Microchip Technology Incorporated (MCHP) Earnings Call Transcript & Summary

All right. I think we're on now. So good afternoon, everyone. This is Chris Caso, semiconductor analyst at Raymond James. Welcome to -- continuing our conference here. Our next presentation is Microchip Technology. Joining us from Microchip is Eric Bjornholt. Eric, thanks for joining us today.

Thanks for having me. Pleasure to be here. Hi, everybody.

Well, great. Well, Eric, why don't I open the floor to you for a minute or so to kind of make some opening remarks. We'll do some Q&A. Well, I'll just pause a second and say there is the ability for our audience to ask some questions. You should see a Q&A box in the upper left-hand corner of your screen. If you put a question on there, that question comes to me by e-mail. And hopefully, I will see it. And I will ask the question in the session. If you prefer, you could just e-mail it to me directly, [email protected]. That's [email protected]. So with that, Eric, why don't I let you take it away.

All right. Well, thanks, Chris. I guess, I'll start with the normal safe harbor. If I make any forward-looking projections in this, I refer you to our risk factors that are highlighted in our SEC documents. But glad to be here today. So Microchip most recently updated our long-term business model. And I think that was kind of the big news over the last couple of weeks that Microchip moved its gross margin target from 63% on a non-GAAP basis to 65%, and our operating margin target from 40.5% to 42%. We're continuing to be very focused on reducing the debt on the balance sheet and have paid down almost $3 billion over the last 9 quarters, and that's a continued focus of the company. We've done a number of debt refinancing transactions over the course of calendar '20, and a more recent one over the last 2 weeks that we can talk a little bit about as part of this discussion today. But that's been a primary focus, to get the capital structure right, reduce our exposure to convertible debt and the structure as we delever and move towards an investment-grade company. I think the other thing of note is we, and others in the industry, are facing supply constraints. And we're managing through that the best that we can, but our bookings activity has been extremely strong. It was strong in the September quarter, and it's continued through today. So the interesting thing is the bookings that are coming in today are more strongly being booked into even the June quarter beyond the March quarter. So March backlog is filled in very nicely. We're getting good visibility from customers, but we do have some constraints on the manufacturing side, both internally and from our external partners that has created some challenges, but we're working through those the best that we can. So with that, those opening comments, I'll turn it back to Chris for Q&A.

Well, that's a really good summary of a lot of things that I'm going to ask you, and we're going to go into in more detail. So that's a good preview to keep everyone tuned in for the whole session. But maybe I'll start with one of the other important disclosures that have happened over the past couple of months, which is the retirement of Microchip's CEO, Steve Sanghi, who's been -- it's an understatement to say, he's been really important to Microchip over the years, and he's been replaced by Ganesh Moorthy, who's been his right-hand man for a number of years now. Maybe you can talk about that transition? What it means for Microchip? Anything different that Ganesh may do at the helm?

Okay. All right. Well, good questions. So I want to start by saying Steve is not retiring. Steve is moving into an Executive Chairman role. He is still going to be involved in all strategic decisions of the company beyond what a normal Board member would be. He's planning on coming March, working 3 days a week. And with that, be involved in all the strategic discussions and analysis that we do and leave some of the more day-to-day stuff in terms of customer interactions, supplier interactions, to Ganesh. So Steve is not going away. He tells me he's still going to do investor events and interact with people on this call. So I think that's a good thing and still be a big part of Microchip. But Steve and Ganesh have worked closely together for many years. Ganesh was actually hired by Steve at Intel back in the early 80s. So they've known each other forever. Ganesh has been at Microchip now for -- I think for about 19 years. And ever since we announced the Atmel acquisition, Steve and Ganesh have been running the company jointly. So they've been kind of 2-in-the-box. All the executive staff members have worked for both of them. And so if you ask Ganesh this question, what's going to change under your leadership? And the answer is going to be, nothing in the short term, right? Nothing in the short term. Him and Steve have been jointly involved in setting the strategy that we're on today. And I think the next thing he would say is, things are going to change in the future just because the environment is going to present us with things that are new challenges. And as that arises, Ganesh will work with executive management and the Board, including Steve, to kind of set that direction going forward. But investors shouldn't expect anything different in the short term. Ganesh and Steve see things very much alike, and this is going to be a very smooth transition.

Okay. Well, that's very good. I guess my follow-up question is going to be what's Steve going to do with the other 2 days a week? But I'll leave that question for Steve. I won't put you on the spot on that one. So maybe we could go into some of the commentary that you've had regarding bookings in the near term. And this has been an exceptional year, to say the least, where as we've gone into the second half of the year, I know that Microchip had issued a letter to customers. I guess, it was coming out of the summer, I believe, which was encouraging them to give you some more visibility and backlog at the time they were booking in very short. But what you're saying now is now they're actually giving you much more visibility all the way into June. Maybe you can give your view of why that's the case?

Okay. Right. So we did issue a letter to customers back in early July, and we were getting a lot of very short-term orders, and we were having challenges responding to that due to manufacturing cycle times and lead times being longer than what customers were placing those orders on. And so we asked customers to give us longer visibility, and it worked well. Not all customers respond at once, some wait until they have a problem, and then you got to address it. But that has continued with the strong bookings through the day. And with that, we are seeing bookings being strong across all geographies. It's not 1 geography. We don't feel that it's 1 end market, we think it is very broad-based. And with that, we're getting very good visibility into the March quarter, and the June quarter is getting a large number of bookings into that. And even beyond, in the September and December quarter, we measure our backlog kind of looking 12 months out in time. And obviously, the further you go out in time, the less backlog you have. But I think customers are acutely aware of the supply constraints that they're seeing in the industry when they're purchasing their bill of materials, they're seeing that and making sure that they are protected, securing their spot in the manufacturing line, so Microchip and other suppliers can meet their needs.

Well, maybe if I could dissect that 1 or 2 ways. And again, one reason for the bookings perhaps is because of the supply constraints and customers wanting to get in the queue. But then the other is just demand. And maybe you can kind of separate both of those things about what you're seeing? And the extent to which those bookings are being driven by each one of those factors?

So I do believe that the demand environment is better. We're seeing that. We're guiding here for an up quarter, up 2.5% at the midpoint in December, which is definitely above seasonal. And I think we're seeing things build nicely into March. We haven't given guidance for March yet. Steve did make some comments on our call -- our earnings call about 2021 and having some confidence about that. And that's based on all the things that we've done over the last few years. The indications that we're getting from customers. Our largest end market exposure, as you know, is industrial, which was hit quite hard during the U.S.-China trade war, I'll call it, the trade tensions. And follow that on with the pandemic, we think that there is some pent-up demand. Our customers tend to not want to launch their new products in a difficult environment, and we're seeing those products starting to come to market now and the demand building. So I think our customers are much more positive about how 2021 is looking. Clearly, we still have the pandemic hanging over all of us, but hopefully, 2020 is -- gets behind us here soon, and we can move on to a more positive 2021. And we're seeing that in our interactions with our customers that they are viewing the environment as more positive than what we've seen in quite some time.

That's very encouraging. And how are you adjusting your manufacturing to support this? Because I think one of the comments you made, and you referred to it as unsupported backlog, meaning orders that you have that you're not able to ship because of your own supply constraints. What steps are you taking to mitigate that? And then perhaps drive a little more revenue growth.

Right. So from a manufacturing footprint perspective, we do about 40% of our wafer fab in-house, about 45% of our assembly and about 54% of our tests. So on what we can control, we are making investments in manufacturing, we're hiring people, ramping up the factories. One of the impacts we've had to our gross margins, although they've held up incredibly well during the down cycle, is we've had underutilization charges coming out of our factories, and that was about $12 million in the September quarter. That's coming down in the December quarter pretty significantly. And really, in March, we expect them to be pretty much eliminated. So that's a short-term driver of gross margin, and we're ramping the factories accordingly. So we laid out some specifics when we announced our new margin target on the gross margin side, in terms of the underutilization charges, that's just going to happen very quickly. Those will go away. But we've set some targets now externally in terms of where we think the internalization can go on the assembly and test side. Do we think that we can take assembly to over 60% over time and final test to over 70% over time? And these all take incremental investments, time from our operations team and our product lines to make sure that, that happens effectively, and we can support our customers. But these are all incremental drivers of gross margin and also allow us to be more in control of our own destiny when it comes to producing the product and getting it to our customers on time. We're also taking steps in our own wafer fabs to increase output there where we can. Again, to some degree, we are people constrained. So we're hiring direct labor, getting them trained and up to speed, so we can start more wafers in the factory and then also making incremental capacity additions there as we can. So we're really focused on the operations side right now in terms of what we can produce internally, working with our suppliers of foundry and assembly and test and lead frames and all those things that go into making the product to secure the capacity we need to meet demand. But it's challenging. We do have a high dollar amount of unsupported demand, and we don't expect that, that's going to go away anytime soon, even though we're making these investments.

And what's the path for doing that? Because, again, for the very near-term and fulfilling that unsupported demand, you could continue to lean on outsourcing, which I guess if the capacity was available, it would be the quickest way of getting there. And then how long does it take to develop that internal capacity, which, again, would be used long-term to drive those margins, but also could be used to drive some of that -- to bring in some of that unsupported revenue?

Yes. So on the assembly and test side, it goes faster typically than it does in wafer fab. Some of this is, is we're already doing some of the production in-house, right? We might be doing 50% of the production of a certain package style in-house and now we're moving to 70%. Or we're seeing upside in demand on that package style and ramping the internal factories to be able to support that. So it can go relatively quickly, but it takes capital, and it does take time, and our operations team can only do so much at a time. So we're going as fast as we can. And we're making progress. But again, the demand environment is challenging. The booking -- it's great to have a large number of bookings, but we want to be able to service that demand when our customers want it, and that's a challenge right now.

Is there any view on when you might be able to catch up? And again, there's a natural seasonality in the business for some segments, kind of slowed down earlier in the year. But I guess, even for you guys, seasonality has actually been tough to call for you recently. So I don't know if there's a good answer there, if seasonality starts to take care of some of the supply constraints.

Yes. It's challenging. What's seasonality, right? It's been very hard to assess over the last couple of years. So we know that the December quarter is above seasonal. Guiding at 2.5% at the midpoint is definitely above what we would expect in a normal December. But it's been quite a while since we've had any sort of normal period. But I think how quickly we can recover on the supply side is going to be tied to what is the true demand going to be? And we obviously haven't set guidance for even the March quarter in 2021 at this point in time. So it kind of depends on what the slope of the revenue curve is, and we'll determine that here over the coming weeks as we approach earnings for the December quarter and giving guidance for next quarter.

Okay. Fair enough. With respect to the new margin targets, what's the correct time period we should view those under? Certainly, it's not just a couple of quarters, and you've had those old targets in place for a while.

Right. So I mean the good thing is, in the short term, just by taking away the underutilization charges, we're going to be roughly at what the old target was, the 63% gross margin as we move through the next couple of quarters. And then beyond that, there's a lot of blocking and tackling that has to happen. We highlighted the assembly and test, and moving more of that internally. We've got quite a bit of clean room space and wafer fab to move into, and it's just a matter of what's the demand going to be. And the higher the demand is, and we can add equipment cost effectively to the facilities, we just run more production and we're spreading those fixed costs over a larger number of products. It increases our gross margins across all the product lines. So that's kind of basic cost accounting there. But it's going to somewhat just depend on, again, what the slope of the revenue curve is and how quickly we can respond with our own manufacturing resources to support it. Also on the gross margin side, we've highlighted that there's some older factories that we've acquired through acquisition that will transition into some of our more efficient factories over time. These are long-term projects. We closed the fab in Oregon pretty early on after the Microsemi acquisition closed. There's another one in Santa Clara that's going to close down in 2021. And there are several others where we're working to transfer the process technologies and production into our Colorado factory, which is going to be quite a step-up in efficiency for that particular portion of the production. So Microsemi only outsourced about 50 -- excuse me, only did about 15% of their production in-house, but there is opportunity to continue to make some improvements on that portion of the business.

Okay. What about from an operating margin and OpEx standpoint, and you increased those targets as well. I guess, with Opex, there's really 2 things going on. One is in the very near-term. You're putting back some of the costs in salaries and profit sharing that had been taken away earlier in the year when we all feared the pandemic would be worse. And I guess the other is embedded in your operating margin guidance is actually a slight uptick in your OpEx as a percentage of revenue. So maybe you could address both of those things?

Sure. So we were very proactive heading into the COVID situation, not knowing what it meant in terms of what the demand destruction in the short-term would be that we had our employees go on pay cut. And we had 99% of the population signed up for that, and really appreciate their support in allowing us to continue to keep everybody employed, drive the things that we need to for the long-term success of Microchip for products and customer support activity, but be able to continue to produce high levels of operating margin and debt reduction from the cash flow. So we really appreciate that sacrifice that our employees made. The Board eliminated those pay cuts in November, and those are all baked into our guidance for the December quarter. We'll have the full impact of that in the March quarter, because they came in part of the way through the current quarter that we're in. And then on the variable comp side, our variable comp programs are all quarterly. And those have been reduced dramatically due to the environment that we've been in. But we need to get employees back to full pay at some point in time, and we can moderate that as the environment allows for us to do it. I think we're known as good operators, and we're going to do that as the P&L can afford it. But if we have a good March quarter, a good June quarter, a good 2021, those things need to come back to a more normalized level. And then there's pent-up demand in the business also to make investments. We've got 25 business units that all have goals for new products and revenue and margin, and you've got to keep feeding the system with resources to make that happen. And so we want to make sure that the investments that we're making today are appropriate to not just drive 40%-plus operating margins in 2021, but beyond that. And so that's the small tweak that we made to the model, but overall, driving operating margins to 42% puts us in pretty elite company in the industry. And the cash flow that, that's going to generate, our ability to pay down debt will continue, and we are very optimistic about our future.

Okay. And just that we did get a question in, before we move on to some other topics. And one question with regard to the unsupported backlog. And are you seeing any of that -- those unsupported bookings going to competitors? Perhaps that they have more capacity in their manufacturing footprint. In other words, perhaps you're leaving some business on the table, given the shortages.

So no. I mean the vast majority of our products are proprietary products. And quite honestly, our lead times have held up better than most in the industry. So we aren't losing business because of this. Yes, we're leaving some revenue on the table in the short term, but these are design-in products where it's having a good communication with the customer. They want the product on December 20, we can't deliver it until January 15, and then meeting that revised target that we've given them in terms of our delivery date. So we're very focused on it, but we're not losing customers or sockets over. We're talking a matter of weeks, in most cases, in terms of when they want it versus when we can deliver it.

As a follow-on to that, I'd probably be remiss if I didn't mention that one of the other semi suppliers, NXP, made some news last week as they issued a letter to customers talking about raising prices, and in their case, raising prices across-the-board because of supply constraints, and we've seen this another -- a couple of other suppliers have issued similar letters to customers. Is that -- it's maybe twofold. Is that an avenue open to you in terms of price increases? Is that something you have been doing? Is it a potential tailwind as you go into 2021?

So we're well aware of the announcements that have been made by some of our competitors relating to price increases, because of business conditions. And as of now, we really haven't made any decisions in that regard. We want to continue to be fair with our customers. You know that we have a process in terms of if we have to pay an expedite fee to a foundry or assembly and test supplier to get a product to a customer when they need it, that we will pass those costs on to the customer through an expedite fee. But in terms of a broad-based pricing increase, that's not something that we've made any decisions on, but it's unlikely that we would do that across-the-board.

Right. There's another question here, and it says, whether you're facing constraints on getting new equipment for back-end test and assembly. And I would imagine that you are not the only one trying to expand that capacity, given the supply constraints. So how constrained are actually the availability of tools?

We have access to tools. We've got -- sometimes, the lead times can push out a little longer than we'd like, but we've got a good pipeline of orders in place with our suppliers of key equipment. And we're getting that in as quickly as we can and ramping the factories. So it comes down to having good relationships, making sure you're doing the proper planning and knowing -- we know what the lead times are for these things. So it's not really a surprise.

Right. So it does sound like there is some incremental capacity you'll be able to address in the first half of next year?

Yes.

Okay. And the other -- second part of that question was, you said, I think, it was 40% of front-end wafer -- front end in-house. What's your target to -- for front-end manufacturing?

Yes. So I don't think that percentage will probably change materially over the course of time. It used to be, we did a very high percentage in-house, and that was really been -- has really been changed through acquisitions that have outsourced it. Typically, once the product is running at a foundry, you're going to leave it there. It's qualified on that process technology, and the particular foundry has the capability to produce it. And in some cases, we just don't have the capabilities in-house to do the things that we use the foundries for. So it's a good balance today, and I expect that mix to stay about the same over the course of time.

All right. Maybe I'll pivot over to something else, which I think is important, which is your cash return strategy and that's an offshoot of perhaps some change in your M&A strategy. And you've talked about this, I guess, maybe in the last 2 earnings calls, really, where -- because what's happened with valuations in the market and availability of targets, you seemed to be signaling that big M&A, probably is not likely to happen going forward. And then there are some changes in how you -- what you do with the cash because you are generating a lot of cash. In the past, you're using that cash to finance new acquisitions. So I guess there's an open question of what do you do with the cash going forward?

Okay. It's a good question. So let me kind of step back to where we were 10-plus years ago to where we are today. So when we started our journey down the acquisition path, Microchip was subscale in revenue, subscale in the products that we had in our portfolio. And fast forward from 2010 to 2020, we went from under $1 billion in revenue to $5.3 billion today. We've got an extremely broad product portfolio of microcontrollers, analog, FPGA, timing, memory, security, et cetera. It's a very long list, connectivity. And so when we look at the product portfolio, we don't really view that there is any sort of gaping hole in the portfolio. So there's not a need from a product perspective that, "Oh, my gosh, we have to have this to be successful." And we see that when we're presenting our total system solution approach to our customers that we can fill in most components on a customers' board with what Microchip has in the portfolio. So it doesn't mean that we have a complete portfolio, but there isn't a gaping hole where we want to go out and spend a whole bunch of money to do a deal. We still have a lot of leverage on the balance sheet due to the Microsemi acquisition. It's been coming down at a nice rate, but we're still highly levered. And so we have a desire within the company to become investment-grade. The leverage needs to continue to come down to do that. And we think that we can outgrow the industry and continue to gain market share year in, year out with the portfolio and the strategy that we have in place. So with that, as we continue to delever the balance sheet, we think there's a very nice opportunity long-term to change our capital return strategy and move away from M&A and go more into an increase in dividends and a share buyback program now. Because we still have deleveraging to do, these are future Board discussions that need to be had, and we do need to go -- continue to go down the deleveraging path. But the cash flow from the business is extremely high. Our leverage is coming down at a very fast pace. And if we get a tailwind behind us in 2021 from a revenue perspective, that deleveraging should go even quicker as the EBITDA really starts to expand. So we're excited about that. We think it's a good message for investors. Microchip has definitely traded at a discount to peers because of the leverage that we've had historically. And we think this capital return strategy should be a good thing for our long-term investors over time in terms of the way Microchip is valued, and what can we can return to them through putting more cash in their pockets through a dividend or reducing the share count through a buyback. And the details of that will come in the future as we make a little bit more progress on the deleveraging side, but it's a nice opportunity.

Do you have any targets in mind in terms of what's the net debt level that would -- at what net debt level would you begin to start returning some cash to shareholders? And then ultimately, what do you consider over the long-term an ideal net debt level to be?

So it's really up to the Board to make those decisions. I can give my 2 cents on that. But we still have deleveraging to do. As I say, we are focused on becoming an investment-grade company. And we've got a ways to go to get there and getting the leverage down. But it doesn't mean that we couldn't start a program earlier. We're going to have to wait and see what the Board decides. But we've been having these very, very miniscule increases in the dividend each quarter. And that could change in the future and not have a significant impact on the debt that we're paying down. But the Board needs to go through that. And once we have any sort of decision on that, we'll share it with investors.

Yes. I mean remember years ago, Microchip had among the highest dividend yield in semis, back at a time when semis really didn't pay dividends. And once you did the M&A, then less cash was available for the dividend, but that could change again, I suppose?

Yes, it could.

Okay. In addition, you've been making some changes in the convertible debt in an anti-dilutive fashion. Maybe you could talk a little bit about what you've done, why you've done it and what still could be done in the future in terms of taking out some of the dilutive convertible debt.

Sure. So we've done a lot in calendar '20. When we started the calendar year, Microchip had about $4.5 billion of principal amount of convertible bonds outstanding, with maturities ranging from 2025 to 2037. So convertibles have a low coupon, but as the company does well and the stock performs, they really become the most expensive form of capital that we have in our structure. So we bought a lot of the converts back from the marketplace and exchanged transactions for cash and shares. And so we did a transaction back in March of 2020, where I assure you it was very, very difficult to raise money. Everyone thought the world was falling apart at that point in time. But we raised $615 million bridge loan through some of our banking partners. And with that, took it as an opportunity to use those funds to retire some of our convertible bonds at -- when the stock price was $71. And that is an extremely accretive transaction today because the share count that would have leaked into the diluted EPS calculation would have been very significant from that. And so we continued with another transaction in the May-June time frame, another one in August. And then most recently, I should probably clarify that the transaction that we just did. So we have found when we're taking out these bonds that there are certain of the bondholders that are really long-only on the company. The hedge bonds or the arbs that are in the bonds are more than willing to have you take them out for cash and shares. But the long-onlys want to hold that paper. So more recently, we issued a new convertible bond, and investors may scratch their head, why would Microchip do that when they're trying to reduce the amount of convertible exposure. What we did is we issued a 4-year instrument, so much shorter than the bonds that were outstanding, and exchange those for the longer-dated bonds. So it allows us to have a time period when we can actually get that convert out of our structure over the course of time over 4 years and reduce our exposure to these longer-dated instruments. The other thing that we did is we put a capped call in place with that new bond, which essentially prevents dilution coming to the EPS until the stock is up 75% and over a 4-year time period. So if the stock is up 75% over 4 years, it's a good situation for our equity investors. And if we get some dilution from it, I guess it's okay. But we put some things in place that will allow us to prevent the significant dilution from these bonds. So we're happy with the moves that we've made in the capital structure, and we'll see how that develops in the future. But the converts principal on the structure has been reduced by about $2.9 billion, down to $1.6 billion as of today.

Right. It sounds like maybe there's still some opportunity for that $1.6 billion to make some moves with that as well as opportunity presents itself?

That's possible.

Okay. We did get another question that came in, and it was a question about your internal inventories. Should we think about those -- your internal inventory coming down over the next 2 to 3 quarters as you support some of this backlog?

So we're ramping our factories as fast as we can. With our guidance for the current quarter, we expect our inventory days to be about flat. At the midpoint of guidance, we'd expect our days to stay at about 120, which is really kind of the top end of the range that we've given historically that we'd want, between 115 and 120 days. And that can go higher in periods of stress or go lower in times when the business is really hot. But that that's allowed us to keep relatively good lead times for our products. And I don't have a forecast of what inventory days is going to do next year. Again, it's just like everything else we've been talking about, it depends on what the demand picture is going to be. But we're trying to ramp our factories to be able to keep the right amount of inventory in place, to keep our lead times as short as possible and service our customers.

And I guess just to round out the conversation on inventory, you've been stating for a while that the channel inventory has been at a record low level. So I suppose that's another issue that's been driving this kind of shortage situation because there's really not a lot on the channel that customers can draw?

So you're right. Distribution inventory has been quite low. I think it hit kind of a 15-year low back in the December quarter of last year at 28 days. And since then, it's moved by a couple of days. It's at 30 as of the end of September, and that's still on the very low end of the historical range, which has been between 27 and 37 days. And as it stands now with the supply constraints, I don't think distribution is in a spot where they're going to be able to build significant inventory. Typically, what you see in a cycle is that lead times push out, distribution feels that they need to keep a bit more on their balance sheet. But they've clearly been leveraging our short lead times over the last couple of years to keep their inventory low.

Right. And the next question indicates the paranoia that all semiconductor investors have and me, too. It's a question of how much of the strength might be OEM restocking at these levels as opposed to end demand? And of course, when things are good, we always get paranoid that they could be too good and it was difficult assessing what's real demand versus restocking. I'm interested in your view on that?

So it's really impossible for us to really know the answer to that. I would say, again, because of the supply constraints, it's going to be difficult for anybody to be building inventory in the current environment. I would say we're in the very early stages of a recovery, right? So typically, it's in the latter stages that, that is a worry. But at this point in time, we don't think that there is significant inventory that's being built up anywhere just because the supply isn't there to support it.

Right. Which is an indication of a view that demand is pretty good, and this stuff is being sold through?

That's right.

While we're talking about distributors, and we got about 5 minutes left. Microchip's had a bit of a different distribution strategy than a lot of the other industry. Your distributors have a little more off the grid and more Microchip-specific than some of the other larger competitors that you've had. How does that change given all the acquisitions that have been done? You do more business, I suppose, with the big global distributors than you have in the past as a function of the acquisitions. But how does that strategy change? And might it change further now that you're a larger company?

Okay. So we do about 50% of our revenue through distribution. I think a little less than 20% of that goes through the global distributors, or 20% of overall revenue goes through the global guys, which for us are Aero, Avnet and Future. We also use catalog distributors like the Digi-Keys of the world that kind of helps seed the market. And then we have many regional distributors that are more locally based, might have a particular expertise in a certain area. And we really like our strategy with distribution. It gives us the opportunity to have the best feet on the street in every major location throughout the world and not be dependent on a single distributor, where they're going to have strength and weaknesses, just like any business. They might be strong in 1 city and not in another country, and blah, blah, blah. So we like our strategy. We're not moving away from it. We are happy to provide distributors with what we would call a demand creation margin for new opportunities that they bring to us that we wouldn't be able to touch otherwise. And we service 120,000 customers globally, the vast majority of those go through distribution. So it's an important factor for us. Many customers choose to go through distribution because they can bundle everything that they need for their system through 1 local purchase through a global distributor. They might like the logistics or the payment terms or whatever. So distributors provide a valuable service. But we augment that, obviously, with our own internal resources to help from a design perspective, and we think we have the right balance of that today.

Right. And with that then, again, has it changed, for example, Microsemi, or even going back to Atmel, with the way that they went to market with distribution as opposed to what's being done on Microchip today. Has that changed any?

I mean I think it's probably modest changes, right? I mean when we acquire a company, we open up over time relatively quickly, our entire channel and sales organization to them, and that tends to be a big benefit, right? The more feet on the street that you have out there, pushing the product in the marketplace, the larger chance you have for success. And so we've integrated the best of Microchip and the best of our acquisitions into the strategy that we have today, and I think it's working well.

Right. And I guess we're just almost up on the top of the hour. One of the things I want to talk about is the other product lines, such as analog, for example, and analog is really broad within Microchip right now. I remember, this is going back a couple of years ago before Microsemi, that the analog business was starting to grow at kind of a slower pace as more of that content was subsumed into the microcontroller itself. Microcontroller is a small part of your business now just because of everything else that's there. And I guess the question would be, how do you see that analog segment growth could still a reportable segment for you? Do you think that's among the fastest? Is it kind of in line with what you'd expect for growth in the rest of the company?

So our analog product line is an important area for growth for us long term. It absolutely feeds into our total system solutions approach to the marketplace. We've always said that it tends to be the microcontroller, the microprocessor, the FPGA or that central computing chip on the board that's the first thing that the engineer selects when building out their system. But by having such a strong presence there, it gives us a great look into the rest of the board that the customer is creating, and we can position the analog products, the timing products, security products, et cetera, around that, and gain a greater share of the bond. So analog is important to us. Yes, there is instances where we're still integrating some of that analog functionality into the microcontroller. But our stand-alone analog products continue to perform well in the marketplace, and we compete against all the big guys and do very well. So it's an important part of our strategy and will continue to be for the foreseeable future.

Well, that's great. Well, I think we're up against top of the hour now. So we're going to have to cut it off there. Eric, we appreciate you being here. As always, very informative session, and thanks for joining us.

All right. Thank you. I appreciate it. Everybody, have a good rest of your day. Take care.

Thanks, everyone.