Microchip Technology Incorporated (MCHP) Earnings Call Transcript & Summary

Hi. Good morning and welcome, everyone, to the Future of Technology Series. This is John Vinh, Senior Semis' Analyst at KeyBanc Capital Markets Europe. We are pleased to have to Steve Sanghi, CEO; and Eric Bjornholt, CFO, from Microchip to join us. The format of this session will be as follows. Steve is going to kick things off with a presentation, and then we'll take Q&A at the end. Please feel free to submit them online at the bottom of your webpage. And with that, I'll pass it over to Steve.

Thank you, John, and [indiscernible]. All right. Welcome, everyone. Before I begin this presentation, I wish to remind you that during this presentation I'll be making some projections and forward-looking statements regarding the future financial performance of Microchip. These involve predictions and the actual results may vary materially. So I refer you to some filings with the SEC regarding some important risk factors for the company. Okay. So I'll begin with our business results. Q1 business results were quite good. We reported sales of $1.31 billion, which was down only 1.3% sequentially and down 1% from the year ago quarter. We produced 61.70% gross margin non-GAAP and non-GAAP operating margin of 38.6%, EPS of $1.56. We guided Q2 to be down between flat and 8% sequentially, and with a non-GAAP operating margin between -- of 37% to 39%. And Q2 guidance for non-GAAP EPS was between $1.30 and $1.52. Announced record dividend, paid down $394 million in debt and paid down about $2.62 billion over the last 8 quarters. So this is a slide I prepared to compare our results with the competitor for the June quarter. On the left-hand side are sequential results, on the right-hand side are year-over-year. Analog Devices quarter ended April 30, so it didn't have the full impact of COVID-19. So you can see sequentially, we had the best performance among our competitors. And the same thing year-over-year if you take the Redpine results out, Silicon Lab will be below. And then looking at the September quarter guidance, we guided -- the midpoint was 6% below the same quarter a year ago. Silicon Lab, again, had Redpine in it, but other than that, there's only one other company, ST Microelectronics, which is above us. So compared to a lot of our larger competitors like Renesas, Xilinx, TI, Analog Devices, NXP, our performance really stands out in market share. I will skip a few slides and really hit the highlights. This is our annual growth based on end market demand. So fiscal year '19 was a record quarter -- record year. Fiscal '20 was just as smidgen below that. And fiscal year '21 is plotted at a run rate just for 1 quarter. Now this shows our 8-bit microcontroller market share. As you can see from the early beginnings, in early '90s, we have come up and become #1 and have remained the #1 for a very long period of time. This shows our market share in 16-bit microcontroller, how it has risen over the years. And now the #5 manufacturer. Here is the market share for 32-bit microcontrollers similarly. And here is all microcontrollers combined. So from early 1990 to 2009, our market share has gone up from #23 to #2 -- #3, sorry. And last year, we gained share compared to Renesas and NXP. And hopefully, in the coming years should be moving that forward. This is the analog growth over the years, both organically as well as through acquisitions. Here also, fiscal year '19 was a record year. '20 was down a smidgen and '21 is just flatter based on 1 quarter in the [indiscernible]. This is the data from IC Insights showing 2018 to 2019 analog growth or decline. You would see that Microchip is the only company that grew in the calendar 2019. Everybody else was down. And this slide shows our FPGA growth based on the end market. This data goes back to the Microsemi heritage. We bought the company at the start of the fiscal year '19 [indiscernible]. So Microchip 2.0, we have been talking about total system solutions for some time, and I'll show you a couple of slides on that. We continue to have leading customer preference to design with our microcontrollers and have multiple growth drivers. Longer term, we have guided to a gross -- sorry for the brief interruption we had in the streaming. The slide is -- all the slides are posted on the Internet. If you miss something, you can probably go back to that. I'm going to start with this slide now that shows expanding Microchip solution through acquisitions. This slide shows how over a decade plus, we have bought a large number of companies which have really broadened the Microchip's product line from a microcontroller analog only to a very broad product line that has networking and USB and Ethernet and automotive products and discrete products and FPGAs and other stuff. And this is really what has allowed us to build a total systems solution for our customers. And as you see this slide, you can see that the central chip on a board, which is often a microcontroller, a microprocessor or an FPGA is surrounded by a large number of chips on that board coming from analog, power management or memory or connectivity or various other devices. And today, Microchip makes all these devices, any block that is shown here, Microchip makes those products. So I will show you 2 slides now, how that total system solution is helping Microchip grow its sales. The first one is in this current server design. There are 4 sections in this: front panel, motherboard, PCIe cards and backplane. And the color coding shows the various business units of Microchip, how they have content in this large design. So there are 2 8-bit microcontrollers, there are 3 32-bit microcontrollers. There's a part from wireless system group. UNG's networking, there are 2 products from the -- USB products from the networking group. There are 5 timing products from the timing group. There are a couple of products from the analog power management group and a product from computing product group. So this is sort of how our sales and marketing and application group are engaging all of the business units to contribute content in this large design where Microchip is able to provide a total system solution. Now this is the current design in production. And this is the next design. And if I go back and forth between this, you can see how many more colors are there, more business units, larger number of products and much, much larger content, how our content is growing from one vintage server design to a next vintage server design. Revenue by end market. So this is the revised end market mix that we introduced in our conference call where data center and computing have now become the second largest business segment, and we are experts with automotive as automotive business has gone down. So industrial continues to be our largest at 28% and rest you can see in the pie. Financial results, guidance and long-term model. So here, you can see the results going back about 4 fiscal years. And I began these results from Q1 of fiscal year '17, where we had just bought Atmel. And the blue section that I highlighted was the operating margin. So prior to that, our operating margin was in high 30s. But after -- just after buying Atmel, Atmel was essentially breakeven, making really no operating profit. So combined together, our operating margins were 27.4%. Then it shows how we integrated Atmel, improved the business and how we plot back into 39.5% operating margin in Q4 of fiscal year '18, which was a record. And then we bought Microsemi in the middle of Q1 of fiscal year '19. And the operating margin dropped again a little bit to 38.9%. But here was rapid integration, and Microsemi operating margins were really around 30%, so it didn't have as negative an impact as Atmel had earlier. But then further improving, you can see that we have maintained the operating margins very respectable in the 38.5% range. During this last 2.5 years, we've had 3 different events. There was a significant inventory correction event on the Microsemi business. Then we have the U.S.-China trade discussion and [indiscernible] which impacted the business negatively. And now we have the COVID-19. So if you look at the top line, the Q2 fiscal year '19 was a record revenue of $1.5 billion, $1.3 billion. And the last quarter at $1.309 billion, that business is down $200 million and our operating margin is about the same. So as that business comes back and we go towards the records on revenue in the coming quarters, you can see that there should be a much higher operating margin leading us to a long-term model, which is shown below, the long-term model being 63% gross margin, 22.5% operating expenses and 40.5% operating margins. This slide shows how we have brought down our net debt. The blue line is a reduction in leverage, which has gone down from 4.86 to 4.24 over these last 8 quarters ago. Going forward, our significant focus continues to be on these 6 megatrends: 5G, Internet of Things, data center, electric vehicle, artificial intelligence and machine learning. And finally, the driver -- Advance Driver Assist are leading up to some of the autonomous vehicles. And all these 6 megatrends are intersecting with our 6 end markets providing significant opportunity for growth. So I'll summarize it. We are a consistent revenue grower with multiple growth drivers across 6 megatrends of the industry. We have a high-margin business model and very shareholder friendly. Our Q2 current quarter guidance is for net sales to be flat to down 8% sequentially. We have paid down $2.62 billion of our debt over the last 8 quarters and substantially all cash generation beyond dividend payment is being used to reduce leverage. And we have a premium long-term non-GAAP financial model of 63% gross margin, 22.5% OpEx, and 40.5% operating income. And with that, I'll pass this back to John for some Q&A. John?

Great. Thanks, Steve, for the presentation and for the opening here. I thought it was great that you put in perspective. If you look at your performance on a year-over-year basis, right, you've outperformed a significant portion of your peer group. I wanted to kind of maybe start with a question along those lines. If you take a step back and look at your results in the first half of the calendar year versus even your expectations at the onset of this correction, right? I think it's fair to say that your results being down only 1% on a year-over-year basis have been much better than feared. I think it's been quite a while since we've had such major global shocks to these economies around the world. Obviously, you've highlighted many of these things, right? You've had large global regions around the world being shut down by extended periods of time, record high unemployment, factories being shut down for extended periods. How do you put this into perspective and kind of reconcile your results versus all these global shocks that we've seen around the world?

So I don't really know every other company's end market mix and we don't really know their supply chain, what the countries are in -- they're in and how their plans were affected. But one thing we have said is that during this cycle, we had a simultaneous supply and demand shock. And every company experienced it in different ways depending on where they were located and where their end market mix go up. So for Microchip, we had a substantial demand shock in the automotive business, some in the industrial business, some in the appliance business. On the other hand, we had a substantial demand increase in our computing business, in our data center business and in some of our communications business, a lot of work-from-home requirements. And in our medical business really was up quite significantly since our parts went into all the ventilators, thermometers, oxygen sensors, blood analyzers and all sorts of medical equipment, which helped. So somewhat everybody's performance will depend upon how their end market mix was, and secondly, how they experienced the supply shock. So we suffered from supply shock in 2 regions, Philippines and Malaysia, where the requirements from the governments were so strict that we couldn't bring people in and out of the plant. And finally, we landed up putting a lot of our employees inside our plant in Philippines and essentially locking down the plants where they were living and working inside the plants and make sure tents and PPEs and cards and sleeping bags and we bought the food in, and that's how we made the quarter. And we have really no idea who did what. And I'm going to try to download another slide here, but there's a slide I showed to one of the one-on-one audiences and all during the day that shows Philippines plant as a hotel, essentially. It shows all the -- how the tents are laid out, and carts are laid out, and how people are living and working from inside the plant and really showed the spirit and commitment of our people, how we made the numbers.

All right. Just following up on that, Steve, I'm wondering if you could give us an update on the supply situation in the Philippines. I think you're starting to see a spike in COVID-19 cases there. And I think there's some concerns that, that country may go back to kind of a lockdown state again.

So in Philippines, we have largely employees working and living in the plant. And we can increase that capacity further. So if there is a second resurgence, it's not going to be worse than the first resurgence. Essentially, whatever happens outside the plant in Philippines, we're fully capable right now to run the production by having our people work inside the plant -- and live inside the plant. Now if their desire runs out finally, they say, no, I got to go home, I mean, that's different. But so far, our employees have been very, very resilient. Many of them have been living and working in the plant now for months.

Got it. Steve, wanted to just get your perspective on your view on what kind of the broader recovery for Microchip do you think it looks like over the next several quarters or so. I know visibility right now is a little bit challenging for you, just given the aging in your backlog. It sounds like that is getting a little bit better. Several of your peers have indicated that they are building inventory. I don't know if they're signaling a V-shape recovery, but I think they are obviously more optimistic that things are going to stack back toward more meaningfully. What are your updated thoughts on kind of the pace or the shape of the recovery for maybe Microchip and maybe the industry?

I don't really know what the industry would do. I think it will be every company, the end market mix, how the end markets behave. So I don't really have a comment to the industry. As far as the inventory build is concerned, I think you saw in the slide I showed, many of our competitors' business has gone down much more than ours had. And they didn't cut production. So as their business went down further, their inventory is higher. Our business didn't go down that much, and therefore, our inventory is more reasonable. I think it's largely a result of revenue decline for the other companies why their inventories have grown. In the current quarter, we're guiding down between 0 and minus 8% for the midpoint of minus 4%. We are expecting our inventory to slightly grow this quarter also. But beyond that, we have a large amount of inventory packed in the die inventory form that we can rapid package around and test it [indiscernible]. We never find that we have to really keep the -- as large an inventory as many of our competitors are talking about. You don't really require that level of inventory. And as the business for them goes down more, there'll be streaming an inventory and taking charges over time that we do not want to do. We are managing working capital much more tightly, paying down our debt. And I think we have taken the step to educate our customers that they need to give us more backlog if they want a better service, they have done that. Our July bookings were excellent, our August bookings were even better. So even the numbers I talked about a week before, August bookings, with 10 days gone, has actually even done better than the average July booking. So our customers are responding to the letter and having the effect -- exactly the effect that we counted on. We're getting short-term backlog, medium-term backlog and also longer-term backlog. With that, I think inventory position is fine.

Got it. Just a follow-up on the inventory question and maybe the moderation in bookings activity you observed in May and June. I know you indicated that you're not aware of any sort of broad-based holding of inventory by your end customers and your distributor inventories where you have very good visibility is pretty lean at 30 days. But given the supply disruptions we observed early this year, isn't it reasonable to think that some of your end customers might have built a little bit of buffer inventory in March and April? And once supply began to normalize in May and June that, that might have contributed to the moderation in bookings that you had observed?

I think that's a very wrong assessment. That's not really what happened. The only customer that I think that has built inventory is Huawei because of U.S. sanctions and constant threat of what they may not be able to buy, I think the world knows that Huawei built some inventory. But beyond that, our tail of 120,000 customers don't have the balance sheet, don't have the credit to build the inventory. And our distributor inventory is very low. When the customers are placing orders, when there is a supply disruption and they're not able to acquire short-term parts, and we're asking them then to pay expedited charges, if they want to expedite the parts, if they were building inventory, they wouldn't be paying expedited charges to build the product. If they want the parts in 2 weeks, and I'd say I can give it to them in 6 weeks. And they say, can you expedite that for me in 4 weeks, they wouldn't be willing to pay expedite charges to get that. So I think the assessment that there's an inventory build because of the disruption, which is totally wrong. We see no inventory build anywhere.

Thanks for the clarification there. And I did want to revisit just kind of the short lead time orders that you observed in the quarter. Wanted to better understand what was behind this? Was this simply of your customers just having limited visibilities as bookings became weaker? Conversely, you had indicated that there were delinquent orders that you weren't able to fulfill in the quarter, right, related to the letter that you had sent out to your customers. I would've imagined that those orders that were delinquent and were pushed out by you would have given you better visibility into the September quarter, and naturally, would have increased the aging of your backlog. Is that not the case?

It did not. The September backlog was so low because of not getting longer-term orders last quarter. But even though we rolled that delinquency into the September quarter, the starting backlog on July 1, which included all that delinquent backlog we're holding, all that together was still 8% below the backlog for June quarter on April 1. So I think the way you need to think is that customers rather not placing long-term orders because then they have not skin in the game. You don't place long-term orders for your grocery, you go to the store and you buy it because the product is available. So if customers could buy all their product, ordering it week at a time, 2 weeks at a time, they'd rather do that. But customers know that there is a semiconductor build time, and if they place the order only a week away, in many cases, they will have factory disruptions because the product will not be available. But when there is a substantial change in the business environment with COVID-19 [indiscernible], customers sort of erroneously assume that the lead times were much, much shorter, and there was a large amount of inventory, and they placed the order asking the part 2 weeks from now and the part would be available. And they were not because we held the inventory at the die and it takes 4 weeks plus to really build the parts into the finished good. So customers just basically because of dispute impact of COVID-19, didn't want to make the commitment and took their orders in very short term. And as a result, they were not able to acquire parts. Now if I had 20 customers, I'll have salespeople just to visit them and tell them what's going on. In the case of 120,000-plus customers, we cannot communicate with them in any other way than to write a letter and post it on the web. The letter is always written for our customers. And 100% of the time the response from our customer has been as desired. And I do not know why analysts and investors read that letter, almost 90% of the time, you guys make the wrong conclusion from the letter. It is written for the customers and have the desired effect. But somehow, Street assumes that the letter meant there was just a lot of upside, a lot of orders and a large number of orders, and it meant something positive. So the next time I write a letter, please don't pay attention to it.

Okay. Just a couple of questions for Eric to wrap up here. On the gross margin front, you've done a pretty good job executing there, despite the fact revenues down, you're able to hold gross margins flat. How much of this is kind of mix versus operational cost savings you've been able to continue to extract? And then assuming we get growth off of September, is there -- can you remind us in terms of the revenue level or timing in terms of when you get back to kind of your targeted kind of long-term gross margin range?

Sure. So I think the gross margins and operating margins, as Steve highlighted on that slide earlier, absolutely a highlight for Microchip. The results on gross margin are always a mix of operational efficiencies that you mentioned as well as what the mix of the products that are being sold are. But we've done a really good job of controlling costs, continuing to make improvements in the business. We've got a lot of self-help stories in terms of what we're working on and internalizing more of assembly and test. Back in November of last year, we announced the restructuring of our Colorado factory and moving some of that high-volume production into our more efficient factories, 8-inch in Arizona and in Oregon. Yes, those projects all continuing on pace and we're making improvements in the business. So that's really the main piece there that we're continuing to cost effectively run our factories, make improvements to drive gross margins. And also we're running the factories a little bit harder in wafer fab, in particular, in the current quarter. We had our employees on rotating time off schedules in our 2 larger fabs, and that's been removed at this point in time. So our underutilization charges are a little -- expected to be a little bit less in September as compared to June. And then the second piece of your question, John, was what?

Just target gross margin, just time frame, revenue level, 63%, right?

So it really depends on the slope of the revenue. We're not able to put a revenue number around it. But we're at 61.7% in the current quarter based on midpoint of guidance. Our long-term target is only 63%. Last quarter, we had $13.9 million of underutilization charges. So if that went away, we'd almost be there already. And then plus some of these other improvements that I mentioned, we've got high confidence in being able to get to that 63%. But not quite ready to put a revenue range around that in terms of when we'll get there.

Okay. And then just to wrap up here. Just on the OpEx front. You mentioned you're paying your employees a bonus this quarter given the strong performance, but keeping the temporary pay cuts in place. What do you need to see before you're comfortable removing these temporary pay cuts? And then what's the right way to think about your new baseline once things start to normalize as there's likely going to be some costs that just don't go back into the model anytime soon, such as I would imagine, travel.

We committed to our employees that the pay cuts will end on January 1, unless done earlier based on seeing stronger recovery. So we saw a pretty good June quarter. So we gave a lot of the pay cut back to the employees. We are maintaining the pay cut in September. If September is better than our anticipation, we could do something similar. So in a way, we have taken either 2, where we have a safety net of pay cuts if the recession is deeper, but if we do better, we can return it to them. Our employees are pretty happy, comfortable with that. And the pay cut will naturally go away at the end of December, unless done earlier by an action from us.

Great. And with that, I think we're out of time. Thank you, Steve, and thank you, Eric, for joining us.

Thank you.

Thanks, John.