Cisco: Financial Machine Or Growth Engine?

| About: Cisco Systems, (CSCO)


Cisco's recent actions to increase its dividend and to increase its share buybacks have bolstered share values significantly.

The company was able to maintain gross margins last quarter in the wake of a significant sequential slide in some of its product categories.

Its operating cash flow was bolstered significantly during the quarter because of the full-year benefit of the investment tax credit, coupled with a material reduction in receivable balances.

The company faces some significant longer-term issues in the form of the widespread adoption of SDN architectures, which, by their nature, are open and agile.

Cisco is seen by its users as having relatively high prices in terms of the constituents of its competitive SDN solution.

What can you possibly say about Cisco that hasn't already been said?

I confess to being a fundamentalist. No, that isn't the same as being an originalist, or some kind of new religion. It simply means I like to value companies based on their cash flow and growth. How that growth is returned to shareholders is something else again. Many readers want and need dividend yield. They are attracted to companies like Cisco (NASDAQ:CSCO) that have high cash flow generation and are able to pay significant dividends. Cisco's current yield, based on today's share price, is 3.67%.

The problem with pursuing a dividend strategy in the tech space is that the qualities of a company that can produce high payouts and high dividends are often the same qualities that threaten the dividend as new technologies replace older technologies and market share begins to shift in favor of newer competitors. I have seen many articles of this site regarding potential ways to value Cisco or to improve one's yield as a Cisco stock holder, or even ways for the company to increase value for shareholders. I don't know if I have much to add to that kind of discussion.

But as many readers and observers know, Cisco faces an existential threat to its heartland business of selling switchers and routers. It has taken many years for this threat to mature, and even now, the real impact is still in the future - it is just that the future is starting to be foreseeable. Of course, the threat is what is called Software-defined Networking (SDN), and as the name implies, it is all about software and far less about hardware. At the end of the day, and despite efforts that have had some level of success, the company still has to sell switches and routers to sustain the cash flow that is necessary to pay those dividends. Is it completely heretical to even think about what might happen in the alternate case? I don't believe at this point I have (or anyone has) all of the answers.

When I first heard about SDN some years ago, I was struck by its elegance and simplicity as a technology that could do away with one of the real choke points in modern networks. But I was also struck by how much the technology needed to change and improve before it really became a significant threat to Cisco, and for that matter, to all of the other networking hardware vendors.

I think that it is a reasonable exercise for readers to be exposed to the other side of the story, in which the company has to adopt to rapidly changing customer behavior with the advent of mainstream SDN technologies. Just how safe might the yield really be? Is there significant danger of capital loss? Just how realistic are the prospects of Cisco being able to pay annual dividend increases? Short of prescience, which I certainly do not claim to have, there is no way to answer all of those questions. So, about all I can do is suggest some potential straws in the wind and conclusions that might flow from those straws.

What the heck is Software-defined Networking, and why should I care? I bought Cisco as a dividend play with no real expectation of capital gain. I don't need for it to grow, just for it to remain stable.

All right, the standard definition of software-defined networking is that it is a term used to describe technologies aimed at making networks as agile and flexible as the virtualized server and storage infrastructure of the modern data center. In an SDN architecture, an administrator can essentially shape traffic flows from a centralized control console without having to touch individual switches. What does that mean in real life? Basically, it means that network administrators can significantly reduce the capacity requirements for switches by shifting resources from where they are deployed to where they are needed. That has already been done in the other areas of the data center through virtualization and through storage efficiency, and has really an enormous influence on what heretofore were thought to be growth markets. If the cloud is considered to be "the big thing" in software these days, then SDN is its analog within the data center. I really do not think that from the point of view of an investor, it is necessary to discuss in any greater detail how SDNs actually function.

There are 3 key technologies that are used to design SDNs. These are what is called functional separation, network virtualization and automation through programmability. I think these terms define themselves to most readers. There are perhaps a thousand of vendors (that figure is from Gartner) and multiple definitions of the market in which they compete.

Gartner's research director for the space maintains that SDN shouldn't be thought of as a product, but rather, should be considered an architecture. He likes it better if the space is called enterprise SDN data centers. OK. And within that space, Gartner has produced some interesting research.

Gartner says the company to watch in the space is a smaller vendor called Arista Networks (NYSE:ANET). Arista is not exactly unknown, with a market cap of $4 billion and a revenue run rate of a bit more than $1 billion. Very few vendors offer a comprehensive set of solutions that will achieve all the goals of SDN. Companies included in Gartner's list of important vendors that most readers will recognize include Cisco, Citrix (NASDAQ:CTXS), Dell, Hewlett Packard Enterprise (NYSE:HPE), IBM Corp. (NYSE:IBM), Alphabet (GOOG, GOOGL) and VMware (NYSE:VMW).

While Gartner says "Cisco has the largest installed base" of any of the vendors in the study, it goes on to say that its products are amongst the highest-priced. The company has a history of creating products with closed, proprietary features. Moreover, Cisco's Application Centric Infrastructure (ACI) (another way of describing SDN) doesn't offer investment protection for existing Catalyst and Nexus customers (Catalyst and Nexus are Cisco switching products) of having the features that some organizations use. Gartner estimates there are fewer than 25 large-scale production ACI installation. (That statistic is a year old, but the new survey has yet to be released.) Enterprise clients continue to report that Cisco has difficulty producing production references for ACI deployment. Again, according to Gartner, product differentiation is increasingly shifting to software, including automation and orchestration, with hardware capabilities like bandwidth... becoming more standardized.

Companies talk about lots of things on conference calls that really ought not to be said. On Cisco's Q1 FY 2015 conference call, former CEO John Chambers said rather pointedly, "we are going to lead and breakaway from most every player in the enterprise and that applies across all products... with our Applications-Centric Infrastructure." Network World, an industry magazine, suggested that Cisco would much rather defeat the entire concept of SDN because of its openness rather than speak to competition with individual competitors.

Okay, I have heard enough about SDN and Cisco's role or non-role in the trend. How does it impact the bottom line?

Early in February, Cisco reported results for what was its fiscal Q2 (ended 1/31). I think that most readers know that the results were considered to be very strong despite minimal revenue growth. The company provided a typical EPS beat and did not change its guidance materially. (The current quarter for Cisco has 14 weeks instead of the normal 13 weeks, which increases estimates by 2-3%. Net of that adjustment, the company is calling for little change in revenues in the current quarter.) The big news was that Cisco increased its dividend by 24% and announced an increased share buyback of $15 billion. The shares rallied sharply in the wake of the earnings. They were up 10% the next day, and have increased by almost 26% from the early February low. That said, the shares have more or less retraced the dip they took along with most tech stock shares between the end of December and early February, and they still have not quite reclaimed the late October highs. I will comment on some detailed financial analysis shortly, but I think of equal significance to the actual numbers that Cisco printed was the comment by company CFO Kelly Kramer. "We continue to shift our business model to more software and subscription recurring revenue." CEO Chuck Robbins declared that, "In just two years, we have built ACI to a $2 billion run rate business that grew once again last quarter over 100%." (ACI is what Cisco calls its flavor of hardware and software that facilitates the development of an SDN architecture.)

Cisco, or so it seems to me, well understands that SDN is one of the few growth engines in the networking space, although the company is also focused on trying to grow security and cloud networking in its portfolio. I think when measuring the success of a transition, it is appropriate to try to determine the migration between what has become legacy to the new flashy stuff, such as ACI for this company.

The two closest components of Cisco's product revenue breakdown that compare to the functionality that is part of the ACI solution are switching and data center. Data center revenues fell 3% to $822 million for the quarter year on year, and were down by 4% sequentially. Switching revenues fell 4% year on year and more than 13% sequentially.

Just looking at the quarterly numbers, therefore, it appears that ACI increased revenues by $250 million, while the declines in switching and data center business as a whole reduced revenues by about $170 million. So, at least notionally, the transition is still a small net positive for the company. Management stated that it believed the declines in switching and data center were the result of difficult compares and some sort of macro malaise that hit the company in the first three weeks of January. Presumably, therefore, we ought to expect some reasonable bounce back this quarter given the better macro trends seen in recent weeks.

But the overall issue, at least for me, is how the trend towards software-defined networks either adds to or subtracts from Cisco's growth prospects. I don't think there is anyone who believes the company's ACI product line-up will not be successful and will not be a dominant player in the SDN space. Cisco may have been a bit slow to this party, but honestly, not all that slow for a large company in the midst of a management transition.

Simply put, what the SDN paradigm is trying to do is to make networks more agile so that they can work effectively with cloud data centers. There are far more complexities to how that is done than is warranted in this discussion. But what SDN architecture does is to reduce the overall demand for switching power in a cloud data center by allowing administrators to shape traffic. Shaped traffic is simply more efficient so that users can achieve a higher level of throughput without investments in additional switches. And the switches that can be plugged into the network can be commodity off-the-shelf components. When Cisco management is asked about what are called "white box switches," it becomes more or less apoplectic.

Cisco's response to SDN has been to attempt to migrate users to its own proprietary version of SDN or ACI. The company is being very proactive in encouraging migration. It is always difficult to evaluate whether the competitive offerings are functionally better or worse. By all accounts, competitive offerings are certainly cheaper than what Cisco offers. The problem with the ACI strategy, it seems to me, is that it is essentially an "all or nothing" strategy. If you join the Cisco ACI ecosystem, you shut out potential competitors and you resign yourself to paying higher prices by doing so. The exchange for this is that you wind up working with a single company, and these days, no one is going to get into trouble for selecting a total Cisco solution.

How does that play out in the real world? It means that Cisco will face extraordinary pricing pressures from users in order to maintain its closed and proprietary systems. Before users give up their ability to plug commodity components into a heterogeneous SDN network, they are going to force the company to at least be somewhat competitive.

So, I think two issues arise. One of these is that the overall demand for data center components in an SDN world is going to be less than heretofore. And the other trend is that whatever hardware will be bought for the new cloud data centers that are used with an SDN architecture, it is going to inevitably have lower gross margins than the products that are being replaced. Slower "unit" growth and more margin pressures - not a good thing.

For obvious reasons, Cisco does not, and never will, report gross margins by product category. But what we do know is that despite a significant fall in sequential quarter hardware revenue, the company was actually able to marginally able to improve gross margins. I think there are very few companies that would be able to make up a 9% sequential quarter drop in hardware revenues and maintain gross margins. Not that Cisco is really in the manufacturing business, but anytime you ship products, there are inevitably fixed costs, and almost no one can react fast enough to offset what would have been a significant decrease from planned volumes. It should be remembered that typically for Cisco, Q2 is sequentially stronger than Q1.

Company management suggested that Cisco was "getting real traction... from our engineers and supply chain on the productivity side." The CFO also remarked about price discipline, which is essentially financial management speak for discipline in terms of price discounting. So, less discounting and improved productivity, but lower sales in two strategic areas, particularly including data center. Is there a correlation? I think at this point, it is hard to say. There are lots of elements that suggest that if Cisco really wants to solidify its position in its version of SDN, or as the company calls it, the next-generation data center, it is going to have to become significantly more aggressive in terms of price. So far, Cisco has chosen a strategy that conforms to the normal price/performance improvements that the industry expects on a continuous basis. My guess is that will not continue to be possible indefinitely into the future.

The company really has a has a long record of squeezing expenses and of buying all the technology necessary to maintain a leadership role. I think, however, that the current situation is a bit unique. The entire essence of SDN, which revolves around openness and nimbleness, is more or less antithetical to Cisco's longer-term strategies in terms of boxing out competitors. And I believe that SDN is the first major innovative architecture to emerge that was designed to reduce the ultimate size of cloud data center build-outs.

I am not sure that all of these cautionary signs are going to much concern investors in the short term. Investors are most concerned by cash flow, as that is the source for dividends and share buybacks. Cash flow rose 36% year on year in the last quarter, and free cash flow rose by 40% to $3.6 billion. When carefully considering the growth in cash flow, it becomes evident that the large increase was primarily a function of the settlement of "a significant tax matter", i.e., the reinstatement of investment tax credit, along with a negotiated settlement with the IRS on past disputes, coupled with acquisition-related/divestiture gains. The increase in deferred revenue balance for the January 2016 quarter was almost negligible on a sequential basis. Deferred revenues were able to grow moderately year on year. Stock-based comp was a bit more than 10% of operating cash flow, and has been stable, relatively speaking, for a few years. The other relatively significant item in cash flow that jumps out is the substantial decrease in accounts receivable, most likely a function of the sales problems that Cisco had in the last 3 weeks in the quarter. While there are other puts and takes in the cash flow statement, the decrease in A/R was just shy of $1 billion this year, compared to $501 million in the year-earlier period. Were one to remove all of the one-time items that made up the cash flow this past quarter, much of the 36% increase in operating cash flow would disappear.

While Cisco has cash balances of $60 billion on a worldwide basis, as is typical for larger and older companies, all but $4 billion is domiciled outside of the United States. Overall, the way the company calculates these things, it returned $4.6 billion, or 75% of free cash flow, to shareholders in the first half of the fiscal year. The company will now be spending about $5 billion/year to pay its new increased dividend. While the company doesn't explicitly forecast operating cash flow, it seems likely to me, given its earnings guidance, that operating cash flow in the second half of the year is likely to be in the range of $5-5.5 billion. The combination of the company's dividend and committed share buybacks is likely to be 75% or more of that amount.

At least for the short term, in terms of cash returns to shareholders, what you have seen is what you are going to get, and perhaps it is all you might expect to get for some time to come. This company is going to have to make acquisitions in order to fill product and technology holes, and with all of the other demands on cash flow, Cisco probably will need to borrow money, since it isn't likely to be repatriating cash to the US anytime until the punitive costs of such a move have been abated.

Valuation and a few final thoughts!

I think that it is fair to say that at this point, Cisco represents the sort of value story beloved by many investors. This is a tech company that returns enough of its cash flow to shareholders such that it can be considered one that will provide a small amount of growth and a fair degree of visible and stable dividend income. Indeed, with 23% of the outstanding shares in the hands of individuals, it is clearly a favorite tech name amongst retail investors. Cisco has an EV/S of 2.15, based on current fiscal year projections. As revenue growth is expected to be just about 2% next year, the EV/S ratio is more or less the same. The company has a non-GAAP P/E of 12.3X on current-year earnings and a P/E of 11.8X based on estimates for the following fiscal year. Its most likely free cash flow, including the one-time events that influenced the just past quarter, is about $11 billion for the year. On that basis, the company has a free cash flow yield on its enterprise value of over 10%. That last one is an impressive figure and ought to be comforting to many investors.

On the other hand, as I have tried to point out, the beer is really not likely to get materially colder than it is now. Lost in the shuffle of increased share buybacks and a greatly increased dividend were the comments about the slowdown seen in the last 3 weeks of the prior quarter. Particularly disquieting was the 13% sequential fall in data center revenue, when that is supposed to be a mainstay of future growth.

I do not really expect that Cisco will lose the SDN wars in some dramatic fashion. My guess, however, is that the increasing adoption of SDN, or Cisco's version of SDN, will further serve to limit growth in the coming years. I think that if the company does not achieve an upside of some visible amount in the current quarter, I might want to revisit that comment. Value stocks cease to be so if the environment changes or their execution capabilities deteriorate. The company is not going to have execution issues now any more than it has really had since the dark days of the tech stock bubble implosion at the end of 2001 and through 2002. But the environment in the networking space is changing, and not in a fashion that will necessarily be of benefit to Cisco.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.