Microchip Technology Incorporated (MCHP) Earnings Call Transcript & Summary

Great. We're going to get started. I'm pleased to have Microchip with our next session and we have Ganesh Moorthy, who's the President and CEO; and we have Eric Bjornholt, who's the CFO. So I'll turn it over to Ganesh first.

Great. Thank you, Tim, and good morning to everyone. Before we get started, we wish to remind you that during the course of this morning, 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. And we refer you to our recent filings with the SEC that identify important risk factors that may impact our business and results of operation. Three quick things before we go into the Q&A. So first, our business over the last month has progressed as expected with -- and it's consistent with the commentary that we provided during our earnings call. We're in the midst of a classic inventory correction for semiconductors and superimposed with a weakening macro environment, which has amplified the down cycle. All markets with the exception of aerospace and defense in all regions of the world remain weak. Customers continue to adjust their demand expectations and derisking their inventory positions and taking advantage of the fact that there is ample supply and very short lead times that they can rely on at this point in time. While the short lead times, we believe, are the best way to help customers and Microchip to navigate an uncertain environment with more agility. And in the near term, it does result in lower bookings and reduced visibility for us. Notwithstanding any near-term macro weakness that might be there, we're confident that semiconductors remain the engine of innovation in so many different applications and markets we serve. Our focus on total system solutions and specific market megatrends that are important to us, is fueling very strong design win momentum that we expect will drive above-market growth in the long term. And our business has historically demonstrated consistent and resilient cash generation gross margin and operating margin characteristics. We remain confident that our non-GAAP operating margins on a trailing 12-month basis should remain above 40% through business cycles. Last but not least, we remain resolutely committed to our capital return strategy since we achieved an investment-grade rating 2 years ago, November 2021 and pivoted to increasing the capital return to shareholders. We have returned in combined amount of that area since then, $3.25 billion to shareholders through the September quarter of this year, a combination of both dividends and share buybacks. Our capital return to shareholders in the December quarter will increase to 77.5% of the free cash flow, the adjusted free cash flow we generated in the September quarter, and we're on a path to return 100% of the adjusted free cash flow by the March quarter of calendar year 2025. And with that, I'm ready for your questions and so is Eric.

Perfect. Great. Thank you. So lead times are going to be under 8 weeks by the end of the calendar year. Customers are still working on inventory, and this is what happens when lead times compress this much, this quickly. But I think all of us are looking at some of the KPIs and the indicators as to when the bookings environment and -- environment might start to improve. So can you talk just about some of the KPIs that you watch and whether you're encouraged by near-term booking activity and maybe as that provides a look into 2024?

The KPIs that we watch are not that different. We're looking at bookings, we're looking at backlog, we're looking at aging. We're also looking at our customers asking for short-cycle shipments -- are at some point in time, are they looking for pull-ins from what they had placed orders out of time to meeting something quicker than they had expected. So those are all the normal parts of what we look for in the business. It's too early in the process at this point to say that those are changing significantly. Customers are still in a state where they don't have clear view on their own demand. And because of that, they are taking a more conservative posture. Our bookings are low as a result, but we also know that bookings will come back. Consumption is where we expect it to be and to, at some point, support that consumption, inventory will drain and we'll see the backlog coming in. But we will at first see a lot of people asking for short-cycle response to their demand. And that's what we're positioning for. We're trying to take our products, our inventory and put them in the right places that we can get the short lead time available. And I think that gives our customers the best position to be agile to the market, gives us the best position to respond to a quick changing environment that we expect.

Great. I wanted to ask a couple of questions about PSP. There's some controversy in the investment community as to whether it has ultimately been good or bad. PSP was born out of the customer demand for supply. You made financial commitments, but you also need to some degree, keep the customer on the hook, so to speak, for those commitments. So really, I had 2 questions. The first is, as we are really at the bottom of the cycle what did you learn from PSP that can make the company better going forward?

Sure. So PSP for us is an unqualified success. And for the customers who participated in it during the time of extreme constraints, the priority they received was a large factor in their success and what they were able to do. Now PSP also is a program or like any program that needs to adjust over time. The conditions in which we launched it, almost 3 years ago in the early 2021 time frame. And that persisted through '21 and '22 were different. And it was a period where demand was 3x, 4x supply lead times for very long. The market was growing at a tremendous pace. And so as that has changed, our program has also changed. Now to make that priority commitment to customers, we needed to make sure that there was a even exchange of what we were trying to do with investments and what customers were trying to do with commitments. That's what PSP was born on. At a time when lead times were 52 weeks, that investment time frame from a customer was going to be over 52 weeks of time. As the lead times have come down, we have brought down the lead time over which or the time over which customers need to make those commitments. They're much shorter today than they were. If the program is still something they find valuable. For many customers, it still is, especially for customers who make very highly valued end products and who have a much more resilient demand picture that they see. For them, they continue to be very much invested in PSP or other supply programs that are there for the long term. Obviously, along the way, environment change and demand picture changed and demand supply has changed. And for us, it's always been about how could we have how early could we have adjusted, when do we know what we knew. It's a constant learning process. And was there some things we could have done maybe a month or 2 earlier than we did. Sometimes it would have been. But we act upon the best available facts at any given point in time. And of course, 3 months later, 4 months later, sometimes facts can be a lot more clear. But we're not trying to Monday morning quarterback where this is that we have evolved the program to provide the flexibility and we have added far more flexibility on the reschedulability of backlog. So the PSP was noncancelable and nonreschedulable. We have been pretty consistent on the noncancelability but have had much more flexibility on reschedulability. And even on both those items, what we have done is really gone to saying, can we find more win-win approaches, where it isn't that all the benefit in '21 and '22 goes to the customer and all the shortfall comes to Microchip when supply demand come into balance. And that win-win has taken many shapes. Each customer we work out with something that is good for them and something that is good for us and adjust where the backlog will be as a result.

And so what is your hit rate in the scenarios where a customer inside a PSP wants to push something out? And of course, you're not going to let them cancel but you're going to work with them. Do you -- in virtually all those cases, are they willing to give you something of value and return? And can you give an example of like what that might look like?

And every situation is different. It depends on the customer and where it's at. But yes, we are the ultimately, and it takes discussion to find where it would be. So it could be that demand was placed on product lines, A, B and C, which they don't need as much of, but they have opportunity to do more on product lines, D, E and F. We're going to do swaps with them. We're willing to get consideration on a new design that we may not have had we not had some form of a supply agreement with the customer. So it takes different forms with each customer. We got to work it out with them so that it's something that they would need and value and something that we could live with and value.

And just second question on that. There's an argument that gets made to me that I hear. Why not just let the customer out of PSP and force them to come back to the table and place a new booking? I guess from your perspective, you want to keep some leverage because you did make a commitment to them. But is that really the reason why you don't want to let them out of PSP and make them place a fresh booking? Because if they did, then you might have a better real-time signal on demand.

So it's not about leverage. It's about making mutual commitments, right? We took PSP and the backlog that came with it and made substantial investments. Investments in our factories, investments with our suppliers, investment in inventory with parts of the supply chain that provide us products, et cetera, in it. What we want is to make sure that we don't evolve into practices as we go through this that then create complete lack of responsibility in subsequent cycles as well. If you know that, hey, I can make noncancelable commitments and then at the drop of a hat, those all go away what we would, in fact, see is backlog, even higher demand even because people will say, "Hey, there's no risk [indiscernible]." We need a balance where there's responsibility on both sides for what we're doing, and there's a mutual win from making that commitment.

Great. Let's actually talk about pricing, if we can. Can you talk about your philosophy on pricing? I think a lot of investors worry that this is some shoe to drop. Is that a valid concern? It seems to me that you're just passing along cost increases. But can you talk about pricing.

So pricing has always been a strategic exercise for Microchip. We are selling customers products that are proprietary products that are sole-sourced products that have a long life cycle ahead of them. If we were capricious in our pricing, we would not have the trust for those customers to keep using us in new designs. So go back in history, 20, 30 years of time, it has always been thoughtful and long term in the way that we have worked with our pricing. Now we had a very unusual 2 years in 2021 and 2022, where the rate of inflation, the rate at which structural costs were growing was far, far more dramatic than what normal cost efficiencies were able to absorb, and we and the industry at large. And so in that period of time, we absorbed some of that cost, but we couldn't absorb them all. And so we passed along some of that in the form of price increases to customers. Those are structural costs built in. So they're not short-term price adjustment that all go away when you have capital equipment, when you have factory expansion, when you have people costs that have gone up during this period of time, those are structural costs that we don't see just saying turn off the switch and it goes away. But we don't expect price increases are a consistent theme going forward. It wasn't in the history. I don't think it is in the future. But there was a couple of years where it was at a rate that was higher than normal. And we do need to get back to where there's pricing consistency throughout the supply chain and peace of mind for customers when they make proprietary design choices.

And this is all negotiated at the time of design anyway, and it's not like you're giving price reductions just because demand is weak.

So that's another great point, which is pricing is at the point of design end. That's when decisions are being made from design end, there's often a 12-, 18-, 24-month time before someone goes to production and then perhaps a 5-, 10-, 15-year window of time in which they are in production sometimes longer. So the intensity of price and the competition is at the point of design, it's not what we're shipping today or if we're going to be shipping in 3 or 6 or 9 or 12 months and where it's at. But at the point of design, we also have new products that are much more price competitive, which have more features that can lower the cost of ownership and where we're at. So we're going to compete at the point of design with the most cost-effective solution that we have, not with some historical product that is in production today.

So just to wrap up that little topic. So is it fair to say that the only downside risk to pricing really would be if foundries lower their price to you, then you'd pass that on to the customer.

Yes, if that should be -- I mean that's a great if question foundries on -- they have the same challenge as an IDM. We know what it has taken to build capacity in our factories. We used to buy equipment for our factories at pennies on the dollar. We often bought them in the secondary market because there was somebody else whose capacity was rolling off. In the last couple of years, most of the capacity we bought has been at a premium to normal pricing, either because prices went up at the equipment makers or we ended up buying 12-inch equipment to run on 8-inch fabs and where we were at. Those same forces of what our foundry partners have run into as well. Labor costs that they've had to endure to be able to hire enough people to get them trained and going, et cetera. We've seen that in our factories as well. So I don't think the foundry pricing other than perhaps spot pricing at any given point in time, which is really not predictable pricing. And if you wanted -- if they have excess capacity next quarter, you can go find capacity at a discount today, but it's not something that is sustainable, it's not something that is predictable.

Got it. Eric, maybe I can ask you about utilization rates. You've taken them down modestly in the fabs over the past few quarters, but still not at a level where you're taking underutilization charges. So can you talk just about the evolution of utilization? I think you've taken it down more on the assembly and test side. And maybe can you also speak to the potential to take some inventory write-downs as well.

Right. So we're about 40% internal source for fab today, and the other 60% is outsourced to the professional foundries. With our internal manufacturing, we are still running the factories at a relatively high rate. It's not at the 100% utilization that we are running at before. So we've had attrition and that utilization is coming down but aren't expecting any material underutilization charges in the current period from that. We are managing it in a way where we are building inventory through the wafer fab process, taking the product through probe and then holding it in die bank, and we'll take it through the assembly and test process when we get an order. So assembly and test has ratcheted down. We actually drained finished goods last quarter. Our own utilization will continue to probably ratchet down modestly as we have attrition but for us, from a cash flow perspective, we feel it's better to build the inventory, have it ready for the next up cycle. And it actually saves us from deploying capital at an earlier stage than we would otherwise if the inventory wasn't in place. There's such a high fixed cost associated with the wafer fab that if we were to cut capacity utilization, we might be able to take up 25% or 30% of the cost. And so we'd rather build the inventory today. And we'll continue to monitor the status of our inventory position and utilization as we go through the coming weeks and months.

And we're blessed with products that have very good longevity, very low obsolescence risk. These products -- and we've in prior cycles as we have built inventory, these products are needed. They're going to be needed for the next 10, 15 years of time in them. So what we're building maybe excess of demand today, but is going to get consumed as the cycle changes.

Right. I just had a question on sort of the guidance as it relates to the trajectory at which things are worsening. And the question really is around whether you assume that things will get worse. And it sounds like you did because you also are saying that, well, the bias to Q1 also is below seasonal. So you're already baking in the fact that things are likely to continue to get worse even into Q1.

That is correct. We have indicated directionally that Q1 will be a decline. That will be less than the magnitude of what calendar Q4 as a decline was. And beyond that, I think we haven't provided any other guidance.

Great. Can we talk a bit about competition in China? It's a topic that we could spend the entire half hour about. But can you talk about that? I mean, when you really winnow it down, it doesn't sound like maybe there's a mid-single-digit risk to your revenue, but it doesn't sound like it's ultimately more than that. Can you kind of play through all those pieces?

So firstly, competition in China is not new. I know it's fashionable to talk about it in the last 6 to 12 months. We've had competition in China for 30-plus years, right? In the early years, there were people that were copying products, cloning products, et cetera, then there were other players in it. The general ability to compete in the microcontroller marketplace requires players who have a long patients horizon and can deal with substantial fragmentation of the business, right? There's not a single product from Microchip and the broad base of products we supply, that is more than a fraction of a percent of revenue where we're at. So to gain share for someone in analog, in microcontrollers and the broad-based product lines we have, it's hand-to-hand combat to be able to go get it. And that fragmentation is not a patient's horizon that many players have. What they would like to get is a quick hit on something which is high volume, maybe it's in the consumer electronics space, maybe it's a power supply for a cell phone or something else along those lines. And some of it will be in the standard microcontrollers as well. But the vast majority of what we do is so fragmented that the patient's horizon it takes to develop the product, win the design, wait till it goes to production, ramp and get out of everything, it's not an easy process to go into. That's been how it is for 30-plus years from where it's at. Now that said, when you look at our business in China, our estimate is about 20% is in Mainland China and off that 20-ish percent, about half of it is designed outside of China. So the point of design influence is not inside China, it's outside China. The other half that is inside China -- designed in China has about another half of it, which is substantially complex products, which is not what the competition is doing. It has much more application knowledge, software tools, et cetera, that a much higher degree. So there's maybe 1/4 of our business in Mainland China, that would be the broad based. And that broad base is highly, highly fragmented in where we're at. But also those products often lack the ability for people to -- who have more reliable applications who know that the specs have the reliability, the likelihood of supply, the availability over the cycles is going to be there that they say, "All right, do I take a reliable source I've used for years? And do I just go take a Chinese alternative that's there." There's going to be cases that it works in, but it's largely in the lower end of the markets, the lower margin end of the market and the ones that we have been less sensitive to in the past.

I guess also maybe is another factor to consider that the workhorse, the workhorse for your fabs node wise is that 110-nanometer and above. And they don't seem to be adding very much capacity and really investing in those nodes either. Is that another factor?

So I don't have enough knowledge of all the different lines. I mean they're obviously investing in trailing edge nodes, it would make sense that they're trying to invest in some of the 300-millimeter rather than 200-millimeter. But I got to believe some of that investment is getting spread into many different nodes. And we haven't really spent time nor have we seen a specific competition that is at 110, 150 or 180-nanometer versus others.

Great. And maybe as part of that, and not just a China comment, but can you talk about total systems solutions as well? That's another factor that seems to insulate you from being cannibalized in China as well. So maybe asking the question, not just related to China, but just as TSS, the program is evolving.

So Total System Solutions is an approach of taking all of our assets silicon software systems. The services that we provide to enable a customer to utilize everything that Microchip has and to succeed as a result of doing it. So our product line has got the capability to be early designed for almost every customer. So early designs that a customer -- early decisions a customer has to make are around the brains of the system. That's where the controllers, processors, FPGAs, dsPIC are all key decision-making. And because we engage the customer early, we're able to identify all the other pieces of the puzzle that the customers are going to need. It doesn't mean we'll win it all, but we can very early begin to position if you need this kind of -- if you're making this kind of application, you're choosing this kind of brains for the system. Here are the other things you're going to need, and here's how we put it together. And in many cases, we will bind them together in terms of how our software, our validation, our reference designs come together to lower the risk for the customer of using our bundle to make it faster for them to get to market and to be able to get total system cost out in the process of doing that. So it is an approach that allows us to take a larger piece of a customer's design and bring benefits to the customer in the process of doing that.

And what portion of revenue do you think is moving in a TSS type of arrangement?

Almost no new design takes place today without our field approaching it from that perspective. Nobody is looking to go get a single chip on a design. If that's all we can get at the end, that's fine. But we go at it with a new design, both us and our partners that we work with on how do you get the maximum amount, taking advantage of all the parts of the portfolio that are going to be needed in the fulfillment of a given application.

And the company has been on quite a path since 2010 when you began this M&A strategy at the time you were thought of as a controller company, and now you do many, many more things. Can you just talk anecdotally or qualitatively about how your breadth and scale helps you in these discussions with customers and where maybe a particular end market where you've got already a big share because you have this breadth, it's actually helping you gain share in other end markets?

So the acquisitions gave us breadth of product line that I think is unparalleled today in terms of where we're at between what we do for controllers, processors, FPGAs, ASICs, dsPIC and then a tremendous amount of analog and mixed-signal capability, that today, analog mixed signal is approaching 30% of the revenue of the company. And most people would not necessarily think of that as what Microchip is known for. But that has fed into our ability to go to customers, engage in designs and win with more complete solutions. So it has been an unqualified success in the TSS strategy that we've been able to put together is the combined assets that we acquired through the acquisitions and have then since continued to invest in, have given us that strength and breadth of portfolio to go after many more applications, many more end markets with more complete solutions that we have done.

Great. Maybe we can talk about capital return. You've got a very methodical strategy. You're going to be at 100% of free cash flow by March 2025 at this rate. Two questions. First of all, during the cycle, have you ever reconsidered that strategy?

We've never reconsidered this strategy. But the Board continues to discuss it on an ongoing basis in terms of adjustments we might make along the way. And so for example, as the interest rate environment rose, we were more thoughtful about how should we -- how quickly should we get to 100% [indiscernible]. And there was lots of feedback from many investors, hey, go there and go there quickly and we wanted to get there more methodically over time, taking advantage of the fact that a portion of the debt could continue to be retired. The mix between dividends and share buybacks has also been thoughtful about how do we get dividends to a point where it can be approximately 50% of the capital return on a long-term basis and where we're at. But I think the fundamental fidelity to getting to 100% capital return, we've been unwavering on that since we made that change.

And just the basis for pursuing the strategy. Is it feedback from shareholders? Is it seeing the valuation that the stock gets relative to some of your peers that pursue that sort of an approach? What was the basis when you -- when the Board was discussing this. What was the basis for pursuing this?

So there are many ways in which your free cash flow can be allocated, there's debt reduction, there's acquisitions and share buybacks, there's dividends for fundamental ways besides whatever you need that is already in your business to run your business. As we brought debt down and brought the leverage well below the levels that people were concerned about, it opened up some optionality. As we looked at the acquisition market and said, where are we from a scale and breadth of product line, we didn't see that as being as critical as it was 10 years before when we were on that path plus acquisitions have gotten far more expensive than when we had gone through them. And the regulatory environment has gotten far more difficult as well in that whole period of time. So if you pull the large acquisitions out, if your debt is down to a level where it's manageable, then capital return becomes a more important part of how you would deploy the free cash flow. And the Board then had to be thoughtful about how we do dividends versus share buybacks, and that was the announcement we made about 2 years ago and kind of the path we would be on and then made share buybacks another consistent part of the capital return strategy throughout this time frame. We were not doing consistent share buybacks prior to that. We have done it opportunistically, but that now became another element of the capital return.

Great. Well, we're just about out of time. So again, thank you, Eric and Ganesh. Thank you so much.

All right. Thank you.