The substance of this article is that Intel stands to lose a big chunk of its gross margin and that it currently has no plausible replacement. If it can't generate the cash, then it won't be able to pay the dividends for which most have been buying the stock.
If you believe that Intel (NASDAQ:INTC) has a bright future, you have to believe one of two things: the Wintel franchise will be as successful as it has been for a very long time, or Intel will find some other meaningful revenue stream to replace the Wintel franchise, something it has failed to do for a very long time.
Intel reported a full-year revenue of $53.3 billion for FY2012. Of this $34.3 billion, or 64.4%, came from the PC client group, $10.7 billion, or 20.1%, came from the data center group, and $4.4 billion, or 8.3%, came from the "Intel other architecture group." Intel reports a gross margin of 58% not broken down by group.
Those numbers hide some important facts that have enormous financial consequences. Processors sold for PC clients (PCs or laptops) are often essentially the same microarchitecture and in some cases the exact same chip as is sold for the data center group, except that the chips sold for data centers sell for 2-5 times as much as chips sold for PC clients. That means that the data center group contributes to gross margin (and profit) well out of proportion to the revenue it generates.
Chips that are sold for other applications right now just aren't that important for Intel. If a substantial number of chips is ever sold into the mobile market by Intel, it will have to deal with the fact that its chips typically cost five times what Qualcomm (NASDAQ:QCOM) chips for the mobile market sell for. No one that I know believes that Intel can sell into that market at the gross margin to which it has become accustomed, and, even if it could, it would have to sell five times as many chips to produce the same top line revenue number.
Intel may well crack the mobile market and be just as profitable as it always has been. It could do that by trying to compete with Qualcomm (unlikely), or, in collaboration with a software vendor, it could produce the chip that runs a product with features sufficiently compelling to justify a chip that is more in the price (and gross margin) range it is accustomed to. Such a product would likely be a tablet, and could conceivably be developed in collaboration with Google (NASDAQ:GOOG) or Microsoft (NASDAQ:MSFT).
Intel has a long history of successful collaboration (or even dependence on) Microsoft, but it has no history of successful collaboration with Google in the PC client space. Microsoft is having problems of its own.
If Microsoft can come up with a tablet solution that needs a fairly high-powered chip that comes at a reasonable price and that consumers really want, then Intel's problems may be over. That's a great many ifs. It's hard to imagine how anything Intel developed in collaboration with Google would be sufficiently differentiated to allow it to compete with Qualcomm. Under the successful Microsoft scenario, you should regard Microsoft as every bit as promising as a long-term dividend payer as Intel.
Microsoft itself is a puzzle. It has never been a high-concept company. Instead, it has taken routes that others could have or would have taken and dominated markets with its less-than-amazing products through aggressive marketing and its very successful collaboration with Intel. Suddenly, it needs an amazing product, something it has never produced before.
The problem with the possible remaining markets is that none of them seems to offer the kind of margins that Intel has become accustomed to--even *if* Intel is successful in those markets. The cleverness that has gone into Intel's current generation of high-end processors is simply stunning, but the market that benefits from that cleverness, and the margin that goes with it, is disappearing, unless, as many have noted, Intel comes up with a tablet that is a big success in the market--which brings us back to Microsoft.
Even without a spectacularly successful tablet solution, all is not necessarily lost for Intel. The server market and the HPC (high performance computing) market are both growing. ARM (NASDAQ:ARMH) architecture or Atom-like low-margin chips are going to be the web servers of the future, thus lopping off a big chunk of the server market. The driver is the cost of electricity and of cooling data centers. Low power rules. The entire worldwide HPC market is now about $10 billion, but, even there, ARM and Atom-like processors will become more common, again because of the cost of electricity.
There are markets that Arm and Atom are unlikely to invade any time soon: high-reliability enterprise computing and high-end analytics for business or national security applications. The problem there is that that's a market that IBM (NYSE:IBM) wants, and IBM doesn't even try to make a profit on its processors, because the profit for IBM is in services.
There does not seem to be a middle ground. Either, as I have surmised in technical forums, Intel is end-of-life, or Intel will be an amazingly profitable investment (along with Microsoft) once it presents a product yet to be seen.
A final note (or face-saving cop-out, if you prefer). Anyone who has been around computers for long enough knows that the business has been almost entirely dominated by black swans, and there is no reason to believe that the future will be any different. The likelihood that anyone's prediction will stand for very long is small. In the meanwhile, stop thinking of Intel as a safe bet and look closely at the products that use its chips and what kind of value proposition those products can offer to potential consumers.