Good day, everyone. Welcome to Microchip's Second Quarter Fiscal 2021 Financial Results. As a reminder, today's call is being recorded. At this time, I would like to turn the conference over to Microchip's President and Chief Executive Officer, Mr. Steve Sanghi. Please go ahead, sir.

Thank you, operator. Good afternoon, everyone. During the course of this conference call, we will be making projections and other forward-looking statements regarding future events or the future financial performance of the company. We wish to caution you that such statements are predictions and that actual events or results may differ materially. We refer you to our press releases of today as well as our recent filings with the SEC that identify important risk factors that may impact Microchip's business and results of operations. In attendance with me today are Ganesh Moorthy, Microchip's President and COO; and Eric Bjornholt, Microchip's CFO. I will first comment on our CEO transition and Board appointments. Eric will then comment on our second quarter financial performance. And Ganesh will then give his comments on the results. I will then discuss the current business environment as well as our guidance. We will then be available to respond to specific investor and analyst questions. So let me begin by commenting on the CEO transition announced today and the addition of Board members. Today, we announced that I will transition to an Executive Chair role effective March 1, 2021. Microchip's current President, Ganesh Moorthy, will step into the role of President and CEO effective March 1, 2021. Ganesh will also join the Board of Directors effective January 4, 2021. I joined Microchip in February 1990 as Senior Vice President of Operations and was promoted to President to lead this company in July 1990. Microchip then had sales of about $60 million, and it was losing about $10 million per year. The main product line at that time was commodity e-comm, and the gross margin of the company was about 30%. The turnaround of the company was documented in my book, Driving Excellence: How the Aggregate System Turned…

Thanks, Steve, and good afternoon, everyone. We are including information in our press release in this conference call on various GAAP and non-GAAP measures. We have posted a full GAAP to non-GAAP reconciliation on the Investor Relations page of our website at www.microchip.com, which we believe you will find useful when comparing our GAAP and non-GAAP results. We have posted a summary of our outstanding debt and our leverage metrics on our website. We will now go through some of the operating results, including net sales, gross margin and operating expenses. Other than net sales, I will be referring to these results on a non-GAAP basis, which is based on expenses prior to the effects of our acquisition activities, share-based compensation and certain other adjustments as described in our press release. Net sales in the September quarter were $1.31 billion, which was flat sequentially and above the high end of our narrowed guidance range from September 9, 2020, when net sales were expected to be down between 2% and 6% sequentially. We have posted a summary of our GAAP net sales as well as end market demand by product line and geography on our website for your reference. On a non-GAAP basis, gross margins were very strong and near record levels at 62.2%. Operating expenses were at 23%. And operating income was an outstanding 39.2%, all better than the high end of our revised guidance from September 9. Our factory underutilization charges decreased from $13.9 million to $12.2 million sequentially as we started to ramp our factories to respond to the stronger-than-expected business conditions. We expect the continued ramp of our factories to lead to lower underutilization charges in the December quarter. Non-GAAP net income was $416.4 million. Non-GAAP earnings per share was $1.56, $0.15 above the midpoint of our…

Thank you, Eric, and good afternoon, everyone. Let's start by taking a closer look at microcontrollers. In a weaker-than-normal macro environment, our microcontroller revenue performed better than we expected. Our microcontroller revenue was sequentially down 1.8% as compared to the June quarter. On a year-over-year basis, our microcontroller revenue was up 0.8%. Microcontrollers overall represented 53.7% of our revenue in the September quarter. Now moving to analog. Our analog revenue was sequentially down 2.3% as compared to the June quarter. On a year-over-year basis, our analog revenue was down 8.2%. The weaker year-over-year performance of our analog revenue as compared to our microcontroller revenue was primarily due to our product lines which originated from Microsemi, which had higher exposure to Huawei, the communications end market in general, the commercial aviation market and the space market. Analog represented 27.6% of our revenue in the September quarter. Our FPGA revenue was up 24.8% sequentially as compared to the June quarter and achieved an all-time record even going back to the Microsemi history. On a year-over-year basis, our FPGA revenue was up 16.3%. I would like to caution investors that although the FPGA revenue trajectory is positive, the revenue does have some lumpiness associated with it because of the large exposure to the aerospace market. FPGA represented 8.3% of our revenue in the September quarter. Our licensing, memory and other business, which we refer to as LMO, was flat in revenue as compared to the June quarter. LMO represented 10.4% of our revenue in the September quarter. In October, we completed the acquisition of 2 small private companies. The first acquisition was New Zealand-based Tekron International, a global leader of timekeeping technologies and solutions for smart grid and other industrial applications. Timing is an operational necessity for real-time smart grid management and monitoring.…

Thank you, Ganesh. Today, I would like to first reflect on the results of the fiscal second quarter of 2021. I will then provide guidance for the fiscal third quarter of 2021. The September quarter continued to demonstrate what the best of Microchip culture and its people represent. Our global team of operations, business units, sales and marketing and support groups all came together in the middle of a global pandemic while working with a pay cut and delivered a superb quarter. Despite the COVID-19 pandemic challenges, we delivered net sales of $1.31 billion that was essentially flat sequentially and down only 2% from the year ago quarter. This is compared to our net sales guidance, which was to be down 4% sequentially at the midpoint as we capitalize on strong turns opportunities in September. We also delivered outstanding non-GAAP gross margin of 62.2%, which were near an all-time record level. We also achieved non-GAAP operating margin of 39.2%, above the high end of our guidance. Our consolidated non-GAAP EPS was $1.56, $0.15 above the midpoint of our guidance. Our bookings were very strong in the September quarter. We began the September quarter with a backlog position on July 1 to be down 8% from the backlog for June quarter on April 1. With strong bookings and strong turns still in the quarter, we ended the quarter at essentially flat compared to minus 4% at the midpoint of our guidance. Now I will discuss our guidance for the December quarter. Our bookings have remained strong in October. We are seeing a good recovery in the automotive, industrial, home appliance and medical devices for elective procedures markets. At the same time, work-from-home-related markets of computing and data centers as well as certain medical devices that surged with the pandemic revert to…

[Operator Instructions] And we'll take our first question from Ambrish Srivastava with BMO.

Steve, congratulations, and you'll be missed. And Ganesh, congratulations to you as well. I guess you'll have to take over Steve's role to keep us on our toes when we get on this call to ask a question.

It's a hard act to follow.

I know, big shoes to fill, but we would be expecting you to. 2 questions for me. One is on the constrained side. Could you just help us understand how has that translated into lead times? And then if you didn't have the constraints, how much of the business or what would the guidance have been? And then the second question is, maybe Eric can help us on this one, is on the capacity side. So a lot of moving parts there. So what was the -- what has been the outsourced versus in-source, both front end, back end? And then the capacity -- the CapEx increase, what should we be expecting both those 2 to be? And then how is that going to impact gross margin?

Do you want to take the lead times, Steve?

No. You go ahead. Take it.

So on lead times, the vast majority of our line items still have a 4- to 8-week type of lead times for standard products. There are specific package combinations that do have longer lead times. What is happening is, for the factors that I described, it is eating into multiple layers of the supply chain, so into the packaging and some of the subassembly involved in the packaging and into the testing infrastructure as many of these things come together at the same time. So we don't have huge issues with supply issues, but we have spot issues with specific package product combinations where they are. But by and large, for the vast majority of our products, we still have pretty good lead times. Go ahead, Eric.

Okay. The second piece of your question was there was -- it was kind of multifaceted there, but we've got some capacity questions. So we are making investments, as we walked through in our prepared remarks, in wafer fab, assembly and tests to increase our capacity. And if you look at last quarter, we did about 39% of our wafer fab in-house, 47% of our assembly and about 54% of our tests. These are relatively slow-moving metrics even with making investments. But over time, we absolutely would expect the assembly and test percentages to go up as we make these investments. But again, they're relatively slow-moving metrics. But the investments that we're making are all gross margin accretive. And you can see that our gross margin, we're guiding up in the current quarter, and we are expecting lower underutilization charges in the current quarter as we ramp our factories. And all these incremental adds to capacity that we're making, whether it's fab assembly or test, should all add to gross margin benefits for us down the road.

But unable to quantify at this point, Eric, as to how to think about the longer-term -- not the gross margin, what would the steady-state assembly and test would look like and front end would look like in-source versus outsourced?

So again, I don't expect the wafer fab to move significantly in terms of the percentage that we do in in-source versus outsourced. As the business grows, obviously, we're making investments. On the assembly and test side, we do expect those percentages to go up. Ganesh, do you want to give a comment on where you think they can go over time?

Yes. They move a little bit more slowly over time, and we expect that they will probably be in the north of 60% longer-term for assembly probably north of 70% for test. And there's a lot of moving parts into going into that. But that's what we like to be at. It gives us some control. It gives us capacity when it's difficult to get outside, gives us some control on our costs as well. And it all pays for itself in short periods of time in the way we measure what we do or don't do.

With the Micrel, Microsemi and Atmel acquisitions, we were at about 70% assembly and 95% test. So as we bought these 2 large companies, Atmel and Microsemi, they were much more outsourced than we were and our percentages dropped quite dramatically. And we've been working our way up and ideally would like to get back to the higher than 60% assembly and probably higher than 80% test, but it's a painful and slow transition because there's just too many variants, too many packages, too many test programs to correlate and all that. So it's an ongoing effort that will go on for years, but there are really no quick movements.

We'll go ahead and take our next question from Vivek Arya with Bank of America Securities.

And congratulations and best wishes to both Steve and Ganesh. Steve, my question is both near term and the growth you are expecting for next year. When I look at near term, in the September quarter, your microcontroller and analog sales were down a little bit. Year-on-year also, they are down a little bit. So I'm curious what is giving you the confidence to say we are at the bottom of the cycle when there is still some macro uncertainty because of elections and lockdowns. I'm just curious to hear those views. And then when you say significant growth for calendar '21, when I look at consensus growth numbers right now, they are for about, I think, 7% or so sales growth. What does significant mean? Does it mean 5%, 10%, 15%? What is significant in your book?

Well, what gives us the confidence is really we have worked through all the -- if you go back to February, March time frame, many of the estimates that the analysts and investors had were a very, very large drop because of COVID, of the type of 20%, 30%. Many people were modeling the business like the 2009 global financial crisis. Even some of our competitors were. We saw it clearly, and we were not modeling the business to be down that much. And we were right, our business was up in March. It was down only 1.3% sequentially in June. It is flat in September. And it's going up in December. So we kind of have been relatively more correct than anybody else. We are seeing substantial backlog building up. The backlog for the current quarter is significantly stronger than the backlog for the last quarter. And this is supposed to be a seasonally down quarter. Still, we think we're going to do pretty well. And then the bookings we're receiving, which are aging into the next quarter and the quarter after, are just very, very large. We're getting very, very strong bookings. Now some of that is concerns about the supply chain, a lot of constraints and all that, so people are giving orders earlier. Not that our backlog is much larger than before at a similar point in time, but that doesn't mean all that becomes growth because people give you orders earlier and then the crawl chart rolls off because you already got the bookings. But despite all that, we're expecting a much, much better March quarter and a significant growth after that. You asked what does significant mean, it means a lot more than 7% that you have a consensus.

And we'll go ahead and take our next question from John Pitzer with Crédit Suisse.

So I'll add my congratulations to both Steve and Ganesh. And Steve, I appreciate all the help over the years. I'm sure a lot of the analysts on the call feel the same way. I guess my first question is, Steve, you're calling sort of for the bottom of the cycle, June, September levels. I think what's very impressive is where your operating margins are despite the fact that we're at the bottom of the cycle. I'm just kind of curious, I know that expenses have been a little bit light this year because you guys have pulled back on things like variable comp. But given that you're not in the market of buying assets and you're really going to be just focusing on operational efficiencies, how should we think about incremental operating margins and kind of where operating margins can go from here?

So I think we have given a longer-term model, which is 63% gross margin, 22.5% operating expense and 40.5% operating margin. We're not quite there. We're getting close. The midpoint of our guidance is about 62.6% in gross margin this quarter, and I don't quite have the number on the top of my head on the operating margin. We got a little bit more to go before we really get to our numbers. And we will be analyzing going into the next fiscal year. We'll be looking at all that and at some point in time, coming back to The Street with a new longer-term target when we think that current targets have either been achieved or within a striking range.

That's helpful. And then for my follow-up for Ganesh. Your commentary around FPGAs in your prepared comments, notwithstanding kind of your caution that's a lumpy business, I'm just kind of curious if you can talk a little bit about kind of your core advantage in that market. That's becoming sort of a rarer asset over time now with some of the M&A activity in the space. What kind of longer-term prospects do you see in the FPGA space for you?

Sure. So in the FPGA market, we play specifically in the midrange of FPGAs, which is measured by how many logic elements there are in it. We don't go after the very high end. There are other players who are invested there. But in the midrange and lower end, we focus on applications that need low power. We have nonvolatile memory on these products that have security needs, that have robustness needs. And those take us into end markets that we really like. Those end markets are defense and aerospace. They're automotive, now with Microchip taking FPGA products into our historical strength, and industrial, once again, Microchip taking FPGA into our historical strength. So those are the end markets that we believe we're able to take, over the long term, the advantages of what we bring with our FPGA products in terms of low power, security, robustness and being able to position it into the markets where we have the strengths.

And we'll go ahead and take our next question from Toshiya Hari with Goldman Sachs.

And congrats to both Steve and Ganesh. Steve, you talked about your December quarter guidance being roughly 5 percentage points above typical seasonality, maybe 6% when you take into consideration the Huawei dynamic. You also talked about end market strength and share gains contributing to that outperformance. What portion of that 5% to 6% outperformance relative to typical seasonality is share growth? And what percentage is end market strength? And you sort of alluded to this before, but are you concerned at all that customers are pulling in end demand and you see a correction in sometime in the first half of '21?

Well, first of all, it's impossible to break down the growth better than seasonality into the 2 components that you described, what portion of the growth is market share and what portion of the growth is really just better end market. That's really very, very difficult on a short-term basis. You could really do a longer-term comparison on a growth rate basis, looking at how everybody else grew. We have no idea how everybody else is going to do this quarter or next quarter or next year. So I don't really have a good answer to that question. But to your question about the correction next year because customers are pulling in, we don't really see customers pulling in demand, we see customers pulling in, placing their orders and then scheduling them into the next year -- next quarter and the quarter after. And this is something we asked for through our letter back in July. We asked the customers then that in addition to the near-term orders that you've been giving us, please give us a longer-term backlog, tell us your requirements in the time of building constraints so we can plan accordingly and put the capacity in place. And customers have responded very strongly. So we're getting very strong bookings. But the component of the bookings that's aging into the next quarter and even a quarter after is very, very good. We have a lot of backlog already for the June quarter. And we have very, very strong backlog for the March quarter. So they're not pulling in deliveries, they're pulling in, placing orders and scheduling the deliveries, so they're not left short in case there are constraints.

Got it. And then as a quick follow-up, Steve, obviously, you've had a very successful career over multiple decades. Anything you feel like you left on the table? I know you're not necessarily leaving the company, but anything on your to-do list that Ganesh and the team can potentially move forward with?

Well, I don't have any regrets in the timing of -- as I'm doing it, it's more -- much more age-related and a family situation and grandkids and all that. But I could really go back to the Microsemi acquisition, which we announced in 2018 and the goals we had set out in terms of earnings per share and overall accretion. And we were talking about achieving $8 by the end of the third year. Some of those got interrupted by 2 major industry events. One was the entire U.S.-China trade war that took some wind out of the sails and then the COVID-19. And due to those 2 events, our credit rating came under pressure. The leverage became much higher in an uncertain situation, which was comfortable in a growing business but not as comfortable in a situation when leverage is high and the business has a lot of uncertainty. Back in the March time frame, analysts and investors were asking me questions regarding what happens if your business goes down 35%. And I was telling them, our business is not going down 35%. And they were not believing it, like why not, what if it happens. So I would say that the last 2 years have been difficult. And if we were able to have a normal level of growth in the last 2 years and all the accretion that we have achieved on the top of that, we would be over $8 of earnings per share today and hopefully start with the near $200 rather than where it is today. And I leave that for Ganesh to put a 2 00 in front of it.

And we'll take our next question from Harlan Sur with JPMorgan.

And let me also offer my congratulations to Steve and Ganesh. I guess, first question, what was book-to-bill in the quarter? And then given the strong demand environment and constraints, now may not be the right time to be executing to this, but I believe that you guys still have a network of very small fabs in your manufacturing footprint. So if you can -- just maybe how much of this is yet to be consolidated? Can you just remind us how much more COGS savings are still to come as you consolidate these smaller fabs and over what period of time?

So we're not specifically providing a book-to-bill ratio for Q2. I think we had given the reason for that many times in the past. Our book-to-bill ratio was very good. And usually, we have seen investors and analysts essentially take the book-to-bill ratio and try to translate that into a growth number, which does not really work because I already said that a lot of the bookings, strong bookings that we are receiving, are actually aging into the March quarter and some even into the June quarter. So bookings -- the bookings we receive for aging over the next 12 months, and you divide that number by billings in 1 quarter, so it's a little bit apples and oranges. Bookings over the next 12 months, aging over the next 12 months, but billings shipped into the last quarter. The number was very good, but providing that numerically, that number, we have seen investors not interpret it correctly. What was the second part of your question?

It had to do with cost of sales, improvement and ongoing consolidation of factories and things like that. So I'll start, and then Steve or Ganesh can add on to that. So we announced last November some restructuring of our Colorado wafer fab, and we're making excellent progress on those fronts and have achieved a large amount of the cost savings that we outlined at that time. And you've seen our gross margins hold up extremely well. There's still work to be done. Our operations teams and both the front-end operations and the back-end operations are extremely busy. And we talked about our capital expansion plans and the improvement that we'll see in gross margin there. And obviously, the more product running through our own factories absorbs the large flywheel of activities that we have on the cost side. So we've got a lot of good things working on gross margin. We've guided at the midpoint this quarter to 62.6%, which isn't very far away from our 63% target. So we feel good about that. And -- but I don't think we're going to talk specifically about some of the specific actions that are being taken, but we're doing well on structuring the operations appropriately. Steve or Ganesh, anything to add to that?

Well, yes. What I will add is that if you look at our gross margin and operating margin performance in this down cycle, I mean, it's been exemplary. And compare it to any of the prior cycles, we did just extremely well. Our gross and operating margins did not go down by many hundreds of basis points. I think gross margin used to go down by 600-plus basis points. Now why is that? I think that's partially the result of diversifying the business, creating several end markets. The acquisitions we did really helped us build that, serving the entire solution of the customer with a total system solution. So the revenue didn't fall as much. The gross margin didn't fall as much. And the operating margin didn't fall as much. And we are sitting at near-record gross margin and just had shy of the record operating margin at the bottom of the cycle. And as we go from here, as the revenue increases and we are ramping all of our factories, as I spoke about, all our high-volume fabs and assembly and test are all on a rapid ramp to provide the growth that we see into next year. We have increased our capital expense budget. So as we achieve that ramp, the incremental cost of the next product we make is much lower than the cost of the product we're making today because you get better absorption and the incremental gross and operating margin, you understand that concept. So it's a very, very exciting time we're going into, where, at the bottom of the cycle, we are near a record. And then, from there, as the factories ramp and the underutilization first goes to 0 and then you go above that and incrementally start dropping gross and operating margin, that would be very, very good. And what we need to do is really put some numbers around it and at some point in time, talk to you regarding what does that mean in the long-term model that we're not prepared to do today.

Got it. Okay. And maybe just a quick follow-up on the product side. I thought that you guys actually introduced the RISC-V-based FPGA product, but I wanted to get your views on any initiatives that the team has in terms of RISC-V open architecture as an addition to your MCU product portfolio.

So we are part of the RISC-V foundation through the Microsemi acquisition. It had started before us. The first point of implementation is on the FPGA SoC product lines. And we did introduce that, as you noted. There are many possibilities with RISC-V, and they are being evaluated within Microchip, but there's really nothing to report at this point beyond what we have done on FPGA.

And we'll go ahead and take our next question from Craig Hettenbach with Morgan Stanley.

Certainly, quite the journey, Steve, in the last few decades, to see that the company evolved just organically and through M&A. On the last point, including M&A, I think one of the things that's been focused now is with total system solutions. Maybe you can just give some context or update of kind of where you stand with that and some of the efforts you're driving through the sales force.

So I think when you look at an embedded control system, a microcontroller-based system, it has a large number of components around it. And over the years, the number of components have increased. Now even go back 10, 15, 20 years ago, it will usually have some sort of power management, A to D converters, reference devices, maybe some discrete devices, some study grams, some memory, flash memory, nonvolatile memory. It will have those kind of things. But over the years, the amount that goes around that microcontroller has increased substantially, and it increased substantially with connectivity. So today, you need a USB, Ethernet, WiFi, Bluetooth, display driver, touch functionality, so just high-voltage sensors and others. So the number of components that go into an embedded control system have really multiplied in the last 20 years. And we began our journey of really selling things around the microcontroller beginning in about '99, 2000, and then we did a small analog acquisition of TelCom Semiconductor in 2001. And with those resources and adding to it, we started building our analog franchise. And then we didn't really start doing the drumroll of acquisitions until about 2010. But the results have been that we have added all those products now available from Microchip, you can buy them as a kit or you can buy them separately. But if you now look at a reference design for an application and almost find any embedded control application in home, in industry, in car and in any place, open an appliance and look at its printed circuit board, you can essentially have everything on that from Microchip today, a microcontroller, a memory, analog, converters, opt-ins, some sensors, power management, drivers, connectivity, Bluetooth, Ethernet, WiFi, 802.11, anything else. And so we feel that we have done all that, and now we have a powerful franchise to be able to sell that entire solution. And the challenge in the last couple of years has been now training the sales force to really be able to take that kind of message to the market, and they're doing very well at it. So at this point in time, we do not really feel a burning desire to have to do another acquisition. The valuations are sky-high, will never meet our taste. And number two, we still have high leverage and we're paying down debt. And we have clearly signaled to The Street that when our leverage goes down and then we're still producing a large amount of free cash flow, a more likely use of that free cash flow is the increasing dividend and stock buyback and all that and not the new set of acquisitions. So I think that's where we are because we feel we have completed the solution, we have an enormous scale and do not have the scale disadvantage anymore. And with that, we are going to grow the business organically.

And we'll go ahead and take our next question from William Stein with Truist Securities.

And I'll add my congratulations on the transition. It's been quite a run. So congrats on that. I want to dig into the concept of book-to-bill and backlog that you've made very positive comments around, Steve. But maybe I can ask about it this way. It sounds like the book-to-bill was very strong. Backlog is up a lot, but it's coming more in the form of duration of backlog as opposed to what's deliverable in the near term. Is there any metric you can give us around that, around the duration or around maybe what portion of March do you think is now filled? And then the concurrent question with that is those behaviors from customers typically happen under 1 of 2 conditions, either the customers suddenly have a lot more confidence or optimism in their business and the other is when they think they're not going to be able to get supply. I wonder if you could comment as to which of those you think is driving the improved duration of the backlog.

So I think it's a combination of it. Obviously, having such a broad customer base of 125,000-plus customers, you often don't really know how the customer is thinking, so you get samples of it as you talk to the large customer and visit them. And we're really not even visiting them these days. The interaction is largely virtual. But I think it's a combination of customers reading about strengthening lead times in the industry, constraints they are hearing. And when they have an experience from not being able to acquire one component, let's say, from one of the other suppliers, the purchasing manager's action often is to really go ahead to secure and place the order on all the components, whether the lead times are going out at a particular supplier or not. And in our case, we specifically advised the customers back in March that we were getting largely short-term orders, and we need short-term orders to make the quarter. That's great. We thank them, but we also needed your longer-term orders so we can more efficiently build the parts and batch process it and place the orders on our suppliers ahead of time and so on and so forth. And what I would say is that the customers have responded extremely well. I mean our customers have always responded to our letters extremely well. And what you have seen is, a quarter later, exactly a quarter after we wrote that letter, and customers have placed a large amount of backlog that ages into the following quarter.

Maybe just want to clarify one thing Steve said there. He mentioned the letter to customers in March. It was actually in July. I know that's what he meant to said, but just for the record.

I meant -- yes, July 7 was the date I thought, somewhere early July.

And we'll go ahead and take our next question from Shawn Harrison with Loop Capital.

My best wishes. An easy question, hopefully, and then a question more on distribution. With the volatility in FPGA associated with the aerospace business, should we assume that you kind of see more of a normalization here in the December quarter? And then second, Steve, how are the distributors reacting to the tightness in supply or the tightening of supply out there? I know you highlighted that channel inventory days are still low, but is there any pressure for them to add more Microchip stock?

So what was the -- I'll take the distribution. What was the first part of the question?

On the FPGA trends. Why don't I take that one, Steve.

Go ahead.

So as we have said many times, FPGA is a more lumpy business. Trajectory-wise, if you plot the last many quarters, you'll find that it is up and to the right. Quarter-to-quarter, there are going to be changes depending on which lumpy business is coming through or is delayed for whatever reason, and that's what you will continue to see. But if you look at it over 8, 9, 10 quarters, you'll see that it is up and to the right as an overall FPGA business for us.

Go ahead, Steve. So on the distribution front, we're getting huge orders from distribution just as well as we are getting it from direct customers. So our bookings were strong in both, in the distribution channel as well as in the direct channel. And the behavior is largely similar where we're getting bookings to make -- we got the bookings to make the September quarter, and we're getting good bookings to fill up the December quarter, but very large number of bookings are really actually aging into the March quarter and some even in the June quarter. So distributors also are layering in the backlog so that, if the lead times push out further, they're not impacted, and they're placing the backlog already going into the next quarter. So when we wrote the letter, the letter was not only to our direct customers. It was also the same message to the distributors, and they have responded in kind. And just overall inventory has been low for quite some time. I think it hit 15-year low. And then from that, it has only come up a day or two. So in the coming year, as we are expecting significant growth, my sense is it's really up to distribution, but my sense is that the distributor will have to increase the inventory and can sit at a 15-year low.

And we'll go ahead and take our next question from Chris Caso with Raymond James.

Yes. And Steve and Ganesh, congratulations to you both. It's been a pleasure working with you both. The question is regarding seasonality. And you mentioned your view of normal December seasonality down 2% to 3% sequentially. I know in the past, since the Microsemi acquisition, you've hesitated to make a call on seasonality because there really hasn't been much of a normal environment since that acquisition closed. Given that you've offered that for December, do you have a view or an updated view on what you'd consider to be normal seasonality for the March, June and September quarters?

We really don't. I think we're going to have to run a whole year normal and kind of start to look at it. If you go back prior to the Microsemi acquisition, then March quarter will usually be up sequentially, I would say, a couple of percent, 2%, 2.5%. This year, March was up 1.3%, although it got impacted at the late part of March with the China not coming back from the COVID-19 crisis. Despite all that, I think we were up a couple of percent and could have been more. So I think the March seasonality is somewhere around 2%, 2.5%. But I think there are less data points on it with the acquisition than would be otherwise. If you take the Microsemi out, I feel comfortable with a couple of 2%, 2.5%. Add the Microsemi in, we don't have enough data points.

Right. Okay. Understood. Just as a follow-up, with regard to the repayment of debt. Eric, this is for you. Just could you give us a sense of what your expectations will be over the next few quarters? Again, assuming that there is some degree of recovery, as you say, you should be generating more cash. And how does that affect the timing on getting to your net debt target, which I believe is 3?

Yes. I mean we've been generating a bunch of cash each quarter. We indicated that we expect about a $300 million debt paydown in the current quarter. There's some positives and negatives. Obviously, if revenue is growing, we're going to throw off more operating margin, and we are specifically trying to invest a bit more in capital as we talked about. So our CapEx was extremely low in both the June and September quarters, and it's going to be higher here in December and March so a bit of an offset. But I would expect $300 million-plus range. And as the business environment improves, that's only going to go up. So we're making really good progress on debt paydown and expect that to continue. And you mentioned the 3x or less, and that's what we're looking for, looking to be an investment-grade rated company. And over time, we will absolutely get there.

[Operator Instructions] We'll take our next question from Janet Ramkissoon.

Congratulations, guys. Steve, thanks very much. It's been a real pleasure to be along your side for the last 30-some years since you became CEO. And again, congratulations to all. I had actually 2, really quick. Can you give a little color on what's going on in the auto business? Do you feel that we -- there is a secular change happening in demand for autos because of COVID and safety? And secondly, can you provide a little more color on Huawei? When did you apply for the license? And what is -- can you give us what your best-case scenario might be if you were to get the license? And when you get it, how fast can you ramp up?

Well, first of all, thank you, I think, you been with us for probably the entire 30 years or so or 27 years as a public company and prior to that even. Long association, so thank you. And I remember you came to visit our booth in Las Vegas at the CES conference earlier this year. Regarding your specific question on Huawei and automotive, so the automotive business is -- saw the largest decline out of any end of market back in the June quarter because many, many factories is downright shutdown. And then in the September quarter, they started to bring the factories back up, and the business was up from June quarter to September quarter. But the September quarter factories were not full from the beginning. They were ramping during the quarter. So there is quite a substantial growth in the automotive segment, at least for us from September quarter to December quarter, as the factories continue to ramp. So the automotive business now from this point on, kind of looks normal. Cars are selling. The inventory is low, so they're rebuilding their inventory. Lots and lots of automotive customers are making investments into electric vehicles, where we have significant content. And the content in electric vehicle is actually higher than the content in a regular vehicle. So automotive business should look good going forward from here. The other part of your question is Huawei license, when did we apply? Let me hand it off to Ganesh to answer that question.

Yes. So our application was within the last month. It is a very uncertain process of how it navigates through department of commerce and whoever else would have to weigh in on it. I think it's impossible to give you an estimate of what might happen and when and what would it mean. The business with Huawei has many products and is not just a single license. We would need multiple licenses, and each one has a separate application you'd have to go through. And we have a prioritized process that we're going through with it. Given the uncertainty of the process, we did not believe trying to count on any revenue made sense. And once we have line of sight to the license, we would still need to work with Huawei on what they would need, when we would be able to ship the product to be able to provide any kind of guidance on what does it mean to our business.

And we'll go ahead and take our next question from Matt Ramsay with Cowen.

Congrats to you both. Ganesh, I wanted to ask a question. I noticed some of the private company tuck-ins and some of the investments that you've been making over the last few quarters regarding the edge opportunities. Our team has done quite a lot of work on edge as an emerging market, both on edge clients and on sort of edge cloud. And I wonder if you might give a little context as to how big of an opportunity you guys might think about this being over the next 3 to 5 years for Microchip and if it's something that might evolve out of your FPGA franchise or out of your microcontroller franchise or if there's need for a bit heavier-handed compute that might be required as you guys approach some of these edge opportunities.

When we think of the edge, it all comes around the mega trend of artificial intelligence and machine learning that we have spoken about. And we think of that in 3 different buckets, so to speak. There's a part of it which is the best known for many people, which is what happens in the cloud. And that is the domain of many people who have very large and highly processor-intensive compute. We do play in the cloud, but our role in the cloud is predominantly around PCIe switches and a few other things that are associated with how CPUs, GPUs from other companies [ are taught ]. The edge is the second part, and edge is an exceptionally important part as you think of factory automation and the industrial IoT because at that edge compute for the factory is where much of that machine learning is going to be taking place and the application of artificial intelligence. There, we have introduced using FPGA as one of the platforms, a number of solutions. Smart embedded vision is one piece of that, which can go into machine vision for factories, physical security, medical vision, depending on that end application. But surrounding these products are many of our standard products, too. It needs microcontrollers, analog, security, things that process. And then the third element of artificial intelligence machine learning we're looking at is all the way to the end where the sensing is taking place. And there again, we're looking at our standard microcontrollers and some of the other products as to how can they do, to a lesser extent, but what is exactly needed at the end nodes, the learning and the inferencing that is needed using standard microcontroller. So it's a much bigger field than just the edge alone in terms of our interest. But certainly, FPGA at the edge is a key part of how we intend to prosecute that.

And we'll go ahead and take our next question from David O'Connor with Exane BNP Paribas.

Great. Congratulations on the results. Maybe a question on my side, going back to the supply chain constraints, which exact category of products are impacted there? And Ganesh, in your prepared remarks, you talked about Huawei, you talked about the mobile phone refresh and some reshuffling of share as well in capacity maybe. It seems more short-term related. So the question is, do you think these constraints dissipate from the March quarter? Are they here to stay with us for some time? And I have a follow-up on the gross margin.

So the general comment would be that it affects the supply chain of people who are packaging product, their supply chain, which can be lead frames, it can be substrates, can be equipment that does bonding and various other things. And so it may not be things that we are directly involved in, but it consumes bandwidth and capacity of the supply chain, both directly what we deal with and then their supply chain as well. And so all of these compete in many cases for either the materials or the equipment capacity that is out there that we would otherwise be using. And then as we go to use them, we find that, in some cases, they're constrained. We've been able to manage through a lot of it. We do have second sources for some of these things, and we do have a lot of internal capability. And as Eric mentioned, we are accelerating, bringing more capacity internal for some of the package types. So what normally happens in business is, when you have these constraints, the companies that are in the business of providing that capacity or that material, respond with what they can do to take advantage of that situation. And so there is a capacity response that they come with. Some of it also, it's possible that there could be a surge in demand that then begins to dissipate. We can't really predict where that is. But we think that they're all going to be constrained through the December quarter, and it's possible some of that will spill over into the March quarter as well. But I don't have any line of sight into exactly when all the constraints will dissipate.

That's very helpful. And maybe a quick follow on the gross margin for Eric. Eric, the -- you talked about the strength in the March quarter, and it seems from just the higher utilization, you could hit that 63% gross margin in the March quarter. My question is on the additional CapEx that you budgeted for calendar year '21. How much of a headwind is that as that capacity comes online? Does that come on slowly through '21? Or is that going to come initially and we have to factor that into -- as a gross margin headwind into '21?

So in terms of when the capacity comes online, it depends whether it's wafer fab, assembly or test. And so it just depends on what qualification we have to go through and the work that needs to be done in our factory. So it comes on gradually over time. I don't view it at all as any sort of headwind to gross margin. It's just a matter of can we get it installed and up and running and get product out the door. So I think gross margin is in good shape. I didn't make a specific comment on gross margin for the March quarter. So I want to make it clear, we were just speaking about the current quarter, but there are lots of things that we're doing in our business and not the least of which is if we get into a better revenue environment as we look forward into '21, that's going to do very good things for our gross margin, and we'll continue to evaluate the long-term model because we're getting close.

I'd like to clarify one thing. You said somehow the capacity or the CapEx adds a headwind to the gross margin. That's entirely opposite of what we will experience. I believe the comment of the CapEx providing headwind to the gross margin comes from when you build a large greenfield fab and you spend $1 billion, then the process has to be qualified and it slowly ramps, meanwhile, the factory is depreciating. That's the kind of headwind to the gross margin probably you're talking about. And we experienced that back in 2003 when we were bringing up our Fab 4 up in Gresham, Oregon. The kind of capacity we're talking about is not that. It's made up of $300,000 to $1.5 million various pieces of equipment in assembly and test and fab and the diffusion tubes and others incremental capacity. And it becomes productive really the quarter after it is added, and it never has a headwind to the gross margin. It's always accretive to the gross margin because the product it produces, it produces, I think, incrementally lower cost than the average product without that. So there is no headwind to the gross margin. There's only attrition to the gross margin.

And we'll go ahead and take our next question from Vijay Rakesh with Mizuho.

I'll add my congratulations for Steve and Ganesh. Just I'll combine my 2 questions. I know looking at the back half here, you're seeing some strength in auto industrial and also a nice recovery in China. Just wondering what your revenue exposure was in that industrial auto and especially -- or into China? And lastly, I'm sure there's some COVID impact on the gross margin line even though margins are very impressive where they have rebounded. Just wondering what that -- if you have kind of sized that COVID impact and that should go away or resume into next year.

I don't know if I got all that. I think the first part was really asking the mix of industrial and automotive. Ganesh, I don't know if we have that the last quarter. We only provide it once a year.

Yes. And it doesn't move dramatically quarter-to-quarter. What we have shown publicly measured at the end of March for the prior year was that industrial was 28% of our revenue. Automotive was 15% of our revenue. Obviously, in the June quarter, there was more headwinds in those 2 end markets, but I don't really have a number. And then those are all reversing as we went into the September quarter and into the December quarters itself. So those end market percentages for us usually don't change that dramatically over time. And then I'll pass it back on the other part of the question.

I didn't understand the gross margin question. Can you restate that?

I was wondering, in terms of COVID on a logistics and operational basis, I would assume there's some impact. So I was just wondering what the COVID impact was to the gross margin, and I would assume that would reverse. It would be a tailwind next year.

It's very small.

The impact of COVID and gross margin was more like in the June quarter when, within our factory in Philippines at a much lower capacity because you couldn't get all the people in, and we had 150 people living in the facility and working and sleeping there at 30% of the factory's capacity. There was no COVID impact in September quarter. I mean minor providing meals and others to the people who are living there, but there was really no meaningful impact in September quarter, and there's no recovery of that next year because there's no impact now.

Right. The amount in the June quarter was $2.8 million, and you'll see that in our press release, but there was nothing that we broke out separately because it was immaterial to the September quarter results.

And we'll go ahead and take our last question from Mark Lipacis with Jefferies.

Great. The last question on Steve's last call. That's quite an honor. Ganesh, congratulations. And Steve, thanks a lot for all the great insights that you saved. I will miss them. I just had a kind of a strategic question for Steve. You kind of described a scenario where the M&A slows down. You're under a deleveraging cycle and then a capital -- kind of a capital return cycle. What does that say -- what should investors take away what that means about the semiconductor industry? And does it necessarily mean that, ultimately, that there's a different set of requirements from Microchip to their customers? Does it mean that there has to be a regearing of Microchip or Microchip 3.0, if you will? How should -- what should investors take away from that? Or is Microchip 2.0 the perfect equation for what we expect to see next in semis and for Microchip specifically?

Well, I think if you look at a Microchip of 10 years ago, it was providing predominantly microcontroller, which was 80% plus of our business, and a small amount of business remaining was either some analog products or memory products, but they were largely we couldn't complete a customer solution. So when we called on a customer, we largely provided microcontroller, maybe a little bit of other stuff they won in. And the customer will surround our microcontroller with analog coming from Maxim, ADI, TI, Intersil or others would buy WiFi connectivity from people who made maybe USB from Cypress, maybe Ethernet from some Ethernet company and so on and so forth. And we, many times -- there's an old story. I think if you have a couple of minutes, I'll tell you. Back in 1993 time frame when we went public, we used to have a partnership with the Maxim where we will get hotel rooms where we will invite the customers and give seminars jointly. And we will take the front half of the room, and Maxim will take the back -- front half of the day and Maxim will take the back half of the day. We'll tell them how to design our microcontroller, and Maxim will teach them how to add analog around their microcontrollers. And it was a great partnership. It lasted for about 5 years. We both benefited, shared the expenses, and we didn't have analog products at that time. And the story's 20 years old. It was with Jack Gifford of Maxim who died many years ago. So many people wouldn't know it. And then Maxim, Jack Gifford wanted Microchip to pay him something because he said we're bringing all the customers and you're benefiting from selling them microcontrollers, and I said this…

That's very helpful, Steve. Really appreciate it.

Thank you.

And that concludes today's question-and-answer session. I'd like to turn the call back over to Mr. Sanghi for any additional and closing remarks.

I think when I thank our investors who have been with us for long, and I certainly have great relationship and a long career with all of you. I'll still be here. I'm not going away. We still have a conference coming up, a Credit Suisse conference in early December, then the next earnings call in February. And then on March 1, I become Executive Chair, but you will still see me at the investor circuit and conference calls and investor conferences and others. So I'll still continue to be involved with Microchip and not going away. Thank you very much.

Once again, that does conclude today's conference. We do appreciate your participation. You may now disconnect your phone lines.