Skyworks Solutions, Inc. (SWKS) Earnings Call Transcript & Summary

Good afternoon, everybody. My name is Gary Mobley. I'm the publishing analyst on Skyworks, and 1 of 3 publishing analysts at Wells Fargo Securities covering the semiconductor industry. We have with us this afternoon the management team from Skyworks, including Kris Sennesael, the Chief Financial Officer; Carlos Bori, who heads up marketing and sales for Skyworks; and we have Mitch Haws as well, who heads up Investor Relations for the company. And so given a limited amount of time, let's go ahead and get right into it. And I wanted to ask you, Kris, first, if you can just give us a brief overview of Skyworks' business, for those who aren't as well versed and haven't really followed the story closely lately. And then from there, we can dig into more detail.

Yes. Great, Gary, and thanks for hosting us here at the virtual Wells Fargo conference. And well, I'm really, really happy, and I feel really good about where Skyworks' business is right now. As you know, we have been doing this for more than 20 years, and we have been -- over those -- that time frame, we have been building a very strong technology leadership position, right, with our very complex but highly integrated RF solutions that we empower most of the smartphones in the world as well as billions and billions of other devices that need a strong, powerful wireless connectivity. We've been advancing the technology. We've been improving the performance of it as we transitioned from 2G to 3G to 4G. And now finally, after a couple of years of intense investments, 5G is finally happening. And again, thanks to our strong leadership position, we're very well positioned there to take that next wave and drive a lot of top and bottom line improvements. 5G, it's a great story from a content uplift. It drives a lot of complexity, and it could potentially also be a great story from a unit point of view. In mobile, which is roughly 70% of our business, we have very strong customer engagements with all the major players, the big 6, as well as the smaller players there. And we can talk a little bit more about that, but that's where we see a lot of the action, the initial action on 5G right now. In addition to that, roughly 30% of our business is in broad markets, in which we have wireless connectivity in IoT, which could be 5G or 4G cellular connectivity as well as WiFi and bluetooth. We provide the RF around that. We have a wireless or wired cognitive audio business that's really on fire, and we play a little bit in the infrastructure market as well. As I said, we reported our September quarter. We were up 16% year-over-year. The mobile business was up 20%, broad markets, double digit. We were up 30% sequentially, and there's a lot of tailwinds here that continue to drive into the December quarter and beyond.

That's a good segue into my next question. As I think about the backdrop for your mobile business, which, as you pointed out, is 70% of the total revenue, the mobile phone market overall, in unit values, may decline roughly 15% this calendar year. However, if I contrast that against your revenue for the mobile business unit, it appears as though you might be on pace to grow high single-digit percent in revenue terms. And so what explains that difference, that divergence? What has allowed you as well as your main competitor, Qorvo, which is also generating similar type growth, to outperform the overall mobile handset market? Is it the 5G content growth story that's driving that?

I'll take that, Gary. Thanks again for having us this morning. And that's a good question. I would say that it does sort of boil down to a common denominator in content. Content's up across all of our mobile customers and has been growing consistently. Complexity is intensifying, particularly in 5G. The user experience, to a large degree, is defined by form factor, speed, latency and power consumption, all of which are highly dependent on RF front-end sophistication. We've grown content every generation with our largest customers, experienced a big step-up with our China customers in 5G as well as they move from discrete to integrated architecture. So yes, in sum, it's a matter of content. It's just -- it's continuing to step up.

Sure. And on that point, you talked about some Chinese OEMs moving from discrete solutions to integrated modules. And I think you might be referencing Oppo and Vivo, which collectively have about 15% market share. Those were some of the last few China handset OEMs to move to an integrated solution. So is that on top of the strength that you've had from your lead customer? Has that been a big contributor to your growth this year as well?

I would say Oppo, Vivo are both up, but we're also having -- and we're also seeing significant content increases in Xiaomi, which is another significant unit player in China. But we're also seeing the available opportunity grow at other customers like Samsung, as an example. I would -- I'd put it this way. If you're going to be in the 5G game and you're going to be successful, there's going to be significant RF front-end content opportunity growth.

Got you. And with the iPhone 12 being available in the marketplace, we see the typical teardowns of the device and tech insights did a pretty good teardown. And it highlights the fact that the iPhone 12 has support for 50 frequency bands, reports that RFFE bill-of-material increases from $33 in the iPhone 11 to $45 in the 12. And that's not even including any contribution from millimeter wave. I understand Apple has always represented the extreme case in terms of frequency band support as they attack the market with sort of a 1 SKU on a global mode approach. But can you speak to the overall trends in the number of frequency bands required in 5G phones? And how this drives demand for more complex RFFE?

Sure, sure. There's a lot of specific information in that TechInsights Teardown report. We don't provide and we don't speak in a lot of detail with respect to our largest customer. But I'll point out some other facts and statistics that may help provide some context. We grew our mobile revenue 30% sequentially into the September quarter. We've added over $300 million of revenue since the June quarter as a matter of fact. So it's clear that content is increasing across all mobile customers, and Skyworks is a major beneficiary of that. In terms of how many 5G bands there were in an iPhone 12, I think -- again, I think that Teardown from TechInsights was on point. The key takeaway being every 5G band that's in an iPhone is incremental to all the 4G LTE advanced pro bands that were there in the generation before. And I think as we've seen in 4G year-over-year, we're going to continue to see incremental 5G bands in their road map for years to come. This is -- remember, this is year 1 of 5G.

Sure. And Kris, maybe you're the guy to answer this question. But per your recent 10-K filing, you have 1 single customer now representing 56% of your total revenue. And per recent teardown, it shows your content growing at this customer from 6 sockets to 10 in the newest version of the phone versus the prior version. And so if this translates into roughly a 30% increase in bill-of-material increase for you guys, and assuming that customer shifts eventually all of its products over to 5G SKUs in terms of sales, doing the math, I mean, that translates into a 40% increase in your sales to this 1 customer. Am I oversimplifying things? Are there other more complex considerations, such as predetermined pricing, step-downs or whatnot?

No. I think you think about it the right way, I would say so. I mean, there's a lot of information in the teardowns. Actually, we do have a slightly higher number of sockets in the phone, but some of them, you can't really see because they are embedded in some of those other modules. But directionally, you think about it the right way. Of course, a lot of it will depend on the units, right? And the units, it's not something that we control, definitely. So there seems to be a lot of excitement about the new 5G phone that came out in part because of the 5G functionality, but also it's a really good-looking phone. Have a great camera and has a lot of features that are really appealing to the consumers. When you do the math with a large customer, of course, they continue to purchase from us the legacy phones, right? The iPhone 11 has not been discontinued and some of those other versions as well. And so there is some ASP erosion on some of the legacy parts that needs to be taken into account. But when you put it all together, yes, there is -- I mean, it's a large customer for us. It's a big part of our overall revenue. But it's definitely going and trending in the right direction.

Sure. To what extent does -- do some of your larger customers influenced your Sky5 road map as well as capacity considerations in Mexicali to believe the vast majority of your products eventually run through and then as well, Singapore and Japan?

Yes. And so -- and we talked a little bit about that. Even if you look at the September quarter, right, we are somewhat supply constrained. So we could have -- if we were able to ship more in the September quarter, we could have shipped more. And the whole -- even if you look at December and potentially even going a little bit into next year, it will remain supply constrained. And so that's definitely the case for Skyworks. It's also with some of our peers and competitors that play into this ecosystem. And part of the reason there is that normally, when the large customer goals, the other customers lay low, but not this year. This year, the other customers who actually came out with their 5G phones prior to the large customer, they have so much positive momentum. They continue to go with that. So that puts a lot of stress on our supply chain. And we have responded to that by continuously expanding our capacity, expanding the capacity in our front-end power amplifier fabs, continue to expand the capacity in the filter operation, although I think we're feeling good about what we have been doing there with some level loading of the factories. But then you get all towards the bottleneck in our back-end facility. And there, we have been expanding the capacity as fast as we can and continue to do so as well as switching on some external capacity if and when available for some of the lower complexity parts. And again, we keep working that as hard as we can because that's the bottleneck right now.

Got you. Okay. Staying on the topic of current events. There was an article last week in DIGITIMES talking about Apple potentially in-sourcing its RFFE needs. And this just didn't seem right to me. So I'm wondering, without specifically talking about Apple, talk about why a smartphone OEM would not want to in-source this or why they couldn't potentially because of the barriers to...

Right. And so that question pops up like every year or every other year. So -- and again, what we do is very specialized and very complex, right? And I talked about that. So you need to have the right technology building blocks, gallium arsenide PAs, filter technology, TC-SAW and BAW. And then of course, we purchase a lot of materials from third-parties, TSMC or GlobalFoundries or Tower Jazz and PCBs and passive components. And then you bring that all together and do some very complex packaging, assembly, double-sided molded BGA and then RF testing. You need to be able to do that at scale, meaning ramping from 0 to 200 million units every year per device, right? And so you multiply that by 10 or 20 or 30 devices, and so you talk about billions and billions of devices. Being able to do that, controlling your quality, your manufacturing, it's really hard to do. And there is no foundry model, right? If we had a chip, a digital chip that can be farmed out to TSMC and Amkor on the back end, I would be worried. But again, what we do is very complex, very unique and it does not exist in the foundry model. I think the DIGITIMES article was maybe a little bit confusing in a sense that when you look at the iPhone 12, there is an antenna package for millimeter wave, relatively simple, where certain components were bought, including the RF, was bought from third parties and was then assembled in an antenna package. But that's far from doing a low-band path, or doing a mid -- high-band path or doing some of the complex RF that we do.

Okay. Point taken. Carlos, I want to bring you back into the conversation.

Sure.

And talk about product cycles within product cycles. Many investors think the 5G product cycle is being somewhat discrete. In other words, once the SKU cuts over to 5G, that's all the content gains we'll see in that SKU until 6G. But could you speak to sort of the complexities of the different cycles, adding support for more frequency bands within the 5G product cycle that can keep this 5G product cycle alive and well and fresh over the next couple of years?

Yes. Sure. Okay. So yes, certainly not a discrete once over cutover from 4G to 5G. I think the way to interpret or to view this 5G growth cycle over the next 5 years, I think looking at the 4G LTE cycle is probably a good reference point, a good proxy for the discussion. Content grew with each generation as you added 4G LTE bands. Complexity increased with every band. And the user experience demanded it. And then we went from more bands to carrier aggregation because the user experience demanded that, too. And now we're at a point where there's upwards of 10, 20, 30 4G LTE bands that are being carrier aggregated in different combinations from low, low bands to low and mid-high bands together and then mid and mid bands. And then on top of that, you added MIMO capability to these phones. Much, much more complex, much higher performance 4G phone you get today than you did 3 or 4 years ago even. I think you're going to see the same thing with 5G. And I think it's important to note that every 5G band that's on the phone today is incremental complexity in and of itself because that 5G phone has all the old 4G LTE bands, all the 4G LTE carrier aggregation, all the 4G LTE MIMO as well. So I think you've got a complex situation in 4G LTE advanced Pro, and you've seen that grow over 3 or 4 years. I think you're going to stack on top of that a similar growth and complexity story with 5G. And I'm really excited about it because what it really means is the manufacturer that can hit the successive technology nodes within that 5-year period in gallium arsenide, in silicon switching, in miniature packaging, in testing, in scale, in switching and low noise amplifiers, in all the different types of materials and technology that we pack in these very small front-end modules, I think the wind is going to be behind their back-end. Like I said, we're really excited about it. We're looking forward to it.

Great. I wanted to shift gears to a discussion about the competitive environment. And no doubt, Qualcomm's entry into the RFFE market has been probably one of the biggest changes in the competitive environment over the last 5 years. Is it fair to generalize that Qualcomm's success in RFFE will be highly dependent on Qualcomm's overall share in basebands? And would it be optimal for Skyworks if Qualcomm are to lose share to cap the baseband solution or some merchant solution like MediaTek or Unisoc?

Kris, do you want to take that? Or do you want me to...

Carlos, you go.

Yes. So we get this question a lot about the Qualcomm's RF front-end solution. I would say, first of all, Qualcomm is a great company. They've got great technology. I'm really impressed with how they've been out front and led the 5G global transition from a baseband perspective. They're phenomenal proselytizing the technology. They've done a good job with millimeter wave as well. So nothing but great things to say about Qualcomm. All that said, I understand why they want to expand into the lucrative RF front-end market, which is good today, but growing quickly. And I would just say this. I, from a sales and marketing perspective, view them as another competitor. And they're going to do what they're going to do and they're going to win share on some of their baseband sockets and phones where they have the baseband and the applications processor, but so are we. Right now, we have a very, very good attach rate on 5G phones, leading 5G phones from the top OEMs around the world that use Qualcomm basebands and application processors. And we plan to continue to be the #1 supplier for Qualcomm basebands. And again, the fact that they're out early with 5G, the fact that they're out early with not just sub-6 gigahertz 5G, but millimeter wave as well makes those phone sockets, those RF front-end architectures in those phones, like it's extremely complex, which is where we shine. So I would say, I'm not excluding any part of the pie. We've always done a good job on RF front ends where a Qualcomm was the incumbent on modems and AP. And our plan is to continue to win in those sockets in the future. And of course, the plan with MTK and Samsung baseband is the same as it has always been. We're baseband-agnostic. It's a key advantage. Our front ends work across all these platforms. And really, at the end of the day, it's a complexity play.

Yes. And looking specifically at sort of the China market, clearly, China has a desire to foster domestic chip industry. And obviously, I guess, roughly half of the 250 million 5G handsets may be sold in China this year. And with that as a backdrop, what are you seeing from a competitive dynamic on the RFFE front in China, looking at some of the smaller startups. I guess Maxim actually isn't a small start up, but maybe Maxim and some other start-ups as well as getting more dynamic there?

We see Maxim you bring up, we see point product solutions from companies like Maxim, and they're starting to win some lower end sockets on more disintegrated architectures. It's a different game. It's a different area that we're playing in. We're building 10 billion filters each year. We're doing 10 million multichip modules in Mexicali every day. It's not a discrete system or point product play at all. It's a very complex system in a chip. And it's taken us a long time to get here. Years of CapEx investment, years of OpEx investment, years of institutional know how that's difficult to document in spec sheets. I think if they put their mind to it and they want to get there, I wouldn't count them out, but I think it's going to take years and billions of dollars to replicate what we have built here at Skyworks. And then, of course, remember that we're not standing still, right? We're moving and improving and innovating at a very fast pace.

Sure. Okay. Switching gears completely to the broad markets business, you are now seeing that business accelerate. I believe the guidance for the December quarter was $300 million, that's roughly 15% year-over-year growth, and that's following, I think, a prolonged period of flattish performance. What's driving that resurgence and the reacceleration across IoT? Is it as well the new audio business? Is it wireless infrastructure, et cetera?

Yes. Okay. So I can answer that in many different ways. I guess, I'd start by saying that going into this pandemic, it was difficult for us to gauge whether our business would be up, flat or down. And as it turns out, work from home has been a powerful driver for us. Everything from the residential gateway to WiFi extenders, security cameras, iPads, wireless speakers, iPods, Amazon Echos, LED lightbulbs, almost everything is up in units from our perspective. In gaming, we're changing the way that gamers play. We're enabling them to cut the wire with a low latency bulletproof connection. OR the way telemedicine is delivered is changing, and the bandwidth required to deliver that is surging. And then with the advent of 5G, the low latency, the massive machine type communications, the super high bandwidth, the ultrareliable connections, it just gets extremely exciting. We're going to be connecting cars, factories, huge dense sensor nodes in smart cities, in smart education. And all that's just now getting started. So I feel like we've been building towards this for years, if not decades. We're finally at the cusp of this huge transition. And the vision of Skyworks, which is connecting everything and everyone all the time, it feels like with 5G, we're on our way in a material, significant way that we really haven't seen before. And through it all, complexity, miniaturization, ultimately, technology innovation, hitting those nodes, miniaturizing everything, that's going to be what differentiates the winners from the losers and again we can't wait for it to start gaining even more momentum.

Okay. Appreciate that. I wanted to, I guess, go back to the topic of current events and ask about the limited license that you have to ship to Huawei. Qualcomm subsequently got a license, I guess, related specifically to the 4G business. Do you really think that makes a difference in Huawei's supply chain at the end of the day? And does it really make a difference to you whether Huawei's mobile handset business survives or whether that -- those 220 million units per year that they have typically sold goes to competitors?

Yes. So Huawei used to be a really big and important customer for us. At a certain point, it was $125 million a quarter, $25 million of that was infrastructure. Unfortunately, that's down to 0. And then of course, we have the mobile business. With the current restrictions, that's actually also down to 0. But it really doesn't change how many phones are being purchased by people in China or anybody in the world, right? Somebody will supply that. And if Huawei can't supply that, then Oppo, Vivo, Xiaomi, or Samsung or Cupertino or any other smaller player will supply that. And we are very well connected with all those players. So it doesn't really matter for us that much who is going to supply into Huawei. We do have a limited license for 4G parts, and so we are trying to figure that out. Maybe there is some upside. You also heard that they have been selling their Honor brand to a different player. And so that's an interesting move. But again, for us, it doesn't really matter that much because we are very well represented with all the players.

Okay. In the last 2 minutes we have, I wanted to ask you, Kris, please tell me you're not going to go out and make a big dilutive acquisition. Given your strong balance sheet and strong cash flow, you're often cited as the most likely buyer in a consolidating semiconductor industry. So I'd love to hear your perspective on how you feel the M&A environment is flowing right now?

Yes. First of all, I mean, again, we feel really good about our organic business, right? We can grow the top line double digit. We can further expand gross margins, operating margins. We have a very strong free cash flow margin and EPS growth. And so we feel good about the organic business. The business generates a ton of cash. All that cash, more than 100% is being returned back to the shareholder. We have our dividend program that we have been improving, and there's further room for improvement as well as our share buyback program. And even today, we continue to be very active. We don't feel we're fully valued at current prices. So that's what we do. In addition to that, yes, we do have on or about $1 billion of cash on the balance sheet and no debt. So we have the optionality to participate in the M&A process. But again, we're going to remain disciplined. There needs to be a clear strategic rationale strengthening our position in that mobile ecosystem or becoming a more diversified company in the IoT connectivity space. But there also needs to be a financial rationale, right? And we are disciplined. We're not going to do anything stupid. We want to see a clear return on investment. And that's how we approach it right now.

Okay. All right. I think we've run out of time. Guys, I appreciate the time that you've given us here today. It's been very helpful. And look forward to future dialogue, and I hope you all enjoy your holidays. Thanks again. Take care.

Thank you very much.

Thanks, Gary.