Description

Over the past several years, the bear case on CSCO has remained largely the same: Cisco’s legendary 60-70% gross margins are poised to decline given increased competition due to the emergence of new emerging switching and networking technologies, as well as the rapid growth of the cloud.  This threat has intensified, if anything, over the past few years as technology leaders such as Google and Facebook have embraced Software Defined Networking vs. Cisco’s traditional networking technology and highly visible companies such as Netflix have largely outsourced their IT to the cloud. 

Even on GAAP earnings adjusted for Cisco’s $35B in cash taxed at 35%, Cisco now trades at 9.1x F13 (E) earnings and 10.2x F14(E) earnings, vs. the market at over 15x. 

At this point, we think that Cisco’s stock price overly discounts the risks to the business and ignores potential upside from recent new product introductions in the core business and ongoing growth in new product areas.  We believe there is a decent possibility that the combination of new product introductions in Cisco’s core networking and switching businesses and ongoing success in several growth areas of the business will lead Cisco to at least maintain, if not modestly grow earnings over the next several years.  Should our base-case of flat to 5-7% annual EPS growth play out, we see the potential for 30-50% multiple expansion from 9-10x, leading to 50%+ potential upside over the next 12-24 months.  

Cisco’s 2013 Revenue Breakdown

Networking                   47%

 Routing                        17%

 Switching                     30%

New Products               30%

 Service Provider Video 10%

 Collaboration                 8%

 Security                         3%

 Wireless                         4%

 Data Center                    4%

Services                       22%

Source: Cisco

For reference, a core component of the bear thesis on Cisco surrounds the potential for Cisco to lose its enterprise networking business due to outsourcing to the cloud which will be provided by service providers.  Note that roughly 50% of Cisco’s revenues, as well as perhaps 2/3 of the service revenues (15% of total revenues), derive directly from the total legacy networking business, though a Credit Suisse analysis of Infonetics data estimates that approx. 33% of Cisco’s total revenue derives from enterprise networking products, while approx. 14% of total revenues derive from networking products sold to service providers.  Assuming an additional one-third of Cisco’s services revenues derive from enterprise networking (for 7% of total revenues), enterprise networking could potentially comprise approx. 40% of Cisco’s total revenues.  And given relatively higher margins of 60-70% estimated in the networking business vs. 50% in new products, Cisco’s enterprise networking business potentially comprises 50% of Cisco’s total gross margin dollars.  The fact that such a large percentage of Cisco’s earnings derive from the core enterprise networking business adds weight to the bear thesis of the decline of on-site enterprise IT on the stock; we think that Cisco can largely maintain this revenue stream through sales of its legacy and new products to both enterprises and service providers.

Key Components of Buy Thesis:

First, the stock is very cheap.

CSCO is trading at 9x F13 GAAP EPS ex-tax adj. cash, and 10x F14(E) EPS ex-tax adj. cash.  From a valuation perspective, not taking into account the cash, CSCO is trading at a near 5 year low on trailing GAAP EPS (11.5x vs. an average of 15.7x), and an absolute low from a relative perspective vs. the S&P 500 (0.74x vs. an average of 1.14).

Also, the FCF yield is 10% and it currently pays a 3.1% dividend.

This valuation cushion would allow CSCO's earnings to decline 30%+ and have the stock still be reasonable relative to the market. 

What are the chances that CSCO's earnings experience such a steep decline in the near future? 

Very low.  Even the bears anticipate that Cisco holds or only slightly grows non-GAAP earnings over the next several years as volume growth helps to offset their assumed margin declines given the continued robust demand environment for networking equipment.  For example, on a non-GAAP basis, excluding stock options (as these are the headline numbers that many investors follow), the Street anticipates that earnings are anticipated to decline 2% in 2014 followed by increasing only 5% in 2015. 

And let’s say that even the bears are too optimistic and earnings decline beyond 2014.  The stock should still work unless earnings declines become truly secular and fall 30%+ from 2013 levels without a valuation offset at some point.  A long term decline in the stock price is certainly feasible given the potential for kitchen-sinking valuations in tech, but at some point valuation should provide some protection.  Also, CSCO has several levers that suggest that the company is more likely to grow EPS in the future than to experience EPS declines.

Second, the bar for 2014 has been set very low. 

In the recently reported 1Q14 quarter (Oct. 31^st end), CSCO guided to an 8-10% y/y decline in 2Q14 sales, and 2Q14 non-GAAP EPS of $0.46, 10% below Street expectations.  It further guided to non-GAAP F14 EPS of $2.00 (-1% y/y) vs. the Street at $2.10.  While the bears would point to this significant revenue guidance miss as a sign of worse performance to come, the two primary sources of the miss – a 20% decline in emerging markets (perhaps 10-15% of total revenues) and the federal shutdown (perhaps 1/3 of the total miss) and product transition issues in switching and routing ahead of new product roll-outs (likely another 1/3) – are likely to stabilize/improve over time.  A 20% decline in the set-top box business (perhaps 10-20% of the revenue miss) is likely to continue but should have a minimal impact on earnings given the low-single-digit margins in this business.  And the weakness in low-end edge routing due to competition from Juniper and Alcatel is worth noting, but is too small to materially impact profits should it continue longer term.  Overall, stabilization of the business at this new lower level is likely to drive 10%-20% improvement in valuation as investors gain confidence that 2014 is the bottom.  In addition, if investors begin to acknowledge the potential for earnings improvement from the successful roll-out of new products, Cisco could experience 30%+ valuation improvement over the coming year.

Further, the Street has already brought down its own expectations for CSCO’s revenue growth over the next three years to 3-5% from CSCO’s 5-7% stated at its 2012 analyst day.  This should help to support the stock should Cisco’s revenues take several quarters to turn around.

Third, this year CSCO has the potential to negate the the bear case that its margins are poised to decline.

The fundamental bear case is that CSCO's gross margins are just too high at 61% and must "normalize" to perhaps 30%-40% at some point, given those margin levels in other tech hardware businesses such as servers and storage.  Adding further credence to this bear case is the emergence of Software Defined Networking "SDN" which allows the customer to pair white-box networking equipment with open source software to save ~75% on hardware costs vs. a Cisco solution, exposing roughly 45% of Cisco's business to potential significant disruption. 

Here are some thoughts on the long-term bear case.  First, SDN is in very early stages and will take a minimum of 2-3 years to become "commercial" enough for most customers to consider it.  This suggests that any revenue loss from this business would be very gradual, and allow Cisco and the other networking vendors time to potentially "catch up" and offer competitive solutions. 

In addition as of August 2013, IT decision makers had mixed views on whether SDN would lead their total networking spend to decline; a full 36% anticipated that spending would increase, and 21% anticipated that spending would not change, per a Credit Suisse Networking Survey.

Second, and importantly, in November CSCO launched with a new series of products, "Application Centric Infrastructure" "ACI" from its spin-in, Isieme, which propose to offer a significantly more cost effective and technologically advanced networking solution to customers than white-box SDN.  Given the buy-in that Cisco has already received from key partners such as F5, Embrane, Symantec, Citrix, Redhat, and from initial customers including Microsoft's Azure Cloud and the second largest cloud provider in Europe, the value proposition appears to be credible. 

Note that the new ACI technology costs customers more and is more profitable to Cisco than Cisco's current solution, suggesting that it could lead to earnings growth for Cisco rather than earnings declines.  So why would customers buy this more expensive solution?  Because it could save customers 75% in total costs vs. white-box SDN solutions and also provide savings versus Cisco’s current solutions while providing improved functionality.  How is this possible?  First, over 75% of IT budgets are devoted to operating expenses, and ACI significantly reduces the need for IT staff through automation of network optimization.  The conversation around SDN has been too focused on lower cost equipment given the need for increased operating expenses to manage these white-box systems and the ultimate need for better performance.  

Importantly, ACI provides this increased flexibility and programmability needed to manage more dynamic networks, and also incorporates key application control and virtualization functionality further up in the stack, thereby incorporating functionality previously offered by other vendors.  Based on customer comments, the products provide superior functionality to any product available on the market today. 

What if ACI doesn't live up to its promise?  Cisco still has several years to improve this and other solutions before SDN can present a meaningful threat.  And given Cisco's #1 rating with IT leaders for its technical roadmap, price/performance, management quality, maintenance and support, distribution, sales force quality, and cloud strategy, as well as its cadre of Cisco certified network administrators working at customer sites, Cisco will be much more difficult to displace than the bears think.

Separately, Cisco recently announced two key new routing products – NCS (Network Control Systems) and CRS-X “Carrier Routing System” – to replace older technologies.  Customer anticipation of the availability of these new products has likely led to lower orders of the current products, leading CSCO to likely experience an acceleration in routing revenue later in the year.  This should provide a further boost to earnings vs. current expectations.

Fourth, Cisco can still succeed despite an enterprise IT migration to the cloud

In addition to Software Defined Networking, the second existential threat to Cisco is that as companies migrate IT to the cloud, Cisco will lose significant revenue due to major cloud providers choosing a white-box solution over Cisco’s products.  This threat to Cisco’s business model is underscored by the fact that the largest enterprise cloud provider, Amazon, has largely chosen to pursue a white-box Software Defined Networking solution, and other large consumer cloud-service providers such as Google and Facebook have chosen an SDN path as well.  While Cisco states that its new ACI platform would provide these large cloud providers with significant savings, these companies have already embedded SDN solutions into their networks, making displacement by new Cisco products potentially unlikely.

Should Amazon continue to dominate the enterprise cloud infrastructure space, it could be difficult for Cisco to recoup revenue lost by enterprise customers who migrate to Amazon’s service.  (Google and Facebook’s selection of SDN solutions currently pose less of a threat to Cisco given their generally consumer-focused cloud offerings.)  However, it seems more likely that Amazon will gain only a portion of the market, and other cloud providers will capture meaningful share over time.  Importantly, a key contender for enterprise cloud market share, Microsoft’s Azure Cloud service, is supporting Cisco’s ACI solution.  Should Cisco’s ACI deliver, Cisco has the opportunity to maintain its current revenue streams by selling its products to Microsoft and other cloud providers as companies migrate to the cloud. 

To put the cloud threat into perspective, it is estimated that Amazon Web Services (AWS), by far the largest cloud provider, will generate $3B-$3.5B in sales in 2013 out of $300B+ in estimated global data center spending. From a high-level perspective, spending on the cloud is still relatively small, and many companies are likely to continue to maintain their in-house IT departments for many IT networking functions.  Further, companies that choose to maintain their in-house IT departments are likely to continue to experience increased network traffic, thereby increasing the demand for Cisco’s products.  Given Cisco’s current cheap valuation, we’re prepared to give Cisco the benefit of the doubt that it is able to make inroads with many top cloud providers and maintain a large proportion of its spending with corporate IT departments for the functions they choose to keep in house.

Fifth, the Huawei threat is likely to remain limited.

Independent of the case that Software Defined Networking will destroy CSCO’s margins and overall demand for its products, there is a separate bear case that low-cost emerging market vendors such as Huawei will gradually eat away market share and force gross margins lower.  1Q14’s recent significant weakness in emerging markets could give credence to this bear case, though there is likely some truth to CSCO’s claim that the emerging markets were very weak due to macro factors and political issues in China (perhaps also due to the fall-out from the NSA security breaches). 

For perspective, Huawei has gained 16% share of the service provider market over time primarily in fast-growing edge routers, and beginning in 2011 the company began making a push into the enterprise, Cisco’s largest segment.  As Huawei’s cost structure and margins are much lower than CSCO (with GM’s in the range of 37-40% and operating margins around 10%, well below CSCO‘s 60% GM’s and 27% operating margin), Huawei offers customers a much lower cost solution.  However, political dynamics make it highly unlikely that Hauwei will ever meaningfully penetrate North America, which represents 33% of the global service provider segment, 42% of the global enterprise segment, and 25% of the global telecom capex per Credit Suisse and Infonetics.  And perhaps more importantly, at least per a recent Credit Suisse survey of IT decision makers, Huawei is perceived to be far worse than nearly all the other vendors on almost all metrics.  (Note that Huawei performed very highly on price/performance/reliability in a survey of IT networking professionals, but relatively low on every other factor.)  IT decision makers’ mixed perceptions of Huawei make it seem unlikely that Huawei is likely to significantly accelerate its market share across networking infrastructure anytime soon.

Key Risks to Buy Thesis

There are two fundamental risks to the strong-buy thesis on Cisco.  The first risk is more perceptual that the market never begins to give Cisco credit for the potential that the bear thesis might not play out.  This is the risk of buying stocks with secular overhangs that could take many years to disprove.  But if the only risk to holding the stock is perceptual and the fundamentals hold, we should at least be unlikely to lose much on CSCO given the current very low valuation.

The second major risk derives from the potential that CSCO’s new products do not deliver a compelling enough value proposition to lead enterprise customers and cloud providers to choose CSCO’s products over a software defined networking solution.  While the initial customer uptake and customer commentary suggests that new products should gain traction, the products are still in relatively early stages and could potentially disappoint.

The third major risk derives from the enterprise shift to the cloud.  Should enterprises gradually move their data management to “whitebox” cloud vendors, CSCO will inevitably lose its networking business from its enterprise customers.  To date, CSCO’s data center business which serves the cloud has been growing strongly, though the two of the largest public cloud providers – AMZN and GOOG – have chosen to pursue a white box approach to networking.  Should the major public cloud providers continue to pursue a white-box solution based on momentum from their previous investments, and enterprises shift increasingly more data to these providers, CSCO is likely to see gradual ongoing declines in its core enterprise networking business. 

[The views expressed are those of the author and do not necessarily represent the views of any other person. The information herein is obtained from public sources believed to be accurate, reliable and current as of the date of writing.  The author will not undertake to supplement, update or revise such information at a later date.  The author may hold a position in the securities discussed.]

Catalyst