Analog Devices, Inc. (ADI) Earnings Call Transcript & Summary

Okay. Thank you, everyone, for joining. I'm Vivek Arya. I cover semiconductors and semi cap equipment at BofA Securities. Absolutely delighted to have the team from Analog Devices join us this afternoon. So Prashanth Mahendra-Rajah, the CFO of Analog Devices; and Mike Lucarelli, the Head of Investor Relations. Warm welcome to both. And maybe Prashanth, just to kind of set the stage for the audience, give us kind of your state of the union, what are the top 2 or 3 things that you're finding most interesting and exciting about the industry and Analog Devices right now? And then conversely, what kind of headwinds are you facing that you want investors to be aware of?

Great. Thank you, Vivek, for having us. We're delighted to join you again for the BofA Securities Conference. We just recently finished our second quarter earnings report, and we reported a pretty fantastic quarter, record revenue, record operating margin dollars, record EPS. And on a trailing 12-month basis, we also hit record free cash flow as a percentage of revenue. So the business is doing quite well. We are seeing broad-based demand across all end markets, and we really feel that we're in this -- in the beginning stages of a solid economic recovery. We are also waiting on the final approval for our acquisition of Maxim. We've received regulatory clearance from all jurisdictions. We are waiting on China, and we look forward to that coming in due course over the coming months. From a challenges and headwinds, I think ADI, similar to many of our peers, is trying to find the right balance here of the significant demand and how we can adjust supply within a reasonable time frame to meet that strong demand. I would say that our view is that it really comes from the secular tailwinds that we've talked about in our business for quite a while, and that, in conjunction with sort of a very broad-based kind of business for a while, and that recovery is creating this really fantastic golden age for semiconductors and analog, the analog industry in particular.

Great. Thank you, Prashanth, for that introduction. Can you -- maybe let's start with the supply side and then we'll go to the demand side. So on the supply side, which specific areas are you seeing the shortages? Are they more in the front end, in the back end? And what steps are you taking to help address these shortages?

Yes. So Vivek, we were proactive in -- towards the end of our prior fiscal year to make some decisions on bringing capacity onboard. So we placed orders for equipment to increase our ability to supply the market in anticipation of the economy beginning to recover with the availability of the vaccine. That's what has allowed us to really book a record second quarter, and it's allowed us to guide up sequentially for the third quarter. So you see the benefits of that coming in. The investment we're making, while you see them, a significant increase versus what we were spending in the second half of 2020, it's still at a relatively modest pace. And we're spending -- this year, we'll probably be at around 5% of revenue, and our typical model is around 4%. Now revenue is a little elevated today. So on a dollar basis, that's a bit more meaningful. But this is the right rate of spend for us to bring in that capacity. And we're focused really on test because for us, test gives us maximum flexibility that as the market evolves or as demand may mature and shift in the future, those tools can be used in multiple end markets and multiple applications. So for us, it is equipment that we know we can get great leverage off of as we go forward. The first part of your question, where are we seeing the challenges? At this stage, it's very broad-based. I mean we are seeing demand strong across all of our end markets, and therefore, the constraints that we're feeling are really across the end markets and that reflects both what we manufacture internally as well as the foundry work that we rely on from our strategic partners.

Got it. And Prashanth, what are you doing to make sure that the demand signals that you're seeing are real, but how are you scrubbing for real demand versus overbooking at this point?

That is a -- it's an excellent question. I think the whole industry is literally dealing with them. It's an area that we have been putting a significant amount of energy into. So I guess, I'll highlight a few things. First, we have a very strong communications link with our customers, which help us understand where is their need for inventory versus where is their need for something that's going to go right on the production floor. And we are working with our customers to prioritize delivery of products to make sure that we are giving the highest priority to products that are needed to immediately hit their production line. The second is that while we use distribution for about half of our business, we use signals through distribution and actually run the business on a sell-through basis. So we have intelligence coming from our distribution partners that helps us understand what they're seeing at the end market. We've recently increased our visibility by asking our customers to stretch to a 90-day noncancelable, nonreturn, and that's given us a little more visibility to make sure that the manufacturing activities are really focused on what is most important. And then beyond that, obviously, we're looking at inventory levels, both at our distribution partners, but also what our sales force is able to advise us in terms of where customer inventory levels are. So it's a multifaceted data elements are coming from lots of places. We are mindful that there is certainly going to be some level of excess ordering, but our role is to make sure we prioritize the production needs for customers to manufacture versus customers who might be ordering just-in-case inventory.

Yes, I'll repeat what Prashanth said back is, you're right, Prashanth said, orders are very strong. When you look at our revenue, what we're shipping, if you look at our revenue from our previous peak to the current peak, for the total company of ADI, our 3-year CAGR from '18 to '21 is about 5%. And I think to me, 5% is kind of a solid long-term growth rate for the company. So revenues are really not -- really elevated. They're really right on trend line. I think, as Prashanth said, order is strong, but revenues are in line with a long-term trend line.

Got it. The one interesting thing, Prashanth, we have observed during these times is just the way different companies are kind of putting in place preferred supply programs or the pricing aspect. And it seems like there is actually a difference. Some of your peers on the microcontroller side, where perhaps there is a little more customization of the product, they are putting in place these restrictions on a longer-term basis. But as you just mentioned, right, you are -- your products tend to have a much longer, right, shelf life, there's a lot more general catalog type of products. So what we have seen from the analog industry is actually being less restrictive, right? I mean it is still more than what you were doing before. So can you talk to us, first of all, is this the right read of the industry? Why is it only 90 days? Why not longer for you?

It's a great question, and I think each company is going to find the answer that works for their products and for their customer base. So I guess a few comments that are important. First, we have a very long tail of small customers. And the ability for customers, who are smaller in size, but represent a -- collectively represent a significant portion of ADI's revenue base, it is difficult for them to be able to work with us on a direct basis or through distribution to make those type of longer-term commitments. It's just not -- it's not something that's in their scope of capabilities. For our larger customers, and I would really call out 2 verticals, in particular, the communications group as well as the automotive group. We know those customers very well, we know their end markets. They don't have the visibility to put that kind of long-term demand on it. So it's a bit of an unfair ask to have a automotive or communications customer to give us sort of a 12-month out view. Will they do it? If we ask them to, but we know that, that's likely going to end up in them ending up on the side of taking the pay versus take in a take-or-pay situation. So we work with all of those customers closely. We are making sure that they get what they need to keep the production lines running, and we're mindful that they don't have that visibility right now. And so asking for that isn't really helpful to them, and it doesn't reflect kind of the partnership that we have built with our customers to work in a very collaborative manner.

Got it. Do you think that impacts you financially in any way? Because there's always a trade-off, right? You can try and work towards a forecast, but then maybe there is a risk to that forecast versus being more flexible and putting your internal safeguards in place. What is the right way? And how did you come to, okay, 90 days is the right number, and it is not 60 days versus being 120 days?

For us, it is the visibility needed for us to fine-tune the last stages of the production process, right? So the -- 90 days is not going to affect what we're running through foundry and the die bank that we're building. 90 days will have limited impact on assembly, but it helps us manage the tail end of the assembly and the test to ensure that as the final specialization of the products is happening to go to customers, we have that well aligned with what customers need to -- it really is -- it helps at the fringe. It is not a meaningful change. Because when you're in a capacity-constrained environment, whether you have perfect forecast for the next 6 months, 12 months or 5 years, you're still capacity constrained. So it is optimizing at the fringe.

Got it. So if, let's say, Vince or yourself are talking with a large industrial customer or a large automotive customer and they ask you, guys, when will this be over? When will I be able to just call you and order within the lead times that I was used to in the past? Would you say that that happens middle of '22, late '22, early '22, what is your crystal ball saying as to when the industry gets back to trend?

Yes. If it was auto, in particular, I think the challenge that industry has is how successful will this ramp in electric vehicles be? I think that it's -- there appears to be very strong interest. And I think most of us are a bit surprised at just how rapidly the adoption of EVs is proceeding. EVs are great for ADI. We do EVs, it's proceeding, we have more content in electric vehicle. So what you're seeing is really the secular demand to consumers having a preference for electric vehicle, which is driving more ADI content. When does supply and demand get back in balance? I don't know that I want to make a prediction on that, except to say that it is certainly not going to be this year, and we're optimistic that it's going to be next year, but I think that really also depends just on how the introduction of the number of electric vehicles that's coming out stabilizes and how and when we see, which ones do well in the market versus don't. And that's just speaking to one particular end market. I could make a similar comment around all of our end markets because we're seeing, again, very broad-based demand. All markets are really growing at a significant pace. And our book-to-bill is above 1, and I don't see that changing in the near future.

Got it. So now let's step through the demand side and your various segments, right? Industrial comms, auto, consumer, right, in that order. So on the industrial side, right. ADI has the most exposure to that market. What is the -- is there an easy way to segment that industrial business in terms of what is secular versus what is cyclical? Because when I look at the growth of that business right now, right, 20%, 30% kind of growth rate, I imagine that, that's well above the trend line. So it obviously has a lot of cyclical lift to it. So how do you think about your industrial business, where it is today? And how do you expect it to grow over the next 3 to 4 years?

Yes, definitely. Well, let's do this. Let's -- our industrials are relatively big basket. So Mike, why don't you break down industrial into the relative components of it and then we can talk about how each of those segments is doing.

Sure. So you're right. Industrial is about 60% of revenue right now. The biggest portion is instrumentation and test, it's about 30% of industrial. The next biggest is factory automation, about 20%. Aerospace and defense is about 20% also, that includes space and avionics and also defense applications. Health care is about 15%, 1-5 percent, of industrial. I'll call energy, just below 10%. I mean, when you think of energy, think of renewables, distribution and monitoring energy, don't think about oil and gas. It's really more about clean energy. And then the rest is what we call broad market, kind of the long tail of the business.

Now all of those markets have some level of correlation to GDP. And when we are in a global GDP expansionary environment like we are today, we are in a global GDP tailwind. But all of those markets also have specific secular drivers. And not to go through all of them now, but let's just pick on 2 as an example. In instrumentation and tests, as we continue to see more and more electronic content being deployed globally, all of that content has to be tested, whether it is your 5G phone, whether it is your memory chips for PCs or for cloud storage, all that testing is putting demand on that instrumentation and test business. So both GDP growth as well as secular driver. And the other one, this next one Mike talked about was factory automation. Again, we are continuing to see an interest and a movement to onshore manufacturing. The Biden administration has been very clear about what they're looking to do to help onshore manufacturing in a number of key technology areas and that onshoring of manufacturing requires you to solve for the labor arbitrage -- the negative labor arbitrage that you get when you bring something back to the U.S. or to Europe and you don't take advantage of the more affordable labor that's in Asia. And to adapt for that, you need more automation. And that feeds right into our broad suite of products, whether it is in safety, it is in precision control, communication of your factory equipment. So again, secular drivers that are driven by the onshoring as well as just manufacturing capacity build out.

Got it. So how sensitive are the sales in that segment to just the fluctuations we see in industrial PMIs as an example, right? Because as you mentioned, Prashanth, right, and Mike that this is such a broad segment that for us to track it, right, I mean, even on an annual basis, right, gets tougher. So how do you make the projection, Prashanth? Let's say, if one had to ask you, what do you think your industrial business will grow in the next 3 years, what would be kind of the process of predicting that growth?

For an investor, I would tell you to model a multiple of GDP, which reflects both the GDP growth plus the content expansion that comes with that. Internally, we look at each of those markets individually, and then we have our own models as well as customer inputs. We have a -- we've got a very good pipeline database that also helps us track our wins and then be able to track those wins from when are we awarded it to when does it go to production, to when does that production hit peak production. And for many of the industrial segments, that tends to be a longer cycle. So much of the socket wins that are converting to revenue today were positions that we won a few years ago, which gives us even more optimism about the future. Because we have high conviction that our win rates with the addition of Linear have really improved, and the benefit of that from a revenue growth standpoint is still in front of us regardless of how the market does.

Got it. Are there certain segments within industrial where you see -- like what are the top 1 or 2 segments where you see the most opportunity for content expansion? Because I think in the auto and the consumer side, it is somewhat easier to visualize the units and content story. Are there parts of industrial -- like over the next 2, 3 years, which parts of the industrial business do you think can grow the fastest for you?

I would say energy. Mike, I don't know if you want to add another one, but I would certainly say the energy segment. Because of the significant investments that are being made across the globe in being more conscious in energy use and the movement away from fossil fuels, our energy segment helps our customers optimize the use of those technologies. We've talked about our ESS product, which is doing extraordinarily well now and that is being used in the deployment of the infrastructure for vehicle charging. We have a phenomenal share position there among our customers in terms of their preferred design solutions to build out that charging infrastructure. So I think energy probably has the greatest secular tailwind. And health care is the other one that I think really comes to mind for me that I think the pandemic has shown us that you can do a lot more with your position or your practitioner on a virtual basis. And now the ability to measure vital signs at kind of clinical-grade quality on a remote basis is an opportunity that is -- that hasn't fully come to its realization and the activity that we have underway with both start-ups as well as established players tell us that everybody is racing to find the right solutions to be able to take your traditional physician visit or hospital visit and move it to a remote operation, the expression Vince likes to use is delivering health care from anywhere.

Got it. And then the last one, Prashanth. Again, when we look at the industrial segment, you have another very large competitor on the analog side, right, who has access to a lot of 300-millimeter capacity, right? They also appear to have a portfolio with microcontrollers and other products. How frequently do you kind of -- is there a change in kind of winning and losing sockets on the industrial side, right? You mentioned that a lot of the growth that we are seeing now are sockets you won a few years ago. How frequently do these things happen? And does having a lot of 300-millimeter capability or having microcontrollers to kind of bundle as part of the solution, are those decision criteria by the customer? Or how do you make sure you retain and win more share in the industrial market?

How we distinguish ourselves with our customers in this industrial segment really is in the ability to provide them the advice and the solution set. It is really that close collaboration, either directly or through the engineering resources that we have at our distribution partner. Remember that for industrial, that is really the long tail of our business. It is, by definition, our industrial customers tend to be more of the -- when I talk about 125,000 customers, they really are that 100,000-plus list of customers. So these tend to be smaller customers who do not have the same engineering capabilities as a larger customer in a vertical like automotive or communications. So having access to ADI's incredible FAE resources, field application engineer resources, having access to ADI's technical support to help them with those design decisions, is really what allows us to keep winning more than our fair share. And once you're designed in, it's not really in their interest to design us out for kind of marginal cost benefits. Because the engineering effort that went in to make sure it worked, doesn't -- it's hard to amortize that over a relatively minor change in price, given the amount of other effort that these customers need and they're incurring labor on.

Got it. Makes sense. Moving Prashanth, to the communications and the 5G side, right, one thing you mentioned in the earnings call recently that you are kind of less exposed to the China 5G market. I believe you said that China is only about a quarter or so of the 5G business. Talk to us about the visibility and the engagements that you're seeing on the 5G side with the U.S. and European customers. Because I think in China, we always see deployments happening in a very big way and then they kind of pause and stop. In the U.S. and European theaters, they seem to be a lot more success based and more gradual and hence, spread over a longer period of time. So talk to us about the visibility that you see in terms of 5G radios where ADI is the leader on the analog front end side.

Right. So maybe if we start at the top, let's remind the audience that our role in communications, particularly in 5G, is the tremendous technology we bring in really the transceiver. We have an incredibly strong position among all the infrastructure players. And what we have seen in the last couple of years is we've seen and enjoyed the benefit of that deployment in China. As that has begun to taper down as well as some of the regulatory restrictions that are in place, we are now looking at a business that is really ex China. And if China continues to deploy in the coming year or 2, then that will be upside for us. But our guide now is based primarily on the activity that we're seeing in Europe and in the U.S. The U.S. completed the C-band auction. So now we are seeing the U.S. manufacturers go into the procurement phase. So at the end of the first quarter earnings call, I would not have been so bullish to say we have orders on the books for U.S. demand. At the end of the second quarter earnings call, we're now able to say we're seeing that order activity start to come in. So U.S. is clearly moving into a deployment phase. We've got a pilot in India that we are active in. So we know that's going to move into its production phase as well. And Europe looks like it's likely to be probably the first calendar half of 2022, where we'll see more of that activity. So our position is #1 position. So as long as 5G deploys, ADI is going to continue to participate in it. And now we see the activity actually beginning to roll out. What we have not begun to enjoy yet is our -- the strength of our position in O-RAN. So I think that you'll continue to see some additional tailwinds for us as O-RAN begins to get more traction. I think we did talk about the win that we've had with Rakuten in Japan. And then even further out, I think 5G is going to go to the private network solution, and it's going to be used in private communications, manufacturing and such. And again, we're going to continue to play well there. So I expect the 5G to be longer in duration than we've probably seen in previous generations. And given the strength of our position, it's going to be a great run for ADI for several years to come.

Got it. What is the right way to think about growth in your communications business? Because when 5G started, there was a lot of excitement that content could be 3 or 4x of what you're seeing in 4G. But when we look at the growth rates on the infrastructure side, they have been somewhat more modest than that. And we thought, okay, maybe it is because 4G is rolling over, and 5G is still too small. So what is the mix of 5G within the communications business? And when will we start to see the benefit of that big content expansion that we thought would happen with 5G?

Right. So 5G today is about half of our business. But in terms of the content comparison to 4G, I'm going to let Mike use the -- provide a more ambiguous answer.

It's about 4x. I mean when I say 4x content, it's about the radios and then the channel... [Technical Difficulty] and 4x is when you go to 64. So when you go to massive MIMO 64 channel, you need to have a 4x content added because there are more so many channels, you need more transceivers, you need more power. Now deployments are mix, some were 32, some were 64, you're seeing some that are even lower. It's 8 by 8 or 16 by 16. So really, we're not sure how it's going to all unfold from a standpoint of total channel count. But channel count is definitely increasing, and we have high share in that channel count. And like Prashanth said, we're still below 50% of 5G. So you are seeing some of the content benefits today, but they'll ramp up more as you go forward. And what we said, I think a couple of earnings calls ago is long-term growth rate for comms is probably high single digits. If you look at the 4G cycle, it was kind of mid-single digits. So you're adding some additional content on top of it, given our share position and new channel count.

I don't want the message to get lost that we have high conviction that the duration of the 5G cycle will be much longer than any of the previous 5G cycle will be much build-out, whether it be 3 or 4G. There's a lot more and there's a lot more opportunities for 5G than we've seen. It's not really limited to just mobile communications.

Got it. And then Prashanth, in the few minutes that we have, I definitely wanted to talk about the battery management system, right? Because I think that's one of the more unique aspects of, among several unique aspects of the business. And you recently mentioned, I think, almost a doubling of -- how much is BMS as part of the company or of the auto business? And what does wireless BMS bring to the table? And when will you start to see that become a more tangible part of your automotive segment?

So BMS today represents about 10%, a little over 10% of automotive revenue in the second quarter that actually was 100% growth, which reflects kind of the momentum that we see behind it. We have the #1 share position. If you look at the top 10 electric vehicles, we have more than half of them using our BMS product. We've introduced a new technology, as you mentioned, the wireless BMS, which takes all of the benefit of precision accuracy that our BMS product gives and adds a number of pretty significant enhancements for our customers. One, it reduces pack weight, which in electric vehicle translates into further range. Second, it allows a much more flexible configuration of the pack so that as you design the battery pack for different shaped vehicles, the amount of engineering effort that's required is significantly less. So it allows you to make that investment upfront from an engineering standpoint and then modify as needed for different vehicle shapes and sizes. And the overall kind of how you think about designing the car changes with the features of wireless, such as the ability to -- the ease by which you can do maintenance and repair on batteries because they are wireless and you don't have some of the service costs necessary to disentangle or to remove the pack. We are seeing great adoption. I know we've been public with the General Motors win. They've been a great partner of ours in bringing that to the market with their Ultium battery line. We have another very large auto OE that we have also won on the wireless side. But the traditional BMS is also continuing to win. I think we mentioned in the call that we've now secured Volvo as well. So both products are continuing to take their positions in the market. It's hard for me knowing how compelling the wireless BMS platform is to not see that being the dominant solution in a few years' time. It's just such a compelling return on investment for our customers.

Got it. So what is the right way to think about growth in the automotive segment over the next several years?

Yes, it's a multifaceted answer because it is a -- what do you model the inflection in electric vehicles and each investor has to make their decision on how they feel that penetration curve is going to accelerate. And then you put a multiplier on that because our content in EVs is a multiple of what we have in traditional combustion engines. So the more optimistic an investor is on the growth of EVs, then the stronger the ADI penetration of that market and the portfolio is. How else would you help someone guide -- model that, Mike?

For us, if you look at SAR, if you use base of SAR, production, we should grow at least a couple of points above that over the long term, and as Prashanth a lot is indicated by EVs. As EVs increase more and more, that portion of our portfolio that BMS portion portfolio...

As well as the LTC power portfolio.

And the power portfolio also, that portion should grow kind of a 30%-plus CAGR over the next 5 years, and that's really high share and penetration of EV is continuing.

Got it. So what I'm hearing is your industrial business can grow multiples of GDP. Your automotive business can grow very strongly. Your comms business can grow at a high single digit, but yet the overall company will only grow at mid-single digits. So what am I missing here?

Well, certainly, I think in the near term, we're going to be limited by capacity constraints, [ that's our stats ]. And then I think over the longer term, we are more optimistic that we should be pushing up kind of closer to that higher single-digit growth, I would say, the -- remember that the model is we have a very broad industrial customer base which brings up this profitability. And then we have a vertical customer base which really amps up the growth. And the balance of those 2 is how we bring that investor model together, which is that high single-digit plus growth with 70% plus gross margins and then with very tight control of our OpEx that drives tremendous cash flow leverage.

Got it. And finally, Prashanth, in the minute or so that we have left, talk to us about user free cash flow, so you're resuming buybacks, what is -- how do you decide whether you should be doing accelerated buybacks or kind of more spread out buybacks? How do you make that decision?

So we will plan to have an investor call shortly after the close of the Maxim deal. At that time, we're going to have in excess of $3.5 billion of cash on our balance sheet. And we're going to have an enterprise that's going to be generating $3 billion or so of free cash flow. So at that time, we're going to share more specifically with shareholders how to think about the capital return. But a couple of things I can say. One, we're very comfortable with our leverage ratio, and we see no reason to do further debt reduction. So expect for us to talk to you about how we take that cash war chest and begin to bring that back to you through dividends and through share repurchases.

Excellent. With that, gentlemen, we are at the end of our time. But thank you so much, Prashanth and Mike for joining us this afternoon, sharing your insights about the industry and Analog Devices. Thanks to the audience for joining. If you have any follow ups, please feel free to write, but with that, we'll wrap up the session. Thank you so much again, Prashanth, and Mike. Appreciate it.

Thank you.

Thank you.