Next up, we've got Photronics Inc., NASDAQ-traded company PLAB, $1.8 billion market cap. We've got Christopher Dayton, the Senior VP of Finance, and Ted Moreau, the Vice President of Investor Relations. I'll turn it over.
Does this move? Thanks, Andy. What I'm going to talk about, obviously, the company, we're kind of entering into this interesting period of time where we're investing in the business because we're taking advantage of what we see as a growing opportunity to diversify the manufacturing of the semiconductor industry throughout the world. Part of that is coming into the U.S. with a reshoring of semiconductor production here, also globally, as the world diversifies away from China and Taiwan. We're investing in other regions as well, we'll get into that. The reason that the company is also interesting is because there is a very significant increase in the dollar value of what we do as you get into more advanced semiconductor production. As we invest in our business, we're looking to capture more of that opportunity.
We'll talk about that in a little bit. These numbers are trailing 12-month as of our most recent April quarter. We have an October fiscal year- end. Generate really good cash flow at the operating level, typically around 25% of total revenue. This year, free cash flow is going to get squeezed a little bit because of a very significant CapEx guidance that we've established for the full- year. Generally speaking, pretty good cash flow. You can see in the lower left-hand corner, the geographic mix is based off of geographic origin. Between China and Taiwan, that's, I think, about 40%, 50% of our business. Korea kicks in. Asia is a very significant percentage of the total revenue, which obviously reflects where much of the world's semiconductor production occurs.
We do have two basically kind of business lines, flat panel display, which is 28% of total revenue on the trailing 12 months, semiconductor or what we call integrated circuits is the other portion of the business. We make a photomask, I don't know if everybody knows what it is, I'll just very quickly kind of walk through it. Basically, we take a piece of glass. In the semiconductor world, it's 6 in by 6 in. In the flat panel display market, it's several meters wide, a couple meters tall. We basically etch into the glass the semiconductor chip design for the semiconductor market or the design for a flat panel display.
In flat panel display world, this flat panel TV. You would need a photomask that would be at least of that size in order to capture the entire schematics of the flat panel display. What happens is, where you take the design that we receive from our customer base, and in the semiconductor world, the customer is any location that manufactures semiconductor production. A lot of our customers end up being fabs that don't have design capabilities, or it could be a semiconductor company that has its own manufacturing operations. A fab being like a UMC or a GlobalFoundries, Taiwan Semiconductor, that sort of thing. A semiconductor company that has its own facilities for operations, such as like a Texas Instruments, Samsung, and whatnot.
We take the design, etch into the piece of glass that design, and then ultimately ship off to the customer. The photomask gets inserted into a lithography tool, very front- end of the semiconductor manufacturing process, and it projects the image down onto the wafer. Front-end semiconductor manufacturing. If you do not have a photomask, you cannot manufacture a semiconductor chip. Very critical to the manufacturing process. Similar events or layout of how you get the image that has been created onto a display for the creation of a display. Historically, the photomask market has been about 1% of the total semiconductor market. That is changing because of the very significant ASP increase that's occurring in the semiconductor market, particularly for memory and very leading-edge chips for AI architectures.
It remains to be seen what percentage of the photomask market will be for the semiconductor market this year or in the next couple of years. Generally speaking, it's been 1%, it's kind of tracked the growth of the semiconductor market. There is a captive versus merchant aspect to the industry. Captives. There's six captives in the world, and they historically represent about 2/3 of the total photomask market. The captives being Taiwan Semiconductor, Samsung, Intel, SK Hynix, Micron, and SMIC also, which stands for Semiconductor Manufacturing International. They're based in China. They tend to manufacture the bulk of the semiconductor chips that are being designed at 3 nm , 4 nm, 5 nm, 7 nm, 10 nm, and they outsource a little bit more at what we consider trailing edge. That's where we participate for the most part.
I'll get into that shortly. For us, we have 11 clean room facilities or fabs throughout the world. Your proximity to your customer is very strategic. One of the ways that you win, obviously you need high-quality products, very little defects, right? That's printed onto the glass because if there's a speck of dust or a speck of dirt that should happen to be on the glass during the wafer. When the image is projected on the wafer, well, then that speck is going to show up on the wafer. You have to have very pristine manufacturing. The other aspect of our business is close proximity, or the reason why you win is because close proximity to the customer. We have one- to- three-week lead times. In semiconductor world, it's typically around a week. In flat panel display, a little bit more like two to three weeks.
Turnaround time ends up being a very significant factor as to why you win. Having your facilities close to your customer base will shorten your shipping times and therefore improve your delivery times. We have three facilities here in the U.S. Our high-end facility is in Boise, Idaho. That facility is capable of handling manufacturing of semiconductor chips down to 7 nm. We have a facility in Texas. That facility is going through an expansion right now from historically about 130 nm and older, to 65 nm. Actually, what's happening between these two facilities is important because as we expand production in Allen, Texas, we are currently producing some projects at 65 nm in Boise, Idaho. But again, that facility can go down to 7 nm .
Once Allen is up and running, we're expecting an initial revenue in the month of October this year, and then grows next year, next fiscal year. As that facility is up and running, we're going to shift those 65- nm projects out of Boise over to Allen and then backfill it with higher-end opportunities that we see, which should improve the ASP profile of the company in the future. We'll talk about that shortly. We have two facilities in Europe, mostly at mainstream. In Asia, we have six facilities, one in Korea, two in China, and the last three are all in Taiwan. We do have two joint ventures. Both of them are in the semiconductor world.
One of them is based in China and one of them is based in Taiwan. That does influence a little bit how you see the financials. We'll talk about that in a little bit later. Our Korea facility, because of demand that we see from the local market in Korea, we are going through an expansion of our Korea facility as well. That is to take that facility from 28 nm and 40 nm down to 8 nm. The timeline of that is about a year behind the U.S. expansion. As we look at how does fiscal 2027 play out, how does fiscal 2028 play out? 2027, U.S. probably very likely leads the revenue growth trajectory, and then in 2028, Korea will kick in and that should drive the revenue growth that year.
As you think about these geometries, it's very important to understand the dollar value of the revenue that's generated at an older geometry, mainstream geometries, such as 90 nm, 130 nm and older, versus what is occurring at the leading- edge. Our business, we break out as mainstream, is anything that's above and beyond 28 nm . High-end is considered 28 nm down to 7 nm today. You kind of see on this far right-hand side, there's very leading- edge, and that's going to be more captives today. As we look at this graph, 90 nm , for one semiconductor chip design, we typically ship about 25 masks, and you need one mask per layer of the design of the semiconductor chip. At 90 nm , again, 25 masks. As you get to 14 nm , the number of layers increases very significantly.
On average, there's 70- 80 photomasks for that one semiconductor chip design. At 5 nm , we're also talking probably in that 80 to maybe even 100 layers or photomasks for that one- chip design. That's really helped to drive the improvement in ASPs as you get into the more leading- edge. Also, the dollar value of each individual layers increases parabolically as we get into the leading- edge. We're talking at 90 nm for the entire mask set for the one- chip design, probably tens of thousands of dollars. At 14 nm, we're talking somewhere, give or take, around $1 million for the set. Then at 5 nm , we're talking a 10x- 20x increase versus 14 nm .
Our investments in the U.S., our investments in Korea, are both designed to try to capture more of this curve at the parabolic move that is occurring. Part of that is due to complexity that's occurring at the very leading- edge, and part of it has to do with what's called extreme- ultraviolet lithography or EUV. There's a little bit of a distinction because the captives handle virtually all of the EUV production that's occurring in the U.S. or throughout the world today. We have an EUV program, but it's for R&D development for semiconductor equipment companies. We don't actually ship masks for EUV production lines today. It's something that I want to make sure is clear. That is part of the reason why we're seeing some of the parabolic move that's occurring at, say, 5 nm .
It's something that we're looking at in the future. We'll see how that plays out, but make no mistake about it, the importance of the very leading- edge on our business model. Display market, talk about that real quickly. That is also going through and about to enter into a upgrade cycle that we can see. That upgrade cycle involves what's called Generation 8.6 AMOLED. As any consumer electronics, largely we're talking smartphones, tablets, and laptops and PCs, upgrades to AMOLED or OLED technology in the coming years, they're likely to utilize Generation 8.6. Relative to LCD screens, that would be an ASP uplift for us. We see it not only as an upgrade cycle that's coming down the pipeline that we think happens later this year. Also, on top of the volumes that we expect, we do see an opportunity for ASP improvement as well.
Again, this kind of just gets into why do you win. Operational excellence. We're very good at producing masks, even in the China market. I referenced where our facilities are. Even in the China market, we have a very cost competitive operations there. The responsive times or our cycle times, close proximity to the customer, designing masks that are pristine, and in line with the specs that the customer gives you, those are all the factors. It's very important to understand, we don't spend a whole lot of money on R&D. That's one of the reasons why we generate a lot of cash flow. Only 2% of our revenue comes from R&D. The reason for that is because we don't design standard off-the-shelf products. Everything that we ship out is a custom product.
Let's say we manufactured some masks for customer ABC, and then they decide to scrap the project for whatever reason, which doesn't happen, but let's say hypothetical. We could not ship that set of photomasks off to any other customer. It's all custom development. You don't need a lot of R&D in order to design your products. Where does the engineering expertise, how do we use it? It's basically on the operational aspects of the clean room facilities, utilizing the equipment to the best of your capabilities, and that's really where any of the engineering dollars are being spent. I mentioned at the beginning of the presentation that we are investing in our facilities. Again, expanding in the United States, expanding in Korea. If you look at historically, in the 2020-2024 timeline, we spend on average about 15% of revenue on CapEx.
That can kind of be considered Everybody thinks of it as kind of maintenance CapEx. That's to fuel your ongoing operations. The spike years that you see are investments in the business. In 2019, it was expansion into China, which really helped drive pretty good revenue growth over the last, let's say, five years. 2025 began the year of this new investment cycle that we're going through. In 2024, we spent $131 million on CapEx. That increased to $188 million last year. That delta was the combination of the U.S. expansion announcement in Texas, and then also we are spending more money to replace old equipment, so we call it End-of-Life Tool Upgrades. We are very good at extending the useful life of equipment in our fabs. We're actually better at it than our peers are, which is great because you're deferring CapEx.
The problem is we kind of hit our limit, and so we're going through this time period of a couple of years where we have to replace old equipment in our facilities. That increase of about $57 million or so, is the combination of the United States and then the End-of-Life Tool U pgrades. For 2026, we set a budget of $330 million for CapEx. That's going to be year two of the U.S. expansion, and every time we announce an expansion, it typically takes about two years of CapEx burden. The second piece is the End-of-Life Tool Upgrades . We said that this year will probably be peak year for End-of-Life Tool Upgrades . Incremental to that is the Korea expansion. Year one of Korea.
That covers kind of like the delta between the $330 million and the $188 million. Everybody always asks, "Well, what are you going to spend next year?" Well, we haven't set a budget for that. We have an October fiscal year- end. You'll probably hear what the budget is going to be when we report the December quarter, so that's going to be in mid-December. I'm sorry, report the October quarter, which will be in mid-December. But I just want to help you kind of walk through how to think about it. The United States, this is year two, so that particular project is likely to come off the CapEx burden for next year. End-of-Life Tool Upgrades , I said this year is going to be a peak year, so that means next year pulls back a little bit. I can't give you exactly how much, but it should be down.
Next year, Korea will be year two of the investment. If you take that all into consideration, that would suggest that the CapEx burden for next year comes down a little bit, right? On the most recent earnings call, and actually we've said at the last couple of earnings calls, we do have a number of projects in the pipeline that we have to determine whether or not we're going to approve or not. For that reason, there is the possibility that we approve incremental CapEx projects, which would be above and beyond what I had just kind of outlined. Don't entirely know what the CapEx number's going to be for next year, but that kind of helps you identify what our thought process is.
If you think about this year on the $330 million, we spent $94 million through the first two quarters of the fiscal year. That would suggest that the back half of the year is going to be a very CapEx-heavy period of time. I can't say that I can't commit to it, I suppose, but that would suggest that any of these other projects probably, if they're going to get approved, is probably late in the fiscal year, so it doesn't burden this year's CapEx, but it's possible. I can't say never. Just kind of walking through again what we've been talking about, why is this an interesting time period? Why are we thinking about investing in the business? Why are there other projects in the pipeline that we see that could generate a pretty good return on the invested capital?
It's because of this regionalization of semiconductor production, reshoring of semiconductor production in the U.S. The captives, I had outlined six of them earlier, they're looking to outsource more. Certainly, they are relevant to where we are investing and why we are investing. The node migration, because every time a semiconductor chip is designed, as that chip goes through the next generation process, maybe it's a 90- nm chip in the past and it's going to 65 nm or 65 nm- 40 nm . That's what we consider node migration. We always see improved ASPs and improved revenue every time that chip goes to the next generation node. We're trying to capture more of that high-end business. On the display side, we believe we're entering into an upgrade cycle that we feel as though is going to last for the next several years.
It's all about execution. Once we get through this CapEx cycle, I don't know how long. It could be a couple of years. I don't know how long it's going to last, but I would imagine that it's all about driving cash flow, and that's something that we're pretty good at doing. Just to kind of review the most recent financial results, revenue was basically flattish year-over-year, was a little bit below what we were expecting. There was a couple of reasons for it. Let me kind of outline initially. There's three reasons for revenue was a little bit lower than what we were expecting. One is utilization rates at our customers, the fabs, is really high right now, that's delaying design starts, and I'll kind of walk through that shortly.
Number two is the demand for memory is going off the charts, the memory suppliers are allocating their memory to hyperscale data center build-outs. Some of the consumer electronics companies are having a hard time capturing memory, they are delaying smartphone releases. That's not to say that they're going to delay the iPhone. They're not going to delay the Samsung Galaxy S20, whatever we're in. The low-end smartphones are having a hard time capturing memory, that's delaying design starts. We're all about design starts. As I kind of walk through that number one point, it'll become a little bit clearer. The other aspect is geopolitical aspects. With the war going on, there was some design delays that had occurred. Let me step back and talk about this fab utilization issue.
The fabs that receive the designs from their customer base, typically, they would then share the designs with us. What was happening was because the fabs are at very high utilization rates, they are not accepting the designs of some of their customers because they said basically, "Well, hey, we're selling all the wafers we can produce. We don't need any more, so you have to wait." That's delaying the release of a design. Clearly, we are tied into design starts rather than wafer starts. If it was wafer starts, then life would be very beautiful right now, I would imagine. It's design starts, and then again, on the memory side, because smartphones are being delayed, that's delaying the release of a semiconductor chip design for the smartphone.
Those were the two biggest factors on why revenue was a little bit below expectations. We still had operating profit margins of 20%. I think, what was cash flow? I can't remember what cash flow it was. It was about $47 million-$48 million in the quarter, still pretty good cash flow. From that perspective, it's still really good. It was just a little bit below what we were originally expecting, unfortunately. This just breaks out the difference between our semiconductor business and flat panel display business. Our flat panel display business actually did really good because again, even though the low-end smartphones and PCs and tablets and whatnot were being delayed, the design releases of the Apple iPhone, the Galaxy products are not being delayed. The high-end smartphones are not being delayed because they're actually able to capture the memory.
Our display business had a really good quarter, which typically is seasonally strong in our fiscal Q2 because you're getting ready for those fall launches, right? Our semiconductor business, yeah, that was the business that had deteriorated in the quarter. We had guided-- I'll go to this page. We had guided basically about flat sequentially. We did say we're starting to see a little bit of improvement, but that's not to say that it's going to be wildly-- all of a sudden, I don't think we're going to see this massive recovery. I think it's going to be kind of gradual, and the reason for that is the memory space is sold out for what's perceived to be in the next three to five years.
How can we predict what the next couple of years are going to look like from that perspective and the influences that had it on our business? I can't predict the utilization rates at the fabs because of AI build-outs are going off through the roof, right? It's a little bit tough to project. With that said, regardless of what happens there, that things are out of our control. What we can control is execution on our U.S. business. We do believe that the United States will drive revenue growth for fiscal 2027, and we don't believe that those factors will influence the U.S. projection for next year. The balance sheet, very strong, $638 million of cash on the balance sheet, 75% of which resides at the two joint ventures that I referenced earlier, and generate a lot of cash flow.
The CapEx, again, we spent about $94 million in the first half of the year on a $330 million budget. This back half of the year is going to be a pretty CapEx intensive. With that, we have about eight minutes left. Let me know, are there any questions? Rudy, yeah, go ahead.
Could you talk about when Allen starts up, what does that do for revenues, if you fill it up, and margins? Are the margins better at 7 nm-8 nm ?
Yeah. I can't talk specifically about the dollar value of revenue that's going to be generated out of that facility. What I would say is, because that facility will accept. I guess I was supposed to repeat the question, all right. U.S. revenue expectations and margin expectations given the U.S. expansion that's coming down the pipeline. What I can say is, Allen facility accepts projects that are currently being designed or manufactured out of our Boise facility, right? That's just a one-for-one replacement. What that does is open up the Boise facility for higher ASP projects, right? Because that's going to free up capacity. The same number of plates or photomasks could go through that facility, but the ASP goes higher, right? That would be a positive for the revenue production out of the United States, obviously.
On the gross margin side, because a lot of the older projects, the mainstream projects, are on fully depreciated tools historically, right? Obviously, that's going to change a little bit because of Allen. A lot of those projects are on fully depreciated tools, the gross margin profile can be very attractive. On a project-for-project basis, there's no discernible trend between high-end and mainstream. Now, with that said, in any given quarter, because we have a very significant fixed cost infrastructure, when we generate more revenue at high-end, that means our ASPs go up, which means on the same number of plates, that means our revenue goes up, we're overcoming that fixed cost infrastructure. The gross margin profile, in theory, should go up next year out of the United States because of more business coming out of Boise at the high- end.
Hopefully I didn't botch that one too much, but I made it clear enough. Any other questions? Yeah.
Yeah. Could you tell us again the difference between AMOLED and OLED, and OLED and how it benefits you, the transition?
AMOLED is part of the OLED. When you see OLED TVs, it's probably shorthand for AMOLED, I believe. I don't know of any differently, but that's how I believe it. The benefit to the consumer is, in theory, it's supposed to have brighter colors, sharper contrast ratios. It's supposed to reduce eye strain. For folks such as myself that can't read anything that's this close to them anymore because of our smartphones, that would actually have been helpful for us about 10 years ago, right? There's benefits to the consumer. The other benefit really is it's supposed to allow the display provider to utilize a thinner and lighter piece of glass. As we get into these foldable phones, it's supposed to reduce the weight.
I don't know if you've held a foldable phone of one of these smartphones before, but they're really heavy. If there's going to be adoption of these foldable phones. One, the price point's got to come down, right? Because I remember looking at one a year and a half ago and it was $2,000, and I said, "I'll take the freak Samsung Galaxy." That's got to be a factor, and then also the weight of it. I can't imagine putting a foldable phone in my jacket pocket and all of a sudden you're walking sideways, right? Or with a tilt. Those are the benefits. For the display maker, they can actually produce. What is it? I think on the same photomask set, more production can occur over the same mask set. I think that's how it works.
Yes. Given that your products are highly customized, how do you think about expanding operational capacity and also how you negotiate ASPs?
Yeah. How do we negotiate ASPs and how do we think about capacity expansion? Typically, the expansion will occur based off of communication with the customer, or the end- customers, and then also evaluating the end- market, right? Are we in expansion time period like we are today? Certainly, the customer isn't going to say, "Yeah, we will commit X dollars if you expand here." It doesn't work that way, unfortunately. I wish it were, but they won't do that. It's a little bit more of a discussion that takes place, "Hey, this is what we're doing. This is our growth plans. Can you help support us in that?
If you do, then we'll give you production." A lot of times it centers around, and this kind of gets into we have these long-term sales agreements that last anywhere from one year to three years with many of the larger customers that we have. In those agreements, and then also in some of these discussions about expanding our business, they have indicated to us, or at least in the long-term sales agreements, they have contracted with us a percentage of market share of business that they give out to the merchant suppliers. That helps us understand what the opportunity could be. They can't commit to volumes, but they will give us a market share. Within those long-term sales agreements, we negotiate pricing that is in effect for the duration of the contract.
On this most recent quarter, yes, revenue came in below expectations, absolutely pricing was not a factor. We have already established long-term sales agreements with pricing, there weren't any effects from that. Yes.
Are you a single-source manufacturer? Is question one, and the second one that came to mind here is do you have raw material risk? Is there a commodity pricing risk?
Raw material risk and then sourcing from customers. Typically, a customer wants to have dual sources because they always want to have backup plans, right? That is the case with our business as well. It can range anywhere from, say, it's a 50/50 split to, in certain situations, it could be a 90/10 split or something, right? It varies depending on the customer, but typically there is a dual-source manufacturing strategy. Obviously, because of the very short cycle times or the very short lead times, they want to ensure that if they have a project that's coming down the pipeline that requires a very quick turnaround, they'll come to us first, and we'll quote them our turnaround time, and either that satisfies them or it doesn't. If it doesn't, then they go to their other source, and then they source it.
As far as the supply chain is concerned, the number one material cost is a piece of glass or quartz plate that we source from Japan. Our access to it has been fine as far as I'm aware. I haven't heard of any disruptions. We haven't seen tremendous cost pressure out of that so far. I think a year ago there was a lot of talk about the tariff situation, that really hasn't had that big of an impact on our business as well, right? We are approaching out of time, but happy to take another question if there is one. Yes, go ahead.
Talking about multiples and priors, what is the market share status and what was the trend of the market shares?
It actually probably requires a fairly lengthy conversation. For the global market on the semiconductor side, it's basically a two-player market. There's ourselves and a company called Tekscend. They're based in Japan. Combined, we're about 70% of the merchant supply of photomasks on the semiconductor side. The China market is a little bit unique. It would probably require a lengthier conversation. On the display side, our market share has been very steady at about 27% of the global market, say for the last five years. It's a little bit more competitive. We have a very distinct advantage on the display side at the very high- end technologically. We're able to do things that others are not, and so we've done a great job on capturing the high- end of the market.
There's more competitors in the more the mainstream portions and the lower- end of the technology curves on both display and IC, particularly in the China market. That requires a little bit more of a conversation. What we are doing on the IC side is moving away from the competitive areas in China. We're focusing very much on the high- end, which is 22 nm , 28 nm , 40 nm , and 65 nm , just to get away from some of that pricing pressure that we do see at the market in IC. All right. Really appreciate your attendance and feel free to reach out to me anytime. My email is on every press release that goes out. Thank you.