Microchip Technology Incorporated (MCHP) Earnings Call Transcript & Summary

All right. Good morning, everyone. Thanks for joining. I think we're live now. So I'm Chris Caso, Raymond James semiconductor analyst. So thanks for joining us at our institutional investor conference. Our next presentation is Microchip Technology. With us from Microchip is Ganesh Moorthy, CEO. And I guess that's the first time I could refer to you as CEO. And I believe, unless you did a presentation yesterday, this will be your first investor presentation as taking over the CEO helm. So congratulations to that, Ganesh.

Thank you.

And we also have Eric Bjornholt from -- Microchip's CFO, and Nawaz Sharif, the VP of European Finance and IR. So gentlemen, thanks for joining us. Ganesh, I'll turn it over to you if you want to make some opening comments. I guess before I do that, I'll also point out that there is a Q&A feature for the investors joining us. If you look at the upper left-hand corner of your screen, you can click that button and the questions will be e-mailed over to me, and I'll be happy to address them. But before I do that, I'll go into some Q&A. But I'll pass it over to you, Ganesh, if you'd like to make any opening statements.

Great. Thank you, Chris. Before we begin today, I wish to remind everyone that in today's chat, we will be making some projections and other forward-looking statements regarding the future financial performance of Microchip. These statements do involve predictions, and the actual results may differ materially. I refer you to Microchip's filings with the SEC regarding some important risk factors about the company. So let me start with a brief update about the business environment, and also consistent with what we have communicated with investors in November, our forward going practice is that we are not providing a mid-quarter for financial update anymore. The business environment continues to be strong with bookings and backlog continuing to set new records. As a reminder, the record levels of bookings and backlog are not an indicator of about just current quarter business, but also a lot of the backlog layers in into the next 12 months as well. And this provides us with more visibility than we normally have, which is very helpful in a constrained environment, to use our capacity more effectively. Demand over the next 3 to 6 months is running considerably higher than our capacity to service it, despite our bringing on more capacity. And as a result, our lead times have extended on a broad base of products, ranging anywhere from 4 weeks to over 40 weeks at this point in time. The exact lead times are product-specific and dependent not only in what the capacity constraints are, but also what the demand intensity. On January 1, we changed our noncancelable window from 45 days to 90 days in order to give us an opportunity to better utilize our manufacturing assets as well as to discourage orders being placed in excess of what is truly needed. Two months after implementing this change, it has worked out as we expected, and will continue to be our standard practice. Four weeks ago, based on broad customer feedback, we launched the Microchip Preferred Supply Program, or PSP. This program offers our customers the ability to receive prioritized capacity starting in the seventh month where they place 12 months of backlog that is noncancelable and nonreschedulable. And after that, as long as the customer continues to place an additional month of backlog every month for a rolling 12-month period, they maintain the ability to get prioritized capacity. Response from our customers to join this program has been swift and broad-based. And at this point, we have over 1,000 customers who committed to the program and still more signing up. We expect many more customers will take advantage of this program in the coming days as they work through their internal procurement logistics. A few words on the supply side, with the strong demand and low levels of inventory, both within Microchip and at our distribution channel partners. We are experiencing constraints in all our internal and external factories and their related supply chains. We started ramping our internal factories in September as well as investing in capital additions to expand our internal capacity and we also work with our supply chain partners who provide us with wafer foundry assembly test, materials and all that to secure capacity. Through the combination of internal and external capacity actions we have taken, we expect our overall capacity will continue to grow every quarter in calendar year 2021. And while our capacity will continue to grow every quarter, we also believe that wafer fab as well as assembly and test constraints, are here to stay with us through all of calendar year 2021, possibly into calendar year 2022. So that concludes my update. Let me pass it back to Chris for the Q&A session.

All right, Ganesh. Well, thank you. And that's a very helpful update, and we'll hit a lot of those points in the Q&A.

I guess I have to start, as again, we're honored to have the first public presentation with you is as formal CEO, although you're not new to Microchip, not new to investors. I have to say, every person, every manager has their own style, and maybe it's a way to start of -- what can we expect differently? What might be something that might be different? And Microchip, I don't know, if it's 2.0, 3.0 under Ganesh.

So as individuals, we all have different styles. And I think over time, rather than my tell you what my style is, you will form your own opinion and my style is, and I'll leave it to data there. I think what matters more than style is substance of how we lead and operate. And by that metric, I think you'll find that what you saw out of Steve and what you'll see from me is very much aligned. We have a shared set of common values. We charge the Microchip culture, and you will see that stability for what we've done. The hallmark of our last 5 years, where I was President and Steve was CEO, working in a 200-block structure was a seamless leadership structure for the enterprise internal and external. And you should, therefore, not expect any changes as I start my tenure as CEO. And obviously, as time goes on, we will face new challenges, we will have new opportunities. And they will be different from the past. And our response will adjust, and you will see how we evolve as a company when those things come upon us. But nothing earthshattering that's taking place today that is different from what it was last week.

Very good. Maybe I'll follow-on with just some questions related to your opening statement, and one of the interesting things, obviously, is that you are adding capacity as you expand -- expect to expand capacity each quarter throughout the year, but you'd expect demand to be higher than that. And obviously, in semiconductors, there's always that lag between when you start to plan for the capacity and when the capacity gets in place. So maybe if you can talk about when the planning had started and that sort of time lag for when you -- the industry and yourself started to understand that, boy, we might be tight supply and when that supply could actually be realized.

So it's different for the different factories. They have different lengths of time. So we began to more aggressively go after a capacity that was needed for the strength in business we were seeing starting in about September. By that time, we were wanting to get all our factories running full-bore, starting to hire for growth, we began to order equipment. And initially, a lot of that was in the back end for the assembly and test, and those you can generally affect in a shorter period of time. If you need to go get fab capacity, that takes a lot longer to be able to get. And of course, as we went through September through the end of the year and through January, all that demand is beginning to pick up. So we could no longer get equipment in the lead times that we were seeing back in September. So equipment lead times have stretched out, and all of that kind of goes into the equation of what capacity can come on and when. Some of it is coming on this quarter, some is taking longer. But in general, many pieces of equipment today are taking almost twice the lead time it was back in September to what we can get today. And so those are affecting the rate at which we can bring some capacity on. But fortunately, we got this thing started early, and we are making progress every quarter with our capacity.

Is there a way of quantifying the amount of capacity that we'll add this year? In other words, if we look at 2021 versus calendar '20, what your kind of revenue capacity would be from an internal perspective? And I suppose from an external perspective as well in terms of the amount of wafer allocation you can get from foundry partners.

Yes. So that has many components, one -- with some which are in our control, which is what are we trying to do and buy for our internal factories; some which is a little out of our control, which is where are we fighting for capacity in a constrained environment that a fab and a foundry isn't going to have a new fab or a new expansion that they're going to be bringing on too quickly on it. So I don't know. There's not a good, easy way to give you that sense of it other than we can see our way to how our capacity will grow every quarter. Some of those are also affected by when are we able to get equipment to come in, and some of that has come in as we had expected, some is coming in later than we had expected, but I don't have any clean way to give you a -- this is the revenue capability we have. We are fighting every single day on how we make that and make it better, and we have a significant amount of unsupported customer demand that we would like to support. And we make progress as each improvement comes on board.

And I guess when it happens, the age-old question in -- is the sustainability of that and the industry. Obviously, because of the long amount of time it takes to put that equipment in and the limited amount of visibility that you have at different points, the industry just overshoots in terms of capacity and that's the down cycle. Is this time any different? It would seem to me this time that as compared to what we've seen in some of the past ones that the extent to which that the industry is short, and yourselves included in that, seems to be more than what we've seen in some past cycles. Would you agree with that statement?

Absolutely. I think both the intensity of demand and the disconnect between supply and demand has never been this large. Certainly in -- I've been in this business 40 years, in the 40 years that I've seen this thing. So in that sense, it is different. But we also know, we're in a cyclical business and at some point, demand and supply will come back into balance, right? Everybody acting in their self-interest works towards that. And we are putting programs in place that allow us to have some degree of intelligently how we add that capacity. So the Preferred Supplier Program is one way in which it also gives us [ indurability ] with that demand so that we don't have it visible today, but it goes away in 6 months in what we're doing. So we're taking some steps. And certainly, we're not trying to add all the capacity all in one shot to service the unsupported demand today, right? As we did back in 2017, we sold some of that every quarter and incrementally approach how we bring that supply and demand together. You'll never get perfection in that, but we're being prudent in how that is happening.

Right. And maybe you could discuss that preferred supplier program. You spoke about 1,000 customers that had signed up for it now. And I suppose the customers don't need to sign up for -- to put all their product in that program. They can select which products they want to put in that program. So how do they handle on your end? And how do you handle on your end to make sure that they have that supply available once they're in the program?

So -- what the Preferred Supply Program does is it gives customers priority on available capacity, right? So if all of the preferred supply demand exceeds what our capacity is, we won't be able to service all of them. But certainly, we are giving those customers priority. It's farther out in time, so it's really months 7 through 12 where typically, not all the backlog is spoken for and capacity is still out there that is yet to be filled in what we do. And we are working to get all of that not only committed, but into our system, into our planning process, into the investments we need to make, in materials and people we need to hire and capital we need to buy and all of that. And we believe that it will give us the visibility we need to be able to get an early signal on when there is a change in the environment. Also give us time over which we can adjust as we see those changes coming in, during which there's nice demand that is in place that is noncancelable and be able to manage more effectively with respect to that. And so -- and right now, customers love it. This is exactly a solution for a problem they have been wanting us to go solve, and the general feedback we're getting from customers is lots of positive feedback and no one else is really providing them a solution that gives them what the PSP program is offering them.

Right. So I guess in the past, there are opportunities to be some sort of a take-or-pay program for certain products to have dedicated supply, but this seems like it's a little bit different then, I suppose?

Yes, it is. We're not forcing anybody to go into it. It's really a customer's choice to take advantage of the Preferred Supply Program, the priority that it affords them. We're not forcing anybody to do all of their backlog. They can pick the parts of the business they feel more confident in and say, "That's what I want to put on PSP." And the other parts where maybe there's more uncertainty, they can decide they don't want to do it. There are certain customers and end markets who have more visibility, and they will do more of it. There will be customers in other end markets that may not have as much visibility, and they may do less of that in what they do. So it's a flexible program that addresses how customers are preparing for their requirements out in time, and it gives them a choice and gives us the ability to then take their choice to go to work on giving them that priority.

Right. And from what you're seeing so far, is that what you're seeing, is that customers are signing in that program for business that would generally be more stable? And certainly, you serve some end markets that have a lot of volatility to them, depending on end markets, and some end markets that are very stable. Is the more stable stuff what's principally getting signed up for with -- under that program?

Yes, I think by definition, there are more players who would have more confidence in their ability to look at it. But I think there are also some people who are surprises to us, but they feel very good about their business. And they are in segments outside of industrial and automotive and aerospace and defense, and they look at this as the way in which they are tackling their market. Perhaps they want to gain share at a time when there's uncertainty and be able to build their products. So we have let all segments play to what they think is the right answer for their business and their understanding of where demand is going to be.

Right. With the capacity expansion that you spoke about, what should we expect for CapEx over the next year? And I suppose this PSP program is helping you to plan that somewhat. But should we expect CapEx still within the normal boundaries for Microchip just as revenue goes higher? Or is CapEx going to spike up as a percentage of revenue as a result of what's going on?

Let me give the ball to Eric and let him speak to it.

Yes. So our CapEx has been extremely low in fiscal '21, which ends at March 31 and then fiscal '20 also. And so we've kind of given investors a bogey of 3% to 4% of net sales in terms of where CapEx will be over time. And it's possible that it could be a little bit higher than that in fiscal '22, which starts in April. So we're making the right investments that we need in the business. And capital is tight now also in terms of lead times on equipment. So we're moving as fast as we can, and we'll do our annual operating plan this quarter and give you a specific number for fiscal '22 when we get to our May earnings call. But likely, you should expect fiscal '22 CapEx to be on the high end of what you've seen for Microchip in terms of percentage of revenue and that general target we've given of 3% to 4%.

And fiscal year '21 is running a little light because we can't get equipment as fast. So some of what we had planned in '21 will roll into '22, and I don't know if anything rolls out of '22 into '23. So some of this, I would say, on a quarter boundary, can move around. But generally, as Eric said, we will have more capital spending in fiscal year '22.

I did notice we have a question just following up on that. And the question is asking you quantify commitments to the PSP program in terms of percentage of your business that you expect. Portion of customers and portion of your business that you expect to be on the PSP programs, is that something you're able to predict to have any visibility on now?

So we are getting early visibility, but we're -- nothing that we are able to provide externally. So I wanted to give you customer count, just to give you the sense of breadth to where these are. And often, many of these customers are some of the larger customers who have the financial wherewithal to be able to make it. So they do have large dollars that go with them. We have a long tail of small and medium-sized customers and they would not necessarily be the ones for whom a PSP program would make sense. Maybe by the time we get to the May conference call, we'll have a far better view to provide for you on kind of how has all that laid itself out in terms of our backlog. And it's still going on. So we're not done. We're still processing how the PSP backlog is coming into Microchip customers. In some cases, they have more than 1 layer between them and how they purchase. So they may have a contract manufacturer, and they may have a contract manufacturer who buys through a distributor. And how the PSP program gets translated all the way to the buying entity takes some time for where we're at. So we know what the intent is, but we may not have completely executed it because how they buy is not always directly from us.

Understood. We do have another question in, and it was a question about your plans to move down the process curve internally down to tighter feature sizes. Do you stay at 0.25? Would you go lower? And maybe as a follow-on to that, you can talk about the decisions to insource versus outsource, particularly on the front end. What's your intention? And especially given now the acquisitions that have kind of increased your footprint somewhat.

So the percent of revenue we run in-house through our internal fabs is around 40%. And it's been in that range since the Microsemi acquisition happened almost 3 years ago. It used to be closer to 50% before that, and it used to be slightly higher than that before the Atmel acquisition. So each acquisition has brought a higher component of external foundry wafers that have become part of the total mix of where we're at. And a lot of when it comes in often comes in at technology nodes that are not suitable for our internal factories. And so there is no intent to bring those in-house because they run best in the factories that they're running in, that's where they're at. But we do continue to have organically developed process technologies that allow some of our products to, on next generation get more features, more cost reduction and so on and so forth. And so on an ongoing basis, we have continued to bring smaller and smaller geometries into our internal fabs. But there are 8-inch fabs, so they have to be technologies that are appropriate for an 8-inch fab and for the cost -- the economics of an 8-inch fab and what it does. And we are still bringing new technologies into our internal fabs that expand our capability, allow us, in some cases, where there's high volume to create a second source to an external foundry. But for the most part, most of our product, if they run the foundry, they're run the foundry. If they run with Microchip, they run with Microchip and with some small exceptions of what to go back and forth.

I wanted to move on to some of the discussion on the end markets, And you discussed, obviously, a lot of this in the earnings call. But as you go through the year, what are your expectations? I mean obviously, auto is in all the news. Auto is very strong right now. We've seen a lot of these work-from-home markets like PC and consumer that have been strong. There's some -- perhaps some concern that maybe some of that will reverse once we can go back to offices and back outside. What are your general expectations for the year? I know it's a very open-ended question because you're in a lot of different markets.

So first of all, I think we are -- have a very nicely diversified end market exposure through both our organic efforts and our acquisitions. And I think it served us very well in 2020 when there was weaknesses in some of the markets, automotive, industrial, consumer. We had strengths in other markets that helped offset some of that. So it kind of cushioned how that whole transition happened. And our product portfolio is also well diversified to play in multiple end markets so that we're not dependent on a single end market for growth or if they have a decline. And then finally, we're selling a complete solution. So we have total system solutions that bring to bear all of what Microchip does. So when I look at end markets, clearly, as we speak, automotive is in the news, but it's not only automotive that's strong. It's really all end markets, probably with the one exception of commercial aviation, it has a ways to go before it comes back. But most of the markets are doing well. Some are doing extremely well because they came out of a very deep hole, as automotive did. Others are sensing second wind, and so there's parts of work-from-home, as it is extended and people are, in fact, making certain investments that maybe they weren't thinking about a quarter ago. We'll see how that plays out. It's really hard to tell what the COVID effects are going to be as we go through 2022, in part because the way at which vaccinations take place and at which people feel comfortable going back into the office is not knowable at this point in time. Kind of we know somewhere in the second half, some form of normality will exist. And even then, I think there are going to be many hybrid work models where not everybody is going to come back in the office all 5 days of the week. And there may be some mixture and how that affects their purchase decisions and do they need something that they use when they're at home, which is different from what they're going to use when they're at work, et cetera. So a lot of moving parts here. We're optimistic, based on how the backlog is shaping up for all these different end markets, and some of that reflects the kind of tailwinds of COVID continuing to play out in 2022.

If I can dig into auto a little bit because, as I said, it's been on the news, been on everyone's mind right now. I mean one of the things you guys have seen -- GM has talked about a desire to build strategic inventory going forward because the vehicle production is being impacted by lack of components. Frankly, that scares me a bit when the customers say that. What's -- what are they telling you right now? Is it really just kind of focus on -- get the short term? And is -- are some of the -- are your customers starting to revisit their thinking towards some of these supply chains?

I -- very definitely. And in automotive and specific, I think the lessons of the last 3, 4, 5 months have been quite painful, as many of them have described in terms of what it has done to their annual planning. And in a more steady-state environment, perhaps some of the planning that had taken place for just-in-time lean inventory, all that worked out just fine. But when you have this dramatic shock down and then a dramatic recovery coming out, right, all the systems kind of broke apart where they're at. So as we have socialized the PSP program with many customers over the last month, certainly, car OEMs, so it's our customers' customer, are extremely interested and making sure that the Tier 1s are all putting in place PSP backlog for their requirements, and then working with the Tier 1s on how they financially support what needs to be done, et cetera, and that's between them on it. But I can't think of a single OEM, an auto manufacturer, who has not seen the benefit of the PSP program to give them that stability and planning that they had wanted and transmitted that to their Tier 1, who are our customers, on signing a quarter.

Right. Okay. I'll pivot over to margins a bit. Maybe Eric, I'll leave this for you. And just talk about the margin levers in the business. And it wasn't long ago that you were actually taking [ underutilization ] that's something that's going away. So maybe talk about the margin levers for the next couple of quarters, as presumably, this recovery continues?

Sure. So as you know, we updated our long-term operating model last quarter in early December and raised our gross margin target from 63% to 65%. And the first thing on that list of about 5 or 6 factors that we outlined for investors was an elimination of underutilization charges, and that happened to a large extent in the December quarter and really expect that to really completely be out of the system come the March quarter. And that's given us some really good gross margin trajectory. We were at 63% for the December quarter. We're guiding at the midpoint of the current quarter to 63.5%. So making good progress on those efforts. And the other things that we've outlined are what I would call kind of more blocking and tackling, right? We've got existing clean room space to expand into within wafer fab and assembly and test. We're insourcing activities on both the assembly and test side with some longer-term goals to get to over 60% of the assembly in-house and over 70% of the final test in-house. You saw a pretty material change in the amount of internal assembly that we did last quarter. I think in one quarter, it went from 47% to 55%. So that's a significant jump. And final test right now is about 57%, so still room to continue to make those investments as we add CapEx over time and take a little bit more control of our own destiny in that area. And then we've got some smaller, less efficient factories that we will move those operations into our larger, more efficient factories over time, long term programs, many of these factories came to us in the Microsemi acquisition and we're working on that. And the last thing I really highlighted in our margin profile going forward is the stability we expect in terms of a disciplined product pricing strategy. And clearly, there's been a lot of increase in cost in the supply chain, and we did make some pricing increases that we've talked about in passing those costs along to our customers. But pricing discipline is something that we've been very focused on over the last few years, and we expect that to continue. So all those things combined, we have confidence in being able to get to that 65% target. Investors should have the expectation that from this point forward, those are kind of small gradual changes as we go along, not a step function increase in any particular quarter. But it's set up very nicely for us.

Right. And if you could follow-on with some of the price increases that you've done, because I guess what you -- 2 things you talked about in the call was, one was raising prices on some individual products. But also as you allocate fixed scarce capacity to customers, I would expect you would naturally be allocating some of that to some of the -- they're not on the PSP program, the customers toward the front of the queue were probably some of the higher-margin customers.

So pricing was a broad adjustment we made in response to input cost increases that we had. Largely that's done. There may be small remnants of that, that we will do as we go along. And so that part of the program is -- at least the major part of the program is complete. Now when we allocate capacity, it isn't always as simple as just go find the highest margin business and allocate the capacity there because you have to think on the longer-term business that you're trying to do with various customers and through the cycles. Just as when cycles get weak, we don't go in and say, "Let's go change the pricing down so that we can get more share and where we're at." So we're thoughtful about it. Clearly, low-margin business gets more attention and scrutiny when you have these environments where this constraints. And in some cases, the low-margin business may have a higher price increase associated with it to be able to give it a better chance, to be able to compete in the overall capacity thing that we have on it. But the PSP program at this point is driven by those customers who are able to provide us 12 months backlog on a noncancelable, nonreschedulable basis with months 7 through 12, where they get the priority for capacity that's available out there.

Understood. Just for the audience, we've got just under 10 minutes to go, so if there's any questions, we'll put a last call for questions. But I wanted to kind of pivot over to a broader strategy. And you've recently said that the M&A that you've done over the number of years is likely to subside as consolidation has kind of run its course a bit. And the focus is on now generating and returning free cash flow. So maybe you could talk about that a bit on how that strategy has changed?

Sure. So after we completed the Microsemi acquisition in mid-2019, we knew we would not have the financial or the organizational bandwidth to do another acquisition for several years as we worked on extracting the value from that acquisition. In the meanwhile, we had also largely reached the twin goals that we had when we embarked on acquisitions as a strategy almost 10 years ago. And those twin goals, one were around scale. We had started close to $1 billion revenue. We were at 5.5-ish by the time we've completed that process, and the portfolio had substantially expanded from what we had back in 2010 to where we are today. And so those were the 2 goals that drove us from an acquisition standpoint over many, many years in what we have done. Now as other acquisitions have happened, even in the period when we were not in the market for it, we do run them through our internal assessments and putting ourselves in the shoes of if we were acquired, would we have done it. And so over the last 2 years, we have observed that valuations have climbed to levels that no longer meet our internal filters. So even if we could have done it, we would not have done it. And of course the number of targets is also starting to reduce. And so as targets reduce and valuations are outside of our internal hurdles, we came to that conclusion and said, "This is not the path that we're going to choose to be able to grow." And we're going to grow with an organic growth path, and we're going to figure out and how we get the markets right, how we get the selling as a total system solutions, right, and make that the foundations on which we're able to grow. And so even when financial bandwidth returns, let's say, in 2022, 2023, we're not expecting that M&A is going to be the way in which we drive our growth.

Right. So -- and I guess the question is, what do you do with the cash at that point? I know you've stated an intention to bring the debt down as your principal driver right now. But I guess -- so what becomes the trigger to start bringing more cash flow back to the investors? Is it...

Yes, good question.

A particular net debt-to-EBITDA ratio? Is it getting back to investment grade on your debt?

Good question. So as you've seen, even in the face of 2 really difficult years, calendar year '19, calendar year '20, right? Fundamental cash generation characteristics of the business have been phenomenal. And despite all the challenges that were out there, we've been able to bring the debt down substantially over that period of time. And as we made progress to use the cash generated to bring the leverage down, the Board has also been looking at what -- if you look ahead and where we are and what we should do and has been discussing options for longer-term capital allocation, especially once we get to investment grade, which is kind of the first major step we want to get to in bringing that debt down. And hopefully, that is a 2022 result that we're able to get to. But with acquisitions being off the table, the Board has decided to pivot towards returning more cash to the shareholders. And we don't have the metrics that we're looking for on like what metrics do we do, et cetera. But we -- we're not waiting to get to investment grade. The dividend increase announced last month is the first step in a glide path of increasing dividends as we head towards investment grade. And the Board continues to look at not only that, but also look at other options, including share buybacks at an appropriate point in time. And when there is something more material to discuss on Board's philosophy that is different, then we'll come back and share it publicly. Right now, the Board would like to do more in returning capital to shareholders.

Sure. And understanding, and it's an unfair question is that you're still in deliberations, but would a -- eventually, once the debt is at the right level once you're investment grade, a TI style program that says, hey, listen, when we get to that level, we can be in a position to return 100% of free cash flow to investors. Is that one of the options on the table?

I think the Board is looking at all options, right? So -- and looking at different scenarios in which we might be and what we should have. And what certainty to go with and what optionality to have with respect to capital return on it. And I think at this point, there's nothing that I can give you as a -- this is exactly where the Board is at. It is -- the Board is actually thinking it through, is making steps that they think are heading in the direction they would like to go to long term, and the dividends and all of that is a quarter-by-quarter decision that we make. We can't know what are we going to do in the next tranche of dividend and how is it different from what we've done in the last quarter. And you saw the first big change in what we announced in February.

And it was. And it was one of the most significant dividend increases you've had in time years ago that Microchip had a much higher dividend yield before you embarked on the M&A strategy. So -- right.

Well, so there was M&A, there's -- obviously, the reduction of debt, there's dividends and the share buybacks. And so it's how you balance across those capital allocation decisions that we had to navigate the last several years.

Sure.

I think I'll probably just make the point that even when we get to investment grade, we're still going to have a relatively significant amount -- dollar amount of debt on the balance sheet, and I think we'll continue to chip away at that. So it's going to be a balance. I don't think there's -- we just wake up one morning and the Board says, "Hey, today, it's 100% return of free cash flow to shareholders." I think you'll see gradual steps that we'll take over the course of time.

Fair enough. And what about reinvestment into the business? And the business is more diversified now than it has been since you've embarked on the M&A. With that, there's probably some decisions you can make within the business on more investment in this area, less investment in some of the other area. Is the intention to kind of keep the OpEx as a percentage of revenue kind of stable at this level? And pick and choose within that? Do you feel like if you spent a little more on the OpEx side, you can grow the business a little faster? What's the view there?

Well, I'll take a quick cut, and I'll let Eric answer this as well. We bumped up our OpEx target marginally from 22.5% to 23%. And I think that is reflective of -- it takes investments to be able to help create a business and sustain a business that has the growth and profitability characteristics of what we have. And within that, the allocation of where it goes is an ongoing business decision that the executive team makes. Go ahead, Eric.

Yes. And there's lots of areas where we need to continue to invest. I mean we've been in a period over the last 2 years where the top line growth has been hard to find, right, with the China trade war, followed by the pandemic. And there is some pent-up demand within our 25 or so business units, within our technical teams that support customers, and we need to keep feeding the system to make sure that we're going to be producing these outstanding operating margins of 42% target here, not just in 3, 4, 5 years, but for the rest of Microchip's existence. So it -- we don't want to underinvest in the business. Will these metrics change over time? It's possible, but we think 23% is about the right level for us to be targeting in the short term.

All right. Well, thank you, gentlemen. I see we're just about out of time. So I think we're going to have to wrap it here. But we appreciate you joining us, as always, and we'll leave it here.

Thank you, Chris.

All right. Thanks, Chris.

All right, thanks, everyone. Thanks for joining.