Good morning. Thank you so much for coming. I'm Stacy Rasgon. I'm Bernstein's Senior Analyst. I cover the U.S. semiconductor and semi-capital equipment space. I can't express what an honor it is to have our guest here today, Ganesh Moorthy, the President and CEO of Microchip, who has never actually been to our conference before. I'm thrilled to have him here. Our chat today will last about 50 minutes. Ganesh will give a few opening remarks, start us off, we'll move into Q&A. I've got tons of questions. If you have questions, there is a Pigeonhole forum. I think there is a QR code on the back of your program you can scan to get into the portal.
You can submit them. I will get them over here on this iPad, and we'll leave time at the end, for those questions. With that, I think we should just get started. Thank you so much for being here today.
Great. Thank you, Stacy. Good morning. Let me start with the obligatory thing I have to do. During the course of this presentation, 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 results and events may differ materially. We refer you to our most recent filings with the SEC that identify important risk factors that may impact Microchip's business and results of operation. I'll start with an update of the quarter, and then we'll go into my actual presentations themselves. The June quarter business is steady. We are on track for the guidance we provided just under a month ago, and that's for this quarter in the June quarter.
As we indicated at the conference call, we still expect that it's highly unlikely we'll have a down quarter, sequentially down quarter, in the September quarter as well. With that, let me go into the Microchip story and a little bit of, you know, what makes us who we are and where we're headed. To begin with, I thought it was important to understand, you know, why do we wake up? Why do 22,500 employees wake up every morning? It's to fulfill our purpose, and our purpose is to empower innovation that enhances the human experience by providing these building blocks that are smart, connected and secure. We want to change the world and change it to be better, utilizing our capabilities. What does this mean?
Just to give you an idea, across a spectrum of different end markets and applications, you can see just about any end market of consequence has something that we're involved in that takes advantage of the many capabilities we bring to them. We'll speak about some of these more specifically in the coming slides itself. This is really what empowering innovation is about. About 18 months ago, we had an Investor Day, and at the time, we introduced what is Microchip 3.0, and this is the next phase of success, and 1.0 and 2.0 set a foundation. I want to take you through some of the key elements and show you some of the facts and figures that go with that. Let's start first with growth and in particular, organic growth, and our strategy there.
Our goal had been that we're able to grow 10%-15% on a compounded annual growth rate on a fiscal year 2021 baseline. That's a longer term baseline, so this is over 30 years of time, you can see how the company has grown. It's over 16% over this timeframe, and in the last 2 years, it's been on a cumulative, both years combined, 2-year basis. It's over 50% of growth that we've had as we've had a tremendous amount of opportunities that we have brought to revenue in the last 2 years. How we do it is what's called total system solutions, and what that means is, pictorially, we have many, many parts of the solution that a customer requires. In the dark blue is all the brains of a system.
Some of the other boxes. In the brains of the system are microcontrollers, processors, FPGAs, et cetera. In some of the other boxes, you'll find the power that's required in every system, the clock or the heartbeat required in every system, the signal chain of how do you go from an analog world to a digital processing, the networking that you're going to need, the security you're going to need. What do you do with all of that when you process it, to be able to do something on the outside, change a the output of your air conditioner or change something else that has to be the output of what you process as well. These activities are driven by six major megatrends.
On one part, we have total system solutions, maximize everything we do in a given application. The other is around focusing on six major market megatrends that we believe we have solutions for and have long-term growth associated with them. The first of them is 5G infrastructure. We're into the third or fourth year of a 10-year journey, so a long way ahead of us. The second is IoT and edge computing. This is one of the elements that also has a lot of AI and ML built into it as well. The third one is data center, and as you will see in a minute, it's the largest portion of our megatrends at this point and also has a significant AI/ ML component that's associated with it.
The fourth one is e-mobility, so electric vehicles of all kinds of shapes and sizes, and the EV infrastructure for charging as well. The fifth one is around sustainability, so anything that is required to be able to reduce energy consumption, reduce water usage, reduce waste, renewable energy, all of those different things that fit under sustainability and what we enable with our solutions for. Finally, is advanced driver assist, and really those advanced driver assist systems are becoming like servers on wheels, and again, has a pretty high AI/ ML component that goes into how those systems are built and what we do to enable them. We had talked about this, but we really have not broken it out.
First time we're breaking this out to show you, where were we two years ago on fiscal year 2021? The pie chart you see was about 34% of our revenue was coming from the six megatrends, 66% was everything else, broad-based of our product lines. Over the course of the next two years, fiscal 2023 was just over for us, 45% of our revenue now comes from these megatrends, and you can see the data center is starting to become one of the larger areas of growth for us. IoT is the second largest, and all the other ones may be smaller in slices, but starting to grow significantly from two years ago to now, and remain the focus of where we think incremental growth for Microchip comes from.
These megatrends are growing at about 2x the rate of the company overall. If you look at the non-megatrends over the course of two years, or the company as a whole, actually, over the two years, we grew at about 55% over two years. The megatrends grew at about 108%-109% in that same time frame. It is pulling up the averages for how we're gonna grow, and it has for many, many more years, legs in front of it in terms of how it's gonna help us from a growth standpoint. I thought I'd pick the data center and double-click for a minute, because there's a lot of talk about what's happening there, et cetera. Here's all the places we play in a data center.
All the way from the storage, to clocks, to security, to the user interface, to power, making it more power efficient, to all the different elements of this, which are more associated with AI/ML, which is giving, like everyone, it's giving us a tailwind as well, where it's into our switches, it's into our data center interconnect, it's into our clocks, it's into our security. All of these are incrementally more critical in a system that has AI/ML performance requirements. Next, I wanna talk about our business model and how that is improving over time, and how the diversified end markets we have are giving us a degree of durability for our business. Here's our revenue by end market.
This is measuring year-over-year, fiscal 23, we just finished, to fiscal 22, and it shows a continuing move towards a more durable end market. Over the course of the last year, our data center business, as a percentage of the total, grew by 1 percentage point, from 18 to 19. Our industrial business, which is the largest piece of our business, grew from 40%- 41%, and the 2 percentage points came away from consumer appliances, which went from about 14 down to 12. Continuing to curate further into the most durable parts of the end markets we have. I thought I'd give you two examples of what does this mean? We are in aerospace and defense. It's part of our industrial business, about 8%, 9% of the company's revenue.
Here's an aircraft represented by, you know, many, many different applications. You can see where are we present in the different applications within an aircraft. You can see it's pretty broad. In every one of these applications, we have multiple products that go into it. On an average commercial aircraft, you're gonna find over $100,000 of content from Microchip. That's one example of how we have complete solutions going into it. I picked an automotive example. Here is an Asian car. It's from Korea. It's Hyundai Kia Motors. It's their electric car, which is starting to come into the U.S., I think it's branded as IONIQ.
Here again, you see, 88 different components from us, across the spectrum of computing and analog and networking and security and wireless, it's built into it. A pretty broad area. You can take any car from any part of the world, and you're gonna find something representative of this, in our content, in vehicles itself. Now let's look at the durability. Here's semiconductor industry cycles, and people from time to time, talk about, "Hey, this is not cyclical industry." Well, it is, and you can see how it is, and this is the industry, not Microchip, and how it's performed, and you can see where we are in the cycle, as we have seen here. Looking at our performance and the durability of that performance.
First, we look at it in terms of gross margin. Gross margin has continued to have higher highs and higher lows. Over this time frame, 15+ years, we actually had many acquisitions that were dilutive to gross margin, so that's what you'll see in the history. As we have stopped acquisitions in the last four or five years of time, you can see the gross margin starting to go up into the right directionally as well. We're right at the top end of where our long-term targets had been, and doing extremely well as we go through this process. A lot of that is happening because our mix is getting better, our absorption is getting better, our factories are running well in terms of how that gross margin is accruing.
Looking at the bottom line, our operating margin, same thing, higher highs, higher lows, getting up into well past of our long-term targets as well. What we have done is, under various scenarios, looked at, well, what might a bottom operating margin look like under some specific scenarios? In a worst-case scenario, what we find, is about a 40% bottom, and, we're well above that. You can see historically, we really have never declined at that level throughout the different cycles as well. The business performs extremely well through the different cycles, both on gross and operating margin, because of the end market exposure we have and the products and technologies we bring.
Lastly, what really matters is, well, how much cash can we generate, and how much free cash flow are we able to generate? You can see this is a very, very consistent story throughout this whole time frame and continuing into more recent years as well. Next, we're looking at our capital investments and our investments in inventory, both of which are important to us. We moderate our capital investments between 3%-6%. Last fiscal year, we were into high investment mode. We were close to 6%. This year, we'll come down probably closer to the 4% range of where we're gonna be. Inventory is a tad high, but these are all highly durable products. They don't really go obsolete.
We also measure ourself on what kind of return on invested capital are we able to generate? Here you can see we're in the 76th percentile of all S&P 500 companies. We put all 500 companies and line up where we are. We're in the 76th percentile of performance for return on invested capital. We look at the capital return, right? What are we doing in terms of free cash flow generation, and then what are we doing with that free cash flow when it becomes available to us itself? Here's our EBIT, adjusted EBITDA, and, you know, we're over 50% in EBITDA performance. Well over $4 billion of EBITDA that we generated in the last fiscal year. This is a money-generating machine.
Our revenue is $8.4 billion, half of it is dropping into EBITDA. When you look at free cash flow, specifically when you look at free cash flow per year, per share, again, you can see the same characteristics of significant free cash flow that we're generating and what it is doing on a per-share basis. We need to do the same analysis of take the S&P 500 and look at free cash flow as a percentage of our revenue, we're in the 88th percentile of all companies in terms of how much free cash flow we generate as a function of what revenue we have.
Over the last five years, we had a fair amount of leverage we took on in the last acquisition that we did back in 2018. The leverage at its peak was close to five, and you can see what we've done through all that free cash flow to be able to systematically bring that leverage down. Under 1.5 at this point, and they're continuing to go down quarter after quarter. As the leverage has come down, we have started to accelerate our free cash flow that we return to shareholders. You can see the last two years, that has an absolute incline up from where we were. We've added a component of programmatic share buybacks as well. We are well on track.
This quarter, we're returning 67.5% of last quarter's free cash flow. We're growing at about 500 basis points every quarter, by the March quarter of calendar year 2025, we expect 100% of free cash flow to be returned to our shareholders. If you then take that cash return as a percentage of revenue, again, across the S&P 500, you'll find that we're in the 77th percentile of all companies that are there. That will go up quite a bit more as we get towards 100% free cash flow return in the next 2 years. You take it all in, we find, you know, we have a compelling valuation.
You look at the growth, 5-year growth rate, well past the 10%-15% range, 16.4% over 30 years. The trailing 12-month non-GAAP gross margin at 67.8%, non-GAAP operating margin at 46.9%, trailing 12 months of adjusted EBITDA. That's over $4 billion, $4.3 billion, to be precise, over 50% of our revenue in that EBITDA. Free cash flow in the 88th percentile, as I showed you, of all S&P 500 companies, and on track to get to a 100% free cash flow return to our shareholders by the March quarter of calendar year 2025. It's premium performance in terms of what we're doing with the business, products, technology, innovation, financials, growth, et cetera.
You look at the valuation, I think we believe there's a compelling valuation. You can see where we are, on and this is as of March, sorry, May 19th, is the date that this was measured, so it may have changed a little bit. You can see what the comparison is. If you want to compare it to S&P 500, it's at a discount. If you want to compare it to SOXX, other semiconductors, it's at a discount, if you want to compare it against a couple of peer companies in the analog mixed signal space, TI, ADI. We think there's a compelling opportunity to acquire these shares at a discount during a time of, you know, premium performance across the board for where we're at.
The winning formula for us is to grow the revenue, grow it profitably, expanding our gross and operating margins, generate tons of cash in the process of doing that, and return the excess cash, free cash flow, to our shareholders. Also how we do it is important to us, and important to us is the culture of how we build the company for the long term and the sustainability that we want to build into that company and our structure, and what we enable our customers for as well. Our bottom line is to deliver differentiated results with what we do. Thank you.
Thank you so much. Very interesting. I feel like I've seen some of those charts somewhere.
A few new ones.
Yeah.
A few old ones.
Yeah. I want to start out, you talked about, like, Microchip 3.0, and you put that out 18 months ago. I was wondering if you'd talk a little bit about the journey, like, through Microchip 1.0 and Microchip 2.0, because I think you go back a ways, your investors kind of looked at you guys as a, just a seller of 8-bit microcontrollers, and that's clearly not the case anymore. Maybe talk a little bit about it. You did talk about recent, you know, five years ago was the most recent M&A, but you've done a lot of M&A over the years. How did the portfolio build out?
What were the things you were focusing on, as you sort of built up, like, the business as it exists today, as you moved to this new model?
Yeah, great question. Microchip 1.0 was, you know, circa started in 1993 or so, and we ran about 15 years. That was basically all about building a differentiated 8-bit microcontroller platform, as you talked about it, and then starting our journey on 16-bit and 32-bit on analog and a number of things, et cetera. It got us to about $1 billion from starting from almost nothing, became a public company, to about $1 billion. At that point in time, you know, we had continued to improve the business, but we also knew we lacked scale and we lacked breadth of portfolio, which is when we began our journey of acquisitions back about 2009 or so.
In Microchip 2.0, we took all the strength of 1.0 and said, "Let us build out the portfolio, and let's build scale." In the course of about eight, nine years of time, we did about 17 acquisitions, 17-20 acquisitions, about seven or eight of them that were public and large, and grew from about a billion-dollar revenue to about a 6 billion dollar revenue or so. We built up that scale, and in every single instance, we bought assets that had lower gross margins, lower operating margins, and improved them to, you know, our overall gross and operating margins pre-acquisition.
Tremendous amount of not only revenue growth, but profitability growth as well, and cash flow growth that came with it. It also expanded out the portfolio, so we now filled out a substantial portion of all the different things we didn't have, many different types of controllers and processors and FPGAs, the complete analog signal chains, the complete networking signal chain, and what we did. That's when we got to the last major acquisition we did, was 2018.
The Microsemi?
Microsemi. By that time, we were about a $5.5 billion dollar company, and we said, "You know, we've filled out the portfolio to a substantial amount," and we had gotten the scale. In Microchip 3.0, we said, "Well, let's take and harness all the assets that we have acquired and build out into organic growth, pay down the debt, and move into a mode of returning capital to shareholders." That's where we're at, and that's most of the journey I talked about in the last few minutes. That's where we are at in the last 30 years or so of us.
Mm-hmm. Were those deals primarily, like, product-driven, or were they end-market-driven? Because I don't know that you were talking about these sort of mega trends as you were doing those deals. Did you need those deals to get into these mega trends, or was it just building up the portfolio?
It's both, right? The products enable you to go into certain end markets. For example, when we did Microsemi, there were three key end markets that we had a minor presence in that Microsemi accelerated to. One was in data center. It was a big end market that they brought on for us.
Mm.
We were there, but we were small. Two was in, communications infrastructure, and three was in aerospace and defense.
Mm-hmm.
Today, we're the largest aerospace and defense semiconductor company, by far, of anybody else combined. Other parts of the acquisitions we did gave us specific portfolios that were around networking as an application, for example, so the whole series of networking in a car, in industrial and various other places. We had a combination of some were just products that built out the portfolio, others that were specific applications, and still others that opened up entire end markets that we didn't have.
Mm-hmm.
They all, in some way, feed into the mega trends as you then pick and choose the combination.
Mm-hmm
That's needed in a mega trend.
Yep. It's funny, if I look at that chart of the pie charts with the mega trends, it's pretty much data center and IoT, and then a bunch of little stuff at this point, right? You talked about the mega trends growing 2x. Is it primarily data center and IoT that are growing 2x, or is everything driving it? Because the other ones are still... I guess small is good, I mean, because small can become big right? We all work on that.
If you study the data, and you can just multiply it out by the revenue, you'll find the smaller ones are actually growing even faster-
Okay
Off of a smaller base.
Yeah.
They're all adding, and they're all adding incrementally to it.
Mm.
They're all important for us. Data center obviously started the largest, remains the largest. Is growing tremendously, as well. All the pieces are growing, and they're all giving us tailwind, which is in excess of corporate average growth.
Got it. Let's talk about data center for a minute. Clearly, your business is growing. There are parts of the data center business that we're all discovering now are growing tremendously. Other parts, maybe not so much. Like, where are you, I guess, exposed within data center? Maybe you could just tell, what is going on in data center right now?
I think it is a tale of multiple stories. Depending on where you're exposed, you have more or less growth that's coming from it. As to step back, you know, data center CapEx is, flattish at this point in time as people pull back.
Do you think flattish is would be flattish?
I, you know, I don't know, right? It's in the realm of where customers and their customers often are building into. Let's assume for a minute it's flatter.
Uh-huh.
Within that, the dollars are being spent differently.
Mm.
As you have seen from all the news in the last few days, and a lot of it is going into AI/ML. We've known that for a while. We have exposure into that space.
Mm.
We have been designed into those platforms, and you can see multiple products that we have that either go, you know, one or two per board, or in many cases, one or two per GPU.
Mm-hmm
That's there. That's what we have focused on, is how do we enter the parts of data center? You know, our primary role was to be the memory interface, or the interface, the networking interface between the memory, and the CPUs and GPUs.
Okay.
We've also, with our PCIe switches, been the interface between GPUs when you have an AI/ML type of system. We've brought on interconnect inside the data center. A lot of these data center products require root of trust in terms of how they start up.
Mm
Where they go, and we're a big part of that. A lot of them require, you know, the standard solutions we have, microcontrollers and power and that type of stuff, that goes into it. Clocks are an important part of how they maintain the performance, and we have very, very precise clocks that are used in these things. Then there are new data, memory interfaces, CXL is one example out of that. We have the right exposure into data center to give us substantial growth, and the business grew much, much faster than Microchip over the last 12 months of time.
Got it, got it. With AI, is that what you call sustainability on the, on the chart, or is AI a piece of data center, or what?
AI goes, it is a very big piece of data center.
Okay.
AI is also a critical part of our IoT solutions.
Okay
Particularly in edge computing. AI is a big part of automotive and the advanced driver assist. Sustainability is more about energy efficiency, renewable energy, reduction in greenhouse gases, in energy consumption, water usage, waste generation, et cetera. That's all, and we do that internally for our own business.
Mm-hmm.
We also enable that in a large way for many, many customers who are in that space.
I'm kind of surprised, especially in this environment, that you don't carve out AI as its own little slice of that pie.
You know, if you go back and look, back in fiscal 2021, we did. What we found was there was so much confusion between because AI is there in the edge, and AI is there in all, in many data center applications.
Right.
Is made there in the. We, you know, we had a duality of, you know, do you put it in this or do you put it in that? We said, "All right, the AI fits into the mega trend that it belongs into. Our AI is represented in those three areas, and it's probably in even others as well.
Mm-hmm
Those are the three that they largely reside in, with data center being the biggest piece of where our AI.
Is it the same kind of play across all of them, as is it power, and it's microcontrollers, and are there specific things, like within end markets, that you're doing around that?
Within the data center market itself, and it actually plays out into automotive as well, into ADAS, these PCIe switches are a very critical part of how GPUs are able to do their parallel processing and be able to deliver the kinds of things that you've seen with AI. To run those GPUs, you need, you know, very precise clocks, you need root of trust. Y ou need data center interconnects and all that. What you do, what we do in IoT and the edge computing is perhaps a little bit different. There, what we want to be able to do is enable, vision systems. Enable processing power at the edge without having the latency of going all the way back into the cloud.
Got it. Do you have a view, I guess, excuse me, on AI and compute growth in the edge versus in the cloud or in the data center?
Not a precise enough number to give you as much as... we're still...
I just qualitatively, I'm just thinking about, like, the, you know, people wondering if this gets centralized. Like, where is more of the growth? Is it at the edge?
I think if you look historically, there has been far more in the cloud. If you look forward. B ecause of the latency effects and the increasing use of AI/ ML, the edge is gonna have explosive growth ahead of it, off of a smaller base than where the cloud has been.
Mm-hmm. Got it. You know, maybe that's a good segue into IoT. Like, I kind of hate that term because I feel like IoT can mean whatever you want it to. What do you guys mean when you say IoT?
It's a, it's a great question. We have typically, described an IoT application that has three pieces to it.
Okay.
Uh, we-
Give your definition. Thank you.
Yes.
Okay.
We wanna make sure that it has some form of smart capability that is built into it. It needs some form of connection.
Mm.
Can be wired or wireless, so it can be industrial Ethernet, can be, you know, Wi-Fi, can be Bluetooth, any number of different ways that it connects. It needs to have security of some form so that connectivity is secure.
You guys support all of those different kinds of...
We do all of those.
Okay.
We have a very broad spectrum of both, you know, standards-based and proprietary, wired and wireless connectivity. Because industrial, as an example, which is what a lot of IoT is, it's a very, very fragmented, very broad-based type of end market for where we have them. Those are the requirements we have. We do have products that are just smart without the other elements. We don't count them as IoT. We have products that are connected, but may or may not be what we're doing from a smartness standpoint.
Mm-hmm.
All 3 elements are we consider to be important to classify it as IoT.
Got it. Got it. Let's talk about EVs a little bit. Again, the slice is small, but it is growing. You gave an example that had, like, a ton of stuff on it, so I'm getting the impression that your EV play is not like I cover ADI, and it's like it's battery management systems, and that's great.
Yeah
It's BMS. You guys seem to have a much wider swath of your attack on EVs. Can you talk a little bit about what the opportunity is there?
Sure
Where you see, like, the biggest bang for your buck?
Firstly, when you think of an EV, there are a great number of applications, you know, in the, from a body control, standpoint, from a convenience standpoint, that are no different from a, you know, an ICE, vehicle.
Mm-hmm.
You know, if you take an EV, we're in the touch screen on both of them. We're in the garage door controls on both of them. We're in the seat controllers. We're in many different aspects of ADAS in both of them. A big play, irrespective of ICE versus EVs. You look at EV specifically, and you now get into anything to do with power. You have power conversion, you have DC-to-DC, you have power management, you have the onboard chargers. You begin to look at motor control. What are you doing from a, you know, motor control that drives the car itself, from any of the compressors that are required? You have a unique requirement in EV, which is it has to generate sound. The National Transportation Security-
Oh
... has said, "An EV must generate audible sound at low speeds so that people walking around with their phones don't get run over because they're not paying attention to what's going on." We have standard microcontrollers. It's the function is called AVAS, A-V-A-S. Some vehicle sound, something or the other, is what it says. That's an example. We're gonna be ubiquitously present in EVs.
Mm.
Now, with our silicon carbide products, we have started to enter into more aspects of it. We want to find, you know, profitable niches within it.
Mm
If you can see where our gross margins are, we want things, find things that can eventually get us there. We're in a new category of product within EVs.
That must be nascent right now, just given where your EV rev-.
It is small and growing fast.
Uh-huh
But also growing into new areas.
Mm.
We're right in the middle of many, many parts of EVs that are starting to use what are called eFuses, replacing mechanical fuses or the other, old-style fuses, as well. That's using our silicon carbide. Of course, it'll go into some of the traction control and other parts, of where it's at. Yes, we have a highly robust, silicon carbide solution, one that doesn't require redundant products being built into the system, has a cost of ownership, everything on it. We will do well with our silicon carbide products as well.
Is it just auto, by the way, for silicon carbide?
Oh, no. In fact, the bigger business at this point in time is industrial.
Mm-hmm.
Aerospace and defense, as an example. Just basic, you know, you build a piece of semiconductor equipment, and it needs these high-power devices.
Mm-hmm.
We got silicon carbide in many, many places inside semiconductor equipment, as an example, into train systems, into, other industrial pieces of equipment. No, we see as big a play in industrial, as there is in automotive, for silicon carbide.
Are your total auto revenues, by the way, they're bigger than that EV slice system?
Correct.
Correct?
I think if you break it out, total auto revenue is about 17% of our revenue.
Yeah.
If you take the EV piece, it's about 2% or 3%.
Okay.
It's about 1/5, 1/6 of the total auto revenue.
Auto's been growing, on an absolute basis, been growing a ton. This is one thing I've been wrestling with for the industry, which is the auto AUTOSAR has not, right? It's quite a bit below, yet auto semis are just, not for you and for everybody, are growing a ton. Is it all just content or something else is going on in that space, at least for you guys?
Yeah. Oh, great question. AUTOSAR, depending on which numbers you're quoting, typically refer to the U.S. part of the market, while auto is a worldwide market. L argest market today is in China.
Yeah
Where it's at. I think some of the numbers get distorted by the fact that the U.S. is only one small part.
Even global, we were doing, I don't know, $95 million pre-COVID.
It's about 85.
75 or 80 or 85 now.
Yeah, in that range.
Yeah.
Yes, content is a second big piece as people build more capabilities.
Mm.
More functions for safety, for convenience, for entertainment, et cetera, that is there, and that is growing. Third is, in that time frame of shortages, et cetera, automakers, in their self-interest, bias their models to have the richest models, the best margins for them in terms of where they're at. The cars that got built were the ones that had the most number of features.
Mm-hmm.
If you and I, as a consumer, wanted to go buy a car, we probably couldn't find one.
Yeah
That was, defeatured because the only ones that were available were fully featured. All those are there. There is an absolute content play where automakers are using semiconductors as a way in which they differentiate their products, differentiate the features that they're offering, making their vehicles more attractive.
You know, you mentioned that your own inventories are a little bit high right now. It doesn't matter because it doesn't go obsolete. Most of the stuff the auto vendors are buying, that's also true. Do you think they're more inclined to hold more inventory now than they were pre-COVID?
Yes.
People talk about, like, the whole, like, just-in-time to just-in-case kind of move.
Yeah. Yeah. Certainly, as I've had discussions with the auto CEOs, both tier ones and OEMs, that's a discussion we have. I think the thoughts are a little different depending on who you speak to and where they're at. Most of them, if you look at the last two years, right, they have the burnt stove, you know, syndrome. They put their hand on the stove, they got burned with what they did in 2020 to what had resulted in 2021 and 2022. They are more conservative from that standpoint.
Mm
They are looking to make sure that their supply chains are maintaining a level of inventory to deal with any, you know, can be a new pandemic, can be a flood, can be a fire, where it's at. Yes, I think they are more thoughtful in that.
Mm-hmm.
I've also been here long enough that I have seen how some of these companies, over the course of time, rapidly forget what the issues were. If you go back to 2011, you know, there was a tsunami and a flood, and there were issues, and the auto companies said, "Oh, my God, I got to do something different." They did for about 12 to 18 months. The half-life of your learning is about 9 to 12 months, and then in about 2 of those cycles, you've forgotten most of what you learned.
That's interesting. You don't think that some of these changes are permanent?
I would like to think so.
Okay.
I think what happens is, in a large organization, you get compartmentalized thinking.
Mm-hmm.
Rather than trying to solve for what optimizes at the whole, you begin to see optimization of the pieces. If you're in a purchasing role, you're gonna start to get measured based on: What was your productivity, what was your dollar spending, what was the thing? It slowly begins to revert back to where that's gonna be. If you go to the general management and the executive levels, clearly there's a very, very clear understanding. At the OEM level, we perhaps have even better understanding of what this issue is and what they need to do.
Mm.
Because if it's their $100,000 vehicles that didn't ship because of $15 of content, right? The amount of profits that they left behind, amount of revenue that they left behind. OEMs have actually gotten far more engaged post this last few years.
Mm
Directly engaged with semiconductor companies, on one hand, for the supply chain, but as much on the other hand, in terms of the innovation cycle and understanding what's possible. They also see that's how they're gonna make their differentiated cars, is to get close to the semiconductor companies. That is one of the good things that have come out of the last two years of things. You know, we have OEM relationship, even if we don't sell directly to them, that have vastly enhanced in the process of the last 2 years.
Got it. Let's talk a little about your own innovation cycle. I wanna ask a little about the total system solution move. Can you talk a little bit, like, the kind of value that's adding to the customers, as well as, like, how much value additional does that allow you to capture? I know pricing in general for the industry has been strong. Is this one of the mechanisms you guys have been using to actually both create as well as capture more value?
Absolutely. When you look at these total system solution slides, often the takeaway, people quickly take away is, "Okay, it's all the silicon that you have that's available." That's only one small piece of what when we think about total system solution.
Mm.
In each solution is a substantial amount of software, and in many cases, the services that are required as well. Let me give you a services example. You buy an Ethernet product from us, we offer a service called LANcheck. LANcheck is a way, LAN is local area network, is a way for you to design your system, put it through our check, and it will tell you, is it gonna be qualified and pass for Ethernet functionality or not? We do that for many, many other things as well. Then we have reference designs, et cetera. By the time you put in the software, the reference devices, the services that we bring in, that's what makes the total system solution sticky.
That's what makes the overall bundle, lower cost for the customer, lower risk for the customer, and higher likelihood of growth, for them as well. That, in the long run, drives both growth and profitability for our customers, and in the process, growth and profitability for us. It is much more than a silicon play, by itself.
Got it. Are you able to price for that? I mean, is that how that works?
You know, we price on value.
Uh-huh.
In the way we go. I mean, you look at how, over time, we have performed, you know, between what we do to integrate value in the silicon, what we do to add software value, system value, our own technical support value in it, right? We want to make sure that, it's not priced high, it's not priced low, it's priced to have the value that the customer perceives.
Got it. One of the other value adds, I think during the pandemic years, that you guys instituted was this PSP program, and I think you were one of the first ones to do this. From what my understanding, and correct me if I'm, if I have this wrong, this was not guaranteed capacity or anything, it was you were promising to prioritize customers for capacity if they gave you, like, long-term visibility and non-cancellable orders. Do I have that right?
Yeah. Very good.
Yeah.
12 months is what it is.
12 months. Okay.
Yeah.
Got it. We had a number of other companies over that period start to move to things like non-cancellable. This is not exactly like a non-cancellable, or maybe it is. I don't know.
It's non-cancellable, non-rescheduleable.
Okay. It sounds like you are rescheduling some of them now. Talk about that, please.
Okay. The PSP program, we launched it in February of 2021. It was right at the start of this, you know, post-COVID cycle.
You were one of the first, I think.
We were the earliest.
Okay.
It was in response to customers asking us, right? Vast majority of our customers, as you saw in the end market slide, have a very, very high-value end products that they're building. In the midst of what was going on, they had not only with us, but they had, you know, all the products they were buying had issues and, and they wanted solutions for what could they get in terms of priority of the available capacity. We socialized this and fine-tuned it based on them, but we came up with PSP as a result of that customer feedback.
Mm-hmm.
It was as exactly as Stacy said, which is, you know, in exchange for 12 months on a continuous basis, so every month, you put another additional month of non-cancellable, non-rescheduleable backlog, we give you the top priority of the supply that we do have. It worked out fantastically. It worked out far better than we even thought. You know, we thought maybe 20%-30% of the customers, you know, volume would be there. Well in excess of 50% of our backlog ended up in PSP, and it gave us the confidence to go ahead and start investing in hiring for our factories, putting more capital equipment in place, buying the materials that we needed to have, et cetera. Those customers got absolutely top priority in supply and did extremely well.
To this day, even as conditions have begun to soften a little bit out there, the PSP backlog, as a percentage of our total backlog, remains almost exactly the same percentage it was 12 to 18 months ago, because people see value in what we're doing. Now, backlog overall has come down and PSP backlog has come down, which has to make sense because backlog was insanely strong for where we're at.
Still pretty high, right?
Still high, coming down, coming to a more normal level, you know, somewhere over the course of the next 12 months, which is where it needs to be so that people can be more responsive to what is happening in their business, et cetera.
Mm-hmm.
Yes, about six months ago, plus or minus, we began to get requests saying, "Hey, you know, wanna be able to adjust, cancel, push out, whatever." What we were clear on is, look, non-cancellable backlog, we want you to be very thoughtful when you put that on. We believe we have the best quality backlog from when people were putting them in place, because I know when, at Microchip, if somebody needs to place non-cancellable backlog on our suppliers, it takes tremendous scrutiny, often takes a CFO's, you know, approval...
Mm-hmm
... to get there. That's the way it is at our customers as well. The backlog was placed with what, at the time, at the best knowledge of what they wanted to be able to place, which had a high quality in it. Things have changed. What we have said is, "No, we're not gonna allow you to cancel backlog because we made investments on that, but we will reschedule, and we will help you out".
Mm-hmm.
At the end of the day, I think it helps us as well to kind of smooth out where this thing is going to land, whatever shape this thing takes and where we are at. We have made tremendous progress in helping customers out. Strangely, from time to time, a customer comes in, wants to cancel, wants to push out, and three, four months later, they're coming right back and say, "Oops, I don't wanna change any of it. Please pull it all back in." Because there is sometimes a tendency to say, "Well, let's just flush everything out." Then, as their business takes shape, they say, "Oh, my God, I canceled stuff I really wanted, or pushed out things...
Nobody knows anything, right?
No, exactly. Where it's at. I think some of that will take place in the coming months as well.
I should ask about that, 'cause, you know, you guys have a reputation of at least being willing to stick your necks out sometimes and tell us what you think, and not everybody does that. I'll be honest, I'm not sure if that's if the right strategy is to try to tell us what you think or to try to tell us nothing. I don't actually know. You guys have, in general, have felt more comfortable doing that. Is there any reason why? I don't know if your philosophy on this differs at all from Steve's, the prior CEO.
Mm
Why have you been willing to do that? I guess you, it seems like you're still willing to do that.
I think one, I'll make a small, subtle change.
Yeah.
I think in the past, many, many years ago, people thought we were trying to make industry calls based on calls for our business. We always called it for our business as we saw it. We weren't trying to make, and now people would make inferences about, Oh, if Microchip says this, then this is probably what it means. We were very clear many years ago saying, Look, we're not trying to call the industry, we're trying to call our business. We think it is important to have a level of transparency for our investor base to be able to know what are we seeing, what are we doing.
Sometimes what we do is, has, you know, people, you know, imagine things about it that is not what we want to think. You know, we write letters to customers often, and some people think, ''Oh, you're trying to manipulate things, whatever.'' No, we're trying to serve our customers. Just as it is for our different constituents, whether that is customers, whether that is investors, you know, we're trying to tell the story and provide the information. T o allow them to make the best informed judgment. For what they're trying to go do. That's just been the ethos of the company in terms of how we communicate.
Got it. I wanna move to the model a little bit, especially like the renewed focus on free cash flow and cash return. Clearly, like you've moved from an M&A, you've pivoted from an M&A strategy to a cash return strategy, and I get that. What I don't quite understand is you have the dividend, and you're sort of like slowly ramping up the buyback. Right now, I mean, you showed us, like, the value, you're at 11 times, and your leverage is already below your 1.5 target. Like, why are you taking so long to get to that 100% of again? Why not get there, like, tomorrow?
Yeah, we would have gone even faster, but something changed in the last 15 months that changed our own thinking as well. The largest change is, you know, what's happened with interest rates and what's happened to our debt. We thought it prudent to continue to pay down a little bit more of the debt. you know, our debt has gone from, you know, or the interest rate, as you know, have gone from 1 percentage point to, you know, close to 6 percentage points at this point in time.
On what you're paying, or so is all.
That's the market rate.
Market rate.
You'll find, right, this week, we're refinancing debt, about $1 billion of debt, which is about 4-ish%.
Mm.
When we refinance, it's gonna be six.
Mm.
In the coming year or so, we have others that we're gonna refinance at lower rates, higher rates. What we thought it prudent is, not knowing which way interest rates were gonna go. N ot knowing what was gonna happen in the macro, recession, deep or not, et cetera. Prudently pay off a little bit more of the debt.
Okay, that's where that incremental cash going.
That's where. Yeah.
Okay.
We double the rate at which we were increasing. We were increasing the capital return by 250 basis points every quarter. We started at 50, 52.5, 55, and then at 62.5, we said, let's double it to 500 basis points every quarter. We went from 62 and a half to 67 and a half. This quarter, 72 and a half. If you just play that out, we're in eight quarters or seven quarters more. W e're 100% there. In the meanwhile, you know, we're continuing to buy back shares, if the right opportunity presents itself at any point in time, we can go faster. You know, we have a line that is available to us to buy more or buy less. Consistent with that, we could buy ahead.
If we thought that we needed to go do that. We thought being programmatic was important to send a signal of how are we thinking about it and how consistently are we gonna do this. Since we're new in that journey, six, seven quarters or so, that's the signal we're trying to provide, is a consistency of how we're gonna do it.
Got it. You know, I noticed a chart you showed as well. It had your free cash flow per share, and it's growing a lot. You have a competitor who has been very successful with creating a story and a narrative around free cash flow per share. That's worked very well. Is that a metric now to which you're managing the company, or you said something that you're?
Yeah.
All right.
We have been, you know, managing to that for a while. We haven't broken it out. We said, "All right, it's time to put that chart, show you what we're doing." You can see what the actions we're taking are, with respect to that, you know, particularly since, we're using free cash flow as a big part of how we're positioning capital return strategy and where we're going. I thought it was important to see how does it compare with the S&P 500, as well, right? It's right in the top 12% for S&P 500 companies and what we're doing.
Got it. We've got a few minutes left. Should we go to the audience-
Please
... lightning round? What end markets are allowing you to continue to grow sequentially when many others are seeing revenue growth declining in 2023?
You know, within each end market are both strengths and weaknesses. Industrial is our largest end market. It's 41% of our revenue. Aerospace and defense are doing really well. A lot of the renewable energy is going really well. Factory automation is doing really well. But I'm sure there's pieces inside it which have challenges as well. Same with automotive. It continues about 17% of our revenue. Again, you know, large pieces of it doing well. There may be pieces that are going slower for whatever reason, et cetera. Data center, we talked about, it's 19% of our revenue. Yes, there are parts of data center which are not doing well, but there's other parts of data center that are doing well. Those are the three big representative parts of our end market. I think it's about 77-ish% of our revenue.
Mm-hmm.
That end market exposure is something we have curated over time and is giving us a better durability in the face of downcycle dynamics in other end markets.
Mm.
We're not saying we're immune to the cycle. We're just saying we're more fortunate in the end market exposure we have.
Got it. That leads into the next one: Would your sales be declining absent your PSP commitments?
You know, it's almost impossible to know because a PSP commitment is a customer's choice. We don't go and ask them every single time, you know, "Are you committed, are you not?" absolutely they are buying product. The fact that PSP continues to be at about the same rate. you know, we're getting new backlog every week. We ship stuff out, we get new backlog in, that PSP, as a percentage of the total, remains at about the same level it was 18 months ago. It's telling you that the program is doing something for customers. It is solving a business problem. At some point, when it's ceases to solve that business problem for them, w e'll cease to offer it as a program. In fact, our customers have gone the other way over the last 12 plus months. To this day, we still sign new long-term supply agreements.
Mm-hmm.
Where people are extending that over a five year period of time, looking to make mutual commitments on both sides. I think it's the nature of the end markets we have, and the customers, and the value that their end products have, that they want to ensure our supply enables their success.
Got it. Margins are very high, at or even above your targets. How sustainable are they?
We believe that they are sustainable gross margins and operating margins subject to whatever short-term things. We've put a floor on the operating margin. I think it's, you know, an absolute, you know, worst case condition in terms of where it's gonna be. I think you can also look at, over time, what have we done for gross margin? What is driving it in this cycle is the, you know, improvement in absorption for what we've had. You can see the mix of the products. We've got far richer products that we're building, more of. Clearly, there's no price pressure in this environment as well. We really have not had price pressure on what we've done, because our pricing is all done at the time of design. They're based on long-term pricing agreements.
You guys weren't taking pricing up, like, during the pandemic, or to offset, like, cost increases? I mean, clearly, you must-
We did in the last two years, but that's a very unique circumstance. I don't expect that that's a long-term capability to go raise prices. We did it to cover the cost increases, to be gross margin neutral on where we're at. You look at the previous 25 years, right? We weren't taking prices up. We did stop taking prices down many, many years ago, before the pandemic. It was crazy that in an industry that was growing 4% or 5%, people were expecting 6%, 7% price reductions every year, and where they were at. We stopped that madness many years ago. I think price increases really are a factor based on cost increase in the last couple of years.
Got it. Any plans to write a book of your own?
You know, some people are good at writing books. Some people are good at reading books. I'm in the latter category. I like to read books. I like to learn from them. Really, my passion is about building the leadership pipeline for Microchip, from whatever I can learn from the books that Steve writes or other people write, and to be able to go do it. I'm a book reader, not a book writer.
Got it. I think with that, we'll, we're out of time. We'll close it. Thank you so much.
Great.
This was great.
Thank you. Appreciate it.