It is hard to imagine that a company that beat its earnings expectations two quarters in a row would take so many hits. On Monday I had the opportunity to delve further into why Advanced Micro Devices (NYSE:AMD) was still getting hit, and now I am addressing it:
- Computing solutions fell 15%;
- Fall in gross profit margin to 36% - down 4% over Q2; and
- A 'disappointing' return on equity.
1. Being that Bank of America Merrill Lynch (NYSE:BAC) was the major one who downgraded AMD, I will address this first. Vivek from BOA posed the following query during the Q3 ER call:
"Now that Intel is starting to become more aggressive in the lower priced segment of the market, also in the tablet category with Bay Trail. And these are segments where you have traditionally held a larger market share on a relative basis. So if you are expecting the PC market to decline 10% next year and Intel continues to be aggressive in that lower priced segment of the market, how do you think your PC sales will trend next year? Just conceptually, I understand it's too early to make a specific prediction"
To be frank, her question was answered multiple times throughout the call. It was first addressed when Rory stated:
"We expect PC industry unit shipments will decline approximately 10% this year and by a similar amount in 2014. We have the right product in IT and will continue to compete effectively in this market"
It was answered again by Devinder:
"In the third quarter, we made progress towards our goal of diversifying our product portfolio, as we successfully ramped our semi-custom products. Specifically, our semi-custom and embedded revenue accounted for more than 30% of total company revenue, exceeding the target of 20% of revenue by the fourth quarter of 2013"
In short, AMD said that they were expecting PC sales to decline, and that the other sources of revenue were part of the strategy to hedge against it. Later in the call it was also stated that AMD would eventually be looking to make 50% of its revenue off of its non-PC sales. As per Rory:
"We see that as an opportunity to drive 50% or more of our business over that time horizon. And if you look at the results in the third quarter, we are already seeing the benefits of that opportunity with over 30% of our revenue now coming from semi-custom and our embedded businesses."
The main issue here, is that these analysts cannot wrap their heads around the idea that Rory had already factored their concerns into account. He is aware that the economy is still ailing, that PC sales are slumping, and that Intel remains a powerful opponent. He gave a total expectation for PC sales to decline another 20% - which still puts them above the 50% revenue point they were looking for.
The big killer for AMD (in regards to PC sales), is not the PC market as a whole, but rather their laptop sales. If you factor into account that it was less than a 7% hit, and that other revenue sources more than made up for it, it is little more than a mole hill. Regardless, the Kabini chip has shown promising reviews, and the Tamesh is supposed to eclipse that.
Yet even here there is much more to be said. TheStreet, Barrons, and The Motley Fool all left out Rory's follow up:
"We delivered the second straight quarter of channel revenue growth based on strong demand for our high-end A8 and A10 APUs and FX CPUs. We expect this trend to continue".
The problem with mixing laptop sales and desktop sales together, is that you miss the bigger picture. While laptops are declining in sales due to notebooks and tablets (something AMD accounted for and even stated as much during earnings), desktop sales remain strong. This is largely due to the low cost of a good AMD build.
It should be noted that no further upgrades will be made to the FX line (the current CPU's offered by AMD without integrated graphics). The reason here is fairly simple. Go to newegg.com or tigerdirect.com (or any other PC retailer site), and read the comments on the AMD FX series. You will note that all but the most expensive chip sold with the new Vishera cores have a 5 out of 5 rating. In short - the customers are quite pleased with this line up. I personally know several owners of these chips who swear by them.
So AMD can be pretty confident that the FX line is good as is (at least for a while). There is no need to sink funds upgrading a product that is not only fairly new itself, but has meant with great success. It also means that AMD knows something we do not. So why does AMD continue tearing up their line of APU's as they are being very well received as well?
For starters, the new Richland chips are just updated Trinity chips (both APU's with the Pile driver based architecture). To those who already had the Trinity APU, the +/- 10% gain was really not worth that much as the Trinity is still being very well received. And yet the Richland is still doing well with high user satisfaction.
For the price point of +/-$150, the Richland A10 APU destroyed anything Intel offers at a comparable price on the graphics side. It is most telling that the comments made by Richland owners speak not only to graphics - but overall performance as well. Unless you are a heavy gamer or running very heavy CPU intensive programs (most users do not), there is simply no need for anything more.
This is why AMD is updating the APU side. AMD once again knows that Intel is not a company to take things lying down. They already recognized the superior capabilities of Intel CPU's. To be honest, they have no way to match Intel's capabilities on the current software lines. Knowing the software side will take years to change, they instead chose to focus on improving their APU line-up where they compete best with Intel.
Enter the Kaveri. If AMD hopes to keep the market share of desktops, Kaveri is their last shot at proving their point (near term). If Kaveri is to be successful to its fullest, it must show the following improvements:
- Built in GDDR5 capability. Currently the APU line uses GDDR3, which can cause bottlenecks between the graphics and memory use. You can even add another gigabyte support of GDDR3 to the board, and bottlenecks will still come up. Therefore GDDR5 is a must, and will show that AMD APU graphics are a force to be reckoned with when integrated into the APU's.
- Improved single-thread performance (the ability of a CPU core to access and process the data flow on a per core basis). Even though the APU was built for working best with HUMA, HSA etc., it still needs to be able to handle current software on a competitive scale. To do this, it needs to show a single-thread improvement over Richland of +/- 10%. The addition of useful 'libraries' has already been addressed in terms of the Kaveri, so there is not 'what if' there.
- Be able to support a greater graphics capability. The problem with the APU is that it is limited to low - mid-low graphics capabilities (even with Hybrid-Crossfire). Among most crowds this has been enough - but to a gamer - not so much. If AMD can upgrade the graphics capabilities on the Kaveri as well, it will have truly hit a home run.
Pushing out Richland with such slight increases was a patch, and will not keep buyers happy for much longer. If Kaveri does not show the needed increases, it will hurt AMD more than help it.
Moving right along, the general consensus is that Rory's statement of expected PC sales decline was a big mistake. I have to go the other way.
- Rory knew that Intel is nipping at AMD's heels, and that Intel holds the key in laptop sales (as OEM's have the bad habit of mixing AMD processors with some of the worst hardware available). He is under no delusion how well Intel does in sales there, and factored it into projected sales loss.
- AMD had anticipated this, and over a year ago in many cases had taken steps to ensure other sources of revenue. The idea that this was unexpected or a new issue shows a great amount of lack of research on those who act like it was a driving reason to down-grade the stock.
The issue then arises if AMD can maintain their revenue stream (and increase it) in order to meet their next payment to Global Foundries in early 2014. They also have bond payments coming up in 2014-2015. Those who found the slip in PC sales as an indicator that AMD was relying too much on the newer streams of revenue, find it unlikely that AMD will meet their goal. Add the fact that Intel is releasing a new series of processors that no one takes lightly, and you have a reasonable doubt.
To sum up this point, AMD's likelihood of failure is slim given the facts as a whole. They already took the downsides into account. The analysts about had a bird, but at least AMD was forth-coming. I am going far out on a limb though, and believe that AMD was really giving a worst-case scenario. As to selling out inventory at a discount, I looked over it again and again, and I do not see how that would equate to a loss in this department (unless the price cuts grow to a somewhat disturbingly high level).
2. The fall in gross profit margin also caused the heads of analysts to spin exorcist style. Last quarter, AMD meant what was expected and the analysts threw a fit. Being that the GPM this quarter fell, the reaction was more severe. However once again - this was expected:
"Gross margin was 36%, down 4 percentage points sequentially and in line with our expectations," - Devinder Kumar (AMD CFO)
Devinder goes on to explain the 4% loss here:
"as we grew our semi-custom business, which has lower than corporate average margins but significant revenue and earnings power as volumes ramp. You will recall that the semi-custom NRE operating model drives significantly lower operating expenses for this business, with the majority of its gross margin dollars falling through to operating income."
This is another case of analysts who are fine with the stock price before the earnings call. Yet right on that day that the earnings call is made, it is a huge issue that no one saw coming. I wish I could say I was surprised. I find it beyond suspicious that you have so many known 'downsides' that go unmentioned when the time comes to knock the price down a peg or two. Sadly, it is the shorts that profit from this - with little to no effort.
3. A disappointing equity return? What is that? Ask TheStreet - who made the statement, "The company's weaknesses can be seen in multiple areas, such as its disappointing return on equity and generally high debt management risk." Of course to find out the details you have to buy their report…I am wondering who they manage to fool with that one. Short point as there is no counters given.
So it comes down to a matter of belief. Some argue that once AMD gets another design win, that the major players will not be able to ignore it anymore. This logic is shaky being that all of the gaming consoles win's, Verizon, etc. meant with nothing more than downgrades. Unless Intel comes out and says 'I surrender' (which will never happen), I doubt any win made will meet with analyst approval for another quarter or so.
Another argument I have heard is that when AMD has eliminated its debt, the company will no longer have anything holding it back. This is something I place less into as every company uses debt as leverage at some point - some just manage it better than others. In this case, if AMD did not incur the debt to Global Foundries, they would have died on their 32nm chipsets anyway. It was a play or quit choice, and they chose to play. I say hurray for them. So it comes down to what it has always been and what it always will be.
Intel is the only reason I can think of that should be a worry for AMD investors. They could either out-price AMD, or beat them at their own game. In short, there are a number of ways for Intel to pull the plug on AMD - the question is, can they? I am not so sure. Intel is a smart company, and for the better part of a decade has been playing it smart. They have the know-how, and the cash. If Intel is to out-price AMD would require Intel putting out cheaper chips that have comparable performance to AMD silicone.
Given that Intel lacks AMD's graphics capabilities, I do not see how they could do this on the APU side. The only way for Intel to out-price AMD, they would have to do so on the FX line. Given that the FX chip is highly regarded in the AMD crowd, this would be difficult for Intel to pull off. They could have played price-war with AMD long ago - so I cannot find a reason as to why they would do so now. Who knows, maybe AMD is succeeding in the age old term 'keeping Intel honest' for the first time in ages. I must say that I do not think this will happen to any real extent- no company likes a war of attrition.
AMD is far better off now than it was a year ago. It has proven that it is aware of its strengths and weaknesses, and knows how to turn both to its advantage. Its methods have been most unconventional, but thinking outside the box is the way to go when the box continues to shrink.
AMD has turned profitable, gained new ground in revenues, realizing a multi-year path toward getting the most out of their product, and a strong base around it. While its debt and competition make it an uphill fight, it has successfully gained ground in overcoming these obstacles. AMD shows no reasonable signs that they will slow down - much less stop. By that I mean that the risks discussed are still there, but they are in much better shape to fight through them and continue to spread their wings.
I do not expect the stock to have a huge break out this year. It has the potential to, but after 5 major firms have downgraded it, I do not see that happening. The first breakout occurred when AMD jumped from 2.40-4.60, and for 1 year, I think we cannot expect anything like that for at least another quarter. This will provide more time for the top 'analysts' to catch up to the rest of us. Sure AMD is a speculative play and nothing is set in stone - but I honestly think AMD can and will pull it off. Time will tell.
Now that the proverbial dead horse has been sufficiently flogged, I look forward to hearing more thoughts on the matter.