching/routing correlates with GDP

-6.0%

-4.0%

-2.0%

0.0%

2.0%

4.0%

-40.0%

-30.0%

-20.0%

-10.0%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

Mar

-08

Jun

-08

Sep

-08

Dec

-08

Mar

-09

Jun

-09

Sep

-09

Dec

-09

Mar

-10

Jun

-10

Sep

-10

Dec

-10

Mar

-11

Jun

-11

Sep

-11

Dec

-11

Mar

-12

Jun

-12

Sep

-12

Dec

-12

Mar

-13

Swit

chin

g/R

ou

tin

g

CSCO routing CSCO switching Developed market GDP

Source: Company data, Credit Suisse Data


1

8 S

ep

tem

ber 2

013

Cis

co

Sy

ste

ms In

c. (C

SC

O)

11

2

Exhibit 146: EMC over CSCO & IBM: Cheaper, with Better Organic Growth and FCF

EMC IBM CSCO

Revenue CAGR 12-14E 8.3% 0.7% 5.2%

EPS CAGR 12-14E 14.1% 9.7% 4.8%

2014 P/E 11.7x 10.0x 11.1x

2014 P/FCF 8.7x 13.0x 10.2x

2014 P/Adj FCF 14.8x 19.0x 13.8x

Financial Details EMC IBM CSCO

2014E 2014E 2014E

Revenue 25,488 105,967 52,293

Gross Profit 16,787 53,068 32,379

GM 65.9% 50.1% 61.9%

Operating Income 7,000 25,964 14,665

OM 27.5% 24.5% 28.0%

Standard FCF, $mn 6,266 15,482 12,408

FCF conversion, % 168.5% 82.6% 114.8%

Adjustments:

Net Acquisitions 0 (2,336) 0

Share repurchase for option dilution offset (2,584) (1,900) (2,844)

Restructuring 0 (675) (400)

CS defined Free Cash Flow, $mn 3,682 10,571 9,164

CS defined FCF conversion, % 99.0% 56.4% 84.8%

1

EMC multiple diverges from IBM CSCOEMC has faster, higher quality growth with cheaper multiple

EMC – Superior FCF generation/conversion even on an “adjusted basis”

Source: Factset, Credit Suisse estimates

*CS defined FCF focues on FCF attributable to shareholders. As such, we penalize

companies on necessary M&A to maintain R&D and sales growth, cash consumed to offset

share dilution, and restructuring charges. For EMC specifically, we penalize on cash use to

maintain ownership of VMW stock and to offset VMW share count dilution as well.

Within the IT sector, we prefer EMC as it trades at a lower FCF multiple

vs. IBM and Cisco with higher growth and better FCF conversion.

EMC trades at a discount with superior growth. On both standard free

cash flow and Adjusted FCF, EMC is cheaper than CSCO and has higher

revenue/EPS growth rates.

EMC has highest FCF conversion even after adjustment. We penalize

companies’ buyback to offset share count dilution, acquisitions to maintain

growth, and workforce restructurings. After adjustments, EMC’s FCF

conversion still stands out.

5.0

10.0

15.0

20.0

Jan

-11

Mar

-11

May

-11

Jul-

11

Sep

-11

No

v-1

1

Jan

-12

Mar

-12

May

-12

Jul-

12

Sep

-12

No

v-1

2

Jan

-13

Mar

-13

May

-13

Jul-

13

Mu

ltip

le

IBM EMC CSCO


18 September 2013


Management Team Cisco’s management team demonstrates a broad level of experience with leaders from

firms such as Motorola, Hitachi Data Systems, and IBM.

Figure 147: Seasoned Team

Name Position CSCO Holding At Firm At Position Experience

Frank Calderoni CFO & Exec. VP <1% May-04 Feb-08 - CFO & VP at QLogic, CFO & Senior VP of Administration

+1.3mn Options at SanDisk, and Controller at IBM

John Chambers Chairman & CEO <1% Jan-91 Jan-95 - President & CEO of TGV Software, Senior VP-US

+4.2mn Options Operations at Wang Laboratories, Board of Wal-Mart

Mark Chandler Secretary & Chief

Compliance Officer

<1%

Jul-96 Oct-01 - General Counsel at Stratacom

Blair Christie Chief Marketing

Officer

<1%

+350k options

Aug-99 Jan-11 - Senior VP of Global Corporate Communications at Cisco

Wim Elfrink Chief Globalisation

Officer

Nov-97 Dec-06 - Senior VP of Customer Advocacy at NetSolve

Robert Lloyd

President of

Development &

Sales

<1% +1.8mn options

Nov-94 Oct-12 - Executive VP, Worldwide Operations, and Senior VP of

US Canada and Japan Operations at Cisco

Gary Moore President & COO <1% Oct-01 Oct-12 - President & CEO at Hitachi Data Systems, CEO at Netigy

+940k options

Pankaj Patel Chief Development

Officer

+800k options Jul-96 Aug-12 - Senior VP at Redback Networks

Randy Pond Executive VP,

Operations,

Processes, Systems

Sept-93 Aug-07 - VP of Finance, CFO and VP of Operations of Crescendo

Communications

Padmasree

Warrior

Chief Technology &

Strategy Officer

Jan-07 Jan-07 - CTO & Executive VP at Motorola


John Chambers, Chairman and Chief Executive Officer

John Chambers has served as chief executive officer since January 1995 and chairman of

the board of directors since November 2006. Chambers also served as president of Cisco.

Prior to Cisco, Chambers served as president and CEO of TGV Software and senior vice

president of U.S. Operations at Wang Laboratories. Chambers holds a JD and B.A. from

West Virginia University and an M.B.A. from Indiana University.

Frank Calderoni, Chief Financial Officer

Frank Calderoni is currently the CFO of Cisco and has held the position since February

2008. Prior to joining Cisco, Calderoni was the senior vice president and CFO of QLogic

Corporation, CFO of SanDisk, and controller at IBM. He sits on the board of directors for

Adobe Systems and Nimble Storage. Calderoni received a B.A. from Fordham University

and holds an M.B.A. from Pace University.

Robert Lloyd, President of Development & Sales

Robert Lloyd is currently the president of development & sales and has been with Cisco

since November 1994. Prior to his current role, Lloyd has held the positions of executive

vice president, worldwide operations, and senior vice president of U.S., Canada, and

Japan Operations. Lloyd also served as president of Cisco's Europe, Middle East, and

Africa (EMEA) region. Lloyd holds a B.Com. from the University of Manitoba.

Blair Christie, Chief Marketing Officer

Blair Christie serves as chief marketing officer, a role she has held since 2011. Prior to this

role, she served as senior vice president of Global Corporate Communications at Cisco.

Christie is an executive director and sits on Cisco's Operating Committee. She holds a

B.A. in marketing and an M.B.A. from Drexel University.

18 September 2013


Financial Statements Exhibit 148: Cisco Annual Income Statement in US$ millions, unless otherwise stated

Income statement FY11 FY12 FY13 FY14E FY15E

Total Sales 43,218 46,061 48,607 50,823 53,656

Cost of goods sold 15,857 17,298 18,207 19,260 20,789

Gross profit 27,361 28,763 30,400 31,563 32,867

Gross margin 63.3% 62.4% 63% 62.1% 61.3%

Products 62.4% 61.2% 61.3% 60.7% 59.4%

Services 67.1% 67.2% 67.0% 67.2% 67.5%

Operating Expenses

Research & Development 5,257 5,066 5,549 5,770 6,319

% of sales 12.2% 11.0% 11.4% 11.4% 11.8%

Sales & Marketing 9,141 9,055 9,048 9,254 9,801

% of sales 21.2% 19.7% 18.6% 18.2% 18.3%

General & Administrative 1,632 1,906 2,121 2,108 2,093

% of sales 3.8% 4.1% 4.4% 4.1% 3.9%

Total Operating Expenses 16,030 16,027 16,718 17,132 18,213

% sales 37.1% 34.8% 34.4% 33.7% 33.9%

Operating Income 11,331 12,736 13,682 14,431 14,654

Operating margin 26.2% 27.7% 28.1% 28.4% 27.3%

Interest and Other Inc/(Exp.), net 151 94 31 106 173

% of sales 0.3% 0.2% 0.1% 0.2% 0.3%

Pretax Income 11,482 12,830 13,713 14,538 14,827

% of sales 26.6% 27.9% 28.2% 28.6% 27.6%

Income Taxes 2,449 2,813 2,847 3,053 3,114

% tax rate 21.3% 21.9% 20.8% 21.0% 21.0%

Net income 9,033 10,017 10,866 11,485 11,714

% Net margin 20.9% 21.7% 22.4% 22.6% 21.8%

EPS Non-GAAP $1.62 $1.85 $2.02 $2.10 $2.13

Diluted shares outstanding (Non GAAP) 5,574 5,405 5,369 5,462 5,502


18 September 2013


Exhibit 149: Cisco Annual Balance Sheet in US$ millions, unless otherwise stated

Balance Sheet FY11 FY12 FY13 FY14E FY15E

ASSETS

Cash and equivalents 7,662 9,799 7,925 13,378 18,527

Short term investments 36,923 38,917 42,685 43,372 44,070

Accounts Receivable - net 4,698 4,369 5,470 6,272 7,002

Inventories 1,486 1,663 1,476 1,835 2,190

Deferred tax assets 2,410 2,294 2,616 2,616 2,616

Financing Receivables, net - 3,661 4,037 4,037 4,037

Prepaid expenses and other assets 4,052 1,230 1,312 1,312 1,312

Total Current Assets 57,231 61,933 65,521 72,821 79,754

Property, plant, and equipment - net 3,916 3,402 3,322 2,763 2,447

Goodwill 16,818 16,998 21,919 22,809 23,735

Purchased Intangible Assets 2,541 1,959 3,403 3,541 3,685

Financing Receivables, net - 3,585 3,911 3,911 3,911

Other assets 6,589 3,882 3,115 3,115 3,115

Total assets 87,095 91,759 101,191 108,961 116,647

LIABILITIES

Current portion LT Debt - - - - -

Short-term Debt 588 31 3,283 3,283 3,283

Accounts payable 876 859 1,029 1,009 1,095

Income taxes payable 120 276 192 192 192

Accrued Payroll Expenses 3,163 2,928 3,378 3,576 3,757

Deferred Revenue 8,025 8,852 9,262 9,805 10,302

Other accrued liabilities 4,734 4,785 5,048 5,344 5,615

Restructuring Liabilities - - - - -

Total Current Liabilities 17,506 17,731 22,192 23,208 24,244

Long-term debt 16,234 16,297 12,928 12,928 12,928

Income Taxes Payable 1,191 1,844 1,748 1,748 1,748

Long-term Deferred Revenue 4,182 4,028 4,161 4,327 4,501

Other long-term liabilities 723 558 1,034 1,034 1,034

Total Long-term Liabilities 22,330 22,727 19,871 20,037 20,211

Stockholders' Equity 47,259 51,301 59,128 65,715 72,192

Total liabilities & stockholders' equity 87,095 91,759 101,191 108,961 116,647


18 September 2013


Exhibit 150: Cisco Annual Cash Flow Statement in US$ millions, unless otherwise stated

Cash Flow Statement FY11 FY12 FY13 FY14E FY15E

Net (loss) income 6,490 8,041 9,983 10,479 10,810

Adjustments to reconcile net income to net cash

Depreciation & Amortization 2,486 2,602 2,351 1,859 1,669

Employee stock based compensation 783 1,401 1,120 1,273 1,144

Provision for doubtful accounts 7 25 66 - -

Deferred income taxes (157) (314) (37) - -

Tax benefit from employee stock plans (71) (60) (92) - -

Net (gains) losses on investments and provision for losses (213) (31) 9 - -

Changes in operating assets and liabilities:

Decrease (increase) in accounts receivable 298 272 (1,001) (802) (730)

Decrease (increase) in inventories (147) (287) 218 (359) (355)

Decrease (increase) in prepaid expenses and other current assets (634) (674) (27) - -

Increase (decrease) in accounts payable/accrued liabilities (28) (7) 164 (20) 86

Income taxes payable (156) 418 (239) - -

Accrued payroll & related expenses (64) (101) 330 198 181

Lease receivables (232) (821) (745) - -

Deferred revenues 1,028 727 598 543 497

Other accrued liabilities (148) 300 196 296 271

Total changes in operating assets and liabilities (83) (173) (506) (144) (49)

Net cash provided by operating activities 10,079 11,491 12,894 13,467 13,573

Investing Activities

Purchases of property, plant and equipment (1,174) (1,126) (1,160) (1,300) (1,353)

Purchases of short term investments - - - - -

Proceeds from sales of short term investments 17,538 27,365 14,799 - -

Purchases of investments (37,130) (41,810) (36,608) - -

Proceeds from maturities of investments 18,117 12,103 18,008 - -

Acquisitions, net (266) (524) (6,766) - -

Minority investments (41) 11 (30) - -

Other 22 166 74 - -

Net cash provided (used) by investing activities (2,934) (3,815) (11,768) (1,300) (1,353)

Financing Activities

Issuance of common stock 1,831 1,372 3,338 - -

Common stock repurchases (6,896) (4,760) (3,022) (3,000) (3,000)

Shares repurchased for tax witholdings on vesting RSU - - - - -

Issuance of Debt 4,109 17 24 - -

Repayment of long-term debt (3,113) - (16) - -

Short-term borrowings, net 512 (574) (20) - -

Other 80 (153) (5) - -

Excess tax benefit from share-based compensation 71 60 92 - -

Dividends Paid (658) (1,501) (3,310) (3,714) (4,071)

Total cash from financing activities (4,064) (5,539) (3,000) (6,714) (7,071)

Change in cash and cash equivalents 3,081 2,137 (1,874) 5,453 5,149

Cash and equivalents, beginning of period 4,581 7,662 9,799 7,925 13,378

Cash and equivalents, end of period 7,662 9,799 7,925 13,378 18,527

Free Cash Flow 8,905 10,365 11,734 12,167 12,221


18 September 2013


Companies Mentioned (Price as of 17-Sep-2013)

AT&T (T.N, $34.75) Accenture Plc (ACN.N, $76.61) Alcatel-Lucent (ALUA.PA, €2.521) Apple Inc (AAPL.OQ, $455.18) Aruba Networks (ARUN.OQ, $17.51) Broadcom Corp. (BRCM.OQ, $27.44) Brocade Comms (BRCD.OQ, $7.78) Check Point Software Technologies Ltd. (CHKP.OQ, $58.54) China Mobile Limited (0941.HK, HK$88.25) China Telecom (0728.HK, HK$4.13) China Unicom Hong Kong Ltd (0762.HK, HK$12.66) Cisco Systems Inc. (CSCO.OQ, $24.37, UNDERPERFORM, TP $21.0) Citrix Systems Inc. (CTXS.OQ, $74.0) Dell Inc. (DELL.OQ, $13.85) Deutsche Telekom (DTEGn.F, €9.921) EMC Corp (EMC.N, $26.81) Ericsson (ERICb.ST, Skr89.45) Extreme Networks (EXTR.OQ, $4.48) F5 Networks (FFIV.OQ, $91.63) Finisar Corporation (FNSR.OQ, $23.66) Fujitsu (6702.T, ¥370) Gigamon (GIMO.N, $39.81) Hewlett Packard (HPQ.N, $21.67) Infoblox (BLOX.N, $42.15) Intel Corp. (INTC.OQ, $23.74) International Business Machines Corp. (IBM.N, $192.16) Ixia (XXIA.OQ, $15.21) Juniper Networks (JNPR.N, $21.42) KDDI (9433.T, ¥4,800) LG (003550.KS, W67,500) Lexmark International (LXK.N, $33.75) Mellanox Technologies Ltd. (MLNX.OQ, $37.23) Meru Networks (MERU.OQ, $3.73) Microsoft Corporation (MSFT.OQ, $32.93) Motorola Solutions (MSI.N, $58.28) NEC (6701.T, ¥223) NTT DoCoMo (9437.T, ¥156,900) NetApp (NTAP.OQ, $43.82) Nokia (NOK1V.HE, €4.752) Oracle Corporation (ORCL.N, $33.26) Palo Alto Networks (PANW.N, $48.06) QUALCOMM Inc. (QCOM.OQ, $69.42) Rackspace Hosting Inc. (RAX.N, $52.62) Radware Ltd (RDWR.OQ, $14.86) Red Hat (RHT.N, $52.57) Ruckus Wireless (RKUS.N, $16.6) SAP (SAPG.F, €55.002) Softbank (9984.T, ¥6,360) Sprint Nextel Corp (S.N, $6.39) T-Mobile US Inc (TMUS.N, $25.09) TIM Participacoe (TIMP3.SA, R$10.06) Telefonica (TEF.MC, €11.27) Tellabs (TLAB.O, $2.34) Texas Instruments Inc. (TXN.OQ, $40.73) Ubiquiti Networks, Inc (UBNT.OQ, $32.57) VMware Inc. (VMW.N, $87.51) Verizon Communications Inc (VZ.N, $48.57) Vodafone Group (VOD.L, 211.25p) Xerox (XRX.N, $10.4) ZTE Corporation (0763.HK, HK$14.66)

Disclosure Appendix

Important Global Disclosures

Kulbinder Garcha and Vlad Rom, 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.

18 September 2013


3-Year Price and Rating History for Cisco Systems Inc. (CSCO.OQ)

CSCO.OQ Closing Price Target Price

Date (US$) (US$) Rating

11-Nov-10 20.58 27.00 O

10-Feb-11 18.92 24.00

12-May-11 16.93 26.00

11-Aug-11 15.92 25.00

10-Nov-11 18.61 26.00

09-Feb-12 20.00 27.00

16-Aug-12 19.00 25.00

28-Dec-12 19.45 NR

* Asterisk signifies initiation or assumption of coverage. O U T PERFO RM

N O T RA T ED

3-Year Price and Rating History for Dell Inc. (DELL.OQ)

DELL.OQ Closing Price Target Price


16-Mar-11 14.22 16.00 U *

23-May-12 12.49 14.00

22-Aug-12 11.68 11.00

16-Nov-12 8.84 9.00

05-Feb-13 13.42 R

* Asterisk signifies initiation or assumption of coverage.

U N D ERPERFO RM

REST RICT ED

3-Year Price and Rating History for EMC Corp (EMC.N)

EMC.N Closing Price Target Price


16-Mar-11 25.31 34.00 O *

18-Aug-11 20.59 27.00

24-Jan-12 25.14 30.00

22-Mar-12 29.13 35.00

18-Jul-12 25.08 30.00 *

01-Aug-12 26.14 *

25-Oct-12 24.14 30.00 O

* Asterisk signifies initiation or assumption of coverage.

O U T PERFO RM

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 secto r, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ra tings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin American and non-Japan Asia stocks, ratings are based on a stock’s total return relative to the average total return of the relevant country or regional benchmark; Austr alia, New Zealand are, and prior to 2nd

18 September 2013


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, 1 2-month rolling yield is incorporated in the absolute total return calculation and a 15% and a 7.5% threshold replace the 10-15% level in the Outperform and Underperform stock rating definitions, respectively. The 15% and 7.5% thresholds replace the +10-15% and -10-15% levels in the Neutral stock rating definition, respectively. Prior to 10th December 2012, Japanese ratings were based on a stock’s total return relative to the average total return of the relevant country or regional benchmark.

Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances.

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

Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation:

Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months.

Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months.

Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months.

*An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cov er multiple sectors.

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

Global Ratings Distribution

Rating Versus universe (%) Of which banking clients (%)

Outperform/Buy* 42% (55% banking clients)

Neutral/Hold* 40% (48% banking clients)

Underperform/Sell* 15% (40% banking clients)

Restricted 3%

*For purposes of the NYSE and NASD ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, an d Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdin gs, and other individual factors.

Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein.

Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: http://www.csfb.com/research and analytics/disclaimer/managing_conflicts_disclaimer.html

Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties.

Price Target: (12 months) for Cisco Systems Inc. (CSCO.OQ)

Method: Our valuation for CSCO's $21 target price is based on a three-pronged approach: i) P/E multiple valuation applying 9.8x to CY14 EPS; ii) discounted cash flow (DCF) analysis with terminal growth of 1% and a WACC of 11%; and iii) Credit Suisse HOLT analysis. Our approach

suggests that the company’s equity value is $114bn to $120bn, or $21 to $22 per share, with a blended value of $21 per share.

Risk: Risks to our $21 target price for CSCO include a stronger than expected macroeconomic recovery boosting IT spend, slower than anticipated SDN development and adoption and an increased commitment to cash return.

Please refer to the firm's disclosure website at https://rave.credit-suisse.com/disclosures for the definitions of abbreviations typically used in the target price method and risk sections.

See the Companies Mentioned section for full company names

The subject company (CSCO.OQ, DELL.OQ, EMC.N) currently is, or was during the 12-month period preceding the date of distribution of this report, a client of Credit Suisse.

Credit Suisse provided investment banking services to the subject company (CSCO.OQ, DELL.OQ, EMC.N) within the past 12 months.

Credit Suisse provided non-investment banking services to the subject company (CSCO.OQ) within the past 12 months

Credit Suisse has managed or co-managed a public offering of securities for the subject company (EMC.N) within the past 12 months.

Credit Suisse has received investment banking related compensation from the subject company (CSCO.OQ, DELL.OQ, EMC.N) within the past 12 months

18 September 2013


Credit Suisse expects to receive or intends to seek investment banking related compensation from the subject company (CSCO.OQ, DELL.OQ, EMC.N) within the next 3 months.

Credit Suisse has received compensation for products and services other than investment banking services from the subject company (CSCO.OQ) within the past 12 months

As of the date of this report, Credit Suisse makes a market in the following subject companies (CSCO.OQ, DELL.OQ, EMC.N).

Credit Suisse has a material conflict of interest with the subject company (DELL.OQ) . Credit Suisse Securities (USA) LLC is acting as a financial advisor to Silver Lake Partners in connection with the announced proposed acquisition of Dell Inc.

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 have not visited the material operations of the subject company (CSCO.OQ, DELL.OQ, EMC.N) within the past 12 months

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 http://www.csfb.com/legal_terms/canada_research_policy.shtml.

As of the date of this report, Credit Suisse acts as a market maker or liquidity provider in the equities securities that are the subject of this report.

Principal is not guaranteed in the case of equities because equity prices are variable.

Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that.

Important Credit Suisse HOLT Disclosures

With respect to the analysis in this report based on the Credit Suisse HOLT methodology, Credit Suisse certifies that (1) the views expressed in this report accurately reflect the Credit Suisse HOLT methodology and (2) no part of the Firm’s compensation was, is, or will be directly related to the specific views disclosed in this report.

The Credit Suisse HOLT methodology does not assign ratings to a security. It is an analytical tool that involves use of a set of proprietary quantitative algorithms and warranted value calculations, collectively called the Credit Suisse HOLT valuation model, that are consistently applied to all the companies included in its database. Third-party data (including consensus earnings estimates) are systematically translated into a number of default algorithms available in the Credit Suisse HOLT valuation model. The source financial statement, pricing, and earnings data provided by outside data vendors are subject to quality control and may also be adjusted to more closely measure the underlying economics of firm performance. The adjustments provide consistency when analyzing a single company across time, or analyzing multiple companies across industries or national borders. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes the baseline valuation for a security, and a user then may adjust the default variables to produce alternative scenarios, any of which could occur.

Additional information about the Credit Suisse HOLT methodology is available on request.

The Credit Suisse HOLT methodology does not assign a price target to a security. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes a warranted price for a security, and as the third-party data are updated, the warranted price may also change. The default variable may also be adjusted to produce alternative warranted prices, any of which could occur.

CFROI®, HOLT, HOLTfolio, ValueSearch, AggreGator, Signal Flag and “Powered by HOLT” are trademarks or service marks or registered trademarks or registered service marks of Credit Suisse or its affiliates in the United States and other countries. HOLT is a corporate performance and valuation advisory service of Credit Suisse.

For Credit Suisse disclosure information on other companies mentioned in this report, please visit the website at https://rave.credit-suisse.com/disclosures or call +1 (877) 291-2683.

18 September 2013


References in this report to Credit Suisse include all of the subsidiaries and affiliates of Credit Suisse operating under its investment banking division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who_we_are/en/This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse AG or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates. The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. CS may, to the extent permitted by law, participate or invest in financing transactions with the issuer(s) of the securities referred to in this report, perform services for or solicit business from such issuers, and/or have a position or holding, or other material interest, or effect transactions, in such securities or options thereon, or other investments related thereto. In addition, it may make markets in the securities mentioned in the material presented in this report. CS may have, within the last three years, served as manager or co-manager of a public offering of securities for, or currently may make a primary market in issues of, any or all of the entities mentioned in this report or may be providing, or have provided within the previous 12 months, significant advice or investment services in relation to the investment concerned or a related investment. Additional information is, subject to duties of confidentiality, available on request. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk. This report is issued and distributed in Europe (except Switzerland) by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority ("PRA") and regulated by the Financial Conduct Authority ("FCA") and the PRA. This report is being distributed in Germany by Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). This report is being distributed in the United States and Canada by Credit Suisse Securities (USA) LLC; in Switzerland by Credit Suisse AG; in Brazil by Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; in Mexico by Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); in Japan by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau (Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; elsewhere in Asia/ Pacific by whichever of the following is the appropriately authorised entity in the relevant jurisdiction: Credit Suisse (Hong Kong) Limited, Credit Suisse Equities (Australia) Limited, Credit Suisse Securities (Thailand) Limited, having registered address at 990 Abdulrahim Place, 27 Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok 10500, Thailand, Tel. +66 2614 6000, Credit Suisse Securities (Malaysia) Sdn Bhd, Credit Suisse AG, Singapore Branch, Credit Suisse Securities (India) Private Limited regulated by the Securities and Exchange Board of India (registration Nos. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777, Credit Suisse Securities (Europe) Limited, Seoul Branch, Credit Suisse AG, Taipei Securities Branch, PT Credit Suisse Securities Indonesia, Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore branch to overseas investors (as defined under the Financial Advisers Regulations). By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the "FAA"), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore branch may provide to you. This research may not conform to Canadian disclosure requirements. In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-U.S. customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. U.S. customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the U.S. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the PRA and regulated by the FCA and the PRA or in respect of which the protections of the PRA and FCA 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 © 2013 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.

Cisco Inititation new.doc

https://www.credit-suisse.com/who_we_are/en/

Cisco Systems Inc. (CSCO) - Credit Suisse | PLUS

Documents

Transcript of Cisco Systems Inc. (CSCO) - Credit Suisse | PLUS