Axcelis Technologies Inc (ACLS) 2016 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Axcelis Technologies Third Quarter 2016 Conference Call. My name is Liz, and I will be your coordinator for today. (Operator Instructions)

I would now like to turn the presentation over to your host for today's call, Mary Puma, President and CEO of Axcelis Technologies. Please proceed, ma'am.

Thank you, Liz. With me today is Kevin Brewer, Executive Vice President and CFO; and Doug Lawson, Executive Vice President of Corporate Marketing and Strategy.

If you have not seen a copy of our press release issued earlier today, it is available on our website. Playback service will also be available on our website as described in our press release.

Please note that comments made today about our expectations for future revenues, profits and other results are forward-looking statements under the SEC's safe harbor provision. These forward-looking statements are based on management's current expectations and are subject to the risks inherent in our business. These risks are described in detail in our Form 10-K annual report and other SEC filings, which we urge you to review. Our actual results may differ materially from our current expectations. We do not assume any obligation to update these forward-looking statements.

Today, Axcelis reported third quarter financial results with revenues of $65.7 million, gross margins of 36.7% and earnings per share of $0.07. Our systems mix in the quarter was 100% mature foundry and logic, supporting continued strength of the Internet of Things. The high-energy and medium current tools we shipped in the quarter were driven by image sensors, RF devices and silicon carbide power devices. The mature process technology segment also drove increased demand for used tool, pushing our CS&I revenues above typical levels.

Q3 was unusual in that it marked the first quarter in two years that we've had no systems revenue from the memory segment or from Korea. 86% of our systems revenue for the quarter came from the U.S. and Europe, with the remaining 14% from China. Typically, we see a mix of about 25% from China and 25% from the U.S. and Europe combined, with Korea usually representing more than a third.

Also during the quarter, at our customer's request, we agreed to work on additional memory application opportunities for the Purion H, which required us to keep the evaluation open. This evaluation is expected to continue into the first quarter of 2017. This decision drove our revenues to the low end of guidance but delivered higher-than-expected EPS as a result of continuing improvements in gross margin in the quarter.

As we anticipated, memory spending for ion implant systems will pick up in Q4 and accelerate during the first half of 2017. We expect some initial high energy tools to ship in Q4 to support both near-term DRAM requirements and a sustained NAND cycle. This will be followed by increased spending activity for additional Purion products in the first quarter and throughout 2017.

Guidance for the fourth quarter is being driven by a slightly slower memory ramp than previously expected. Customers are beginning to buy in Q4 but are scheduling some of their implant spend in the first quarter.

As a result, we are forecasting a relatively flat Q4 with revenues of $65 million to $70 million, gross margins of 36% to 38%, operating income of $3 million to $4 million and an EPS of $0.04 to $0.08. However, in Q1, we expect a material improvement in revenues, gross margin and EPS driven by the forecasted increase in memory spending and continued strength in the mature technology segment.

The slight delay in memory spending in Q4 will also impact our 2016 market share expectations. We now expect to maintain our implant market share of approximately 18% based on a total available market of $850 million. However, the strong start to 2017 will put us back on track relative to continued share gains. We now forecast 2017 market share in the 22% to 28% range based on a total available market of $950 million.

Our long-term industry outlook remains the same. The sustained memory cycle has started. The Internet of Things continues to support strong business for both systems and service in the mature process technology segment, and we continue to control costs and make gross margin goals, driving improved earnings per share. Our long-term market share objective remains at 40%.

We have done a good job executing against our primary 2016 objective of setting the table for 2017 with new Purion penetrations. To date, we have placed Purion products in 13 new customer fabs. Seven of these sites are new Purion customers and six are new fab locations for existing Purion customers. A key element of this success has been the introduction of product extensions to our base Purion product family. These extensions have been designed to address the specific needs of customers producing not only memory devices but also image sensors and power devices.

These extensions include the new Purion EXE and VXE, providing the highest levels available in a production implanter, the high temperature Purion M with 150-millimeter silicon carbide capability to the power device market and significant productivity and source life enhancements for the Purion H. These new product extensions account for nearly 40% of these new placements.

These new penetrations are coming from expanding our customer base through new sales and evaluation placements, increasing the number of qualified recipes at existing customer sites and securing capacity production buys as evaluations close.

We are currently working on multiple additional evaluation opportunities for both the Purion H and Purion VXE. Interest in both products continues to be very high due to their technical differentiation. The Purion H has significant advantages due to its unique spot beam architecture, and the Purion VXE is the only system that can deliver the very high-energy levels desired by image sensor customers.

Continued development of additional product extensions, enhancements and new products in the Purion product family are key to Axcelis' growth. To further this effort, we recently hired Dr. Russell Low as Executive Vice President of Engineering. Russell will be responsible for developing and executing programs to ensure the success of the Purion platform. Russell will work closely with Bill Bintz, who will shift his primary focus back to the marketing area as the EVP of Product Development. Having worked together in the past, Bill and Russell are looking forward to collaborating with Axcelis teams across the business to drive programs that will regain our market share leadership. Bill will also focus more of his time supporting efforts to continue the expansion of our customer base. This is critical in keeping us on track to take full advantage of multiple large projects and fab expansions anticipated in 2017 and 2018.

Customer fab projects over the next couple of years are expected to support the continued growth of the Internet of Things using mature process technology, advanced logic technology to support mobile and data center applications and a large sustained memory build supporting accelerating solid-state storage requirements.

At Axcelis, our top line growth and market share gains will be driven by our new product extensions, strength in mature process technology, the sustained memory cycle and the innovation benefit customers receive from having multiple suppliers. This innovation benefit is measured by productivity, yield and cost of ownership. Our customers realize this benefit by splitting their business evenly between two strong implant suppliers and gaining significant engineering expertise to support their process development efforts.

Now I'd like to turn it over to Kevin to discuss our financials.

Thank you, Mary. Before I review the second quarter financials, I'd like to take a minute to cover our gross margin objectives and results.

We remain focused on gross margin improvement and understand its importance to driving higher earnings per share. Our long-term business model has not changed with our gross margin target at 40% or greater, driven by continued cost out activity, tiered pricing strategies and higher volume.

In Q3, gross margin was 36.7%, down from Q2 as expected. The initial shipments of our new Purion M for silicon carbide applications carry higher cost associated with the development cycle and some higher cost material. We continue to deliver quarter-over-quarter margin improvement on Purion products through value engineering projects, supply chain optimization, lower manufacturing costs and reductions in install and warranty costs.

As Purion H and Purion M products mature and additional cost out initiatives are realized, we expect margins on these products to be more closely aligned with Purion XE.

Since ramping production of the full Purion product line with the Purion H in Q1 of 2015, system standard margins has improved 680 basis points on a rolling four-quarter average. New product line extensions such as the EXE, VXE and silicon carbide tool enable tiered pricing strategies that when combined with ongoing cost out initiatives with planned volume increases drives overall business gross margins to greater than 40%.

At any given quarter, we can expect to see some fluctuation in gross margins driven by a mix and other one-time events. However, the important thing to remember is we are making steady progress on lowering our cost of goods sold, which in turn is driving system margins higher.

Looking at our third quarter results. Q3 revenue was $65.7 million compared to $64.5 million in Q2. Q3 system sales were $30.9 million compared to $33.7 million in Q2. Q3 CS&I revenues finished at $34.8 million compared to $30.8 million in Q2. Q3 sales to our top 10 customers accounted for 70.4% of our total sales compared to 73.8% in Q2, with three of these customers at 10% or above. Q3 system bookings were $23.1 million compared to $41 million in Q2, with a Q3 book-to-bill ratio of 0.86 versus 1.13 in Q2.

Backlog in the third quarter finished at $20.6 million compared to $28.9 million in Q2. Q3 combined SG&A and R&D spending was $20.5 million, flat to Q2.

SG&A in the quarter was $12 million with R&D at $8.5 million. In Q4, we expect SG&A and R&D spending to be approximately $21.5 million, primarily driven by higher evaluation costs required to support expansion of our Purion footprint.

Gross margin in Q3 finished at 36.7% compared to 39% in Q2. In Q4, we expect overall gross margins in the 36% to 38% range, in line with our 2016 year-end target.

Operating profit in Q3 was $3.6 million compared to $4.6 million in Q2. Q3 net income was $2.2 million or $0.07 per share, above guidance and consensus. This compares to $2.9 million or $0.10 per share in Q2.

Q3 inventory ended at $117 million compared to $110.6 million in Q2. This increase is driven by the Purion XE material purchases required to support near-term memory requirements.

Q3 accounts payable were $20.2 million compared to $26.8 million in Q2. Q3 receivables of $45 million compared to $63.5 million in Q2.

Q3 total cash finished at $72.5 million compared to cash balance of $67.8 million in Q2. We expect Q4 cash to be in a mid-$70 million range.

Overall, I'm very pleased with our financial performance in the third quarter. We generated solid earnings despite the absence of memory spending. We continue to tightly manage expenses while making the necessary investment to expand our Purion tool footprint. System level margins improved again on a rolling four-quarter average, and our balance sheet continues to be strong, which I view as an important requirement for our continued growth.

Now I'd like to turn the call back to Mary for closing comments.

Thank you, Kevin. As a result of the full Purion product family, the robust IoT market and stronger operating model, Axcelis is now able to sustain profitability in periods of low implant memory spend. The new Purion product extensions are driving further adoption, share gains and higher margins. As the industry gears up for strong build cycle, we expect our Purion penetration success to position us for growth in both 2017 and 2018. Share gains in these years will continue to be driven by a broader customer base and increased spending in both the memory and mature process technology segments. Our gross margin improvement initiatives remain on track, and we believe that we are well positioned to deliver our long-term model, which will yield attractive earnings for our shareholders.

With that, I'd like to open it up for questions.

(Operator Instructions) Your first question comes from the line of Craig Ellis with B. Riley. Your line is now open.

Thanks for taking my question and congratulations on the ongoing margin progress in the business. The first question, I jumped on a little bit late, was just a higher level question on the back half of the year. Relative to three months ago, can you just frame how memory is playing out versus what you had expected and then the same thing versus foundry and logic?

Sure. Well, we did expect memory to begin to pick up in Q4 and then gain additional steam moving into 2017, both Q1 and beyond. The best information that we had really as recently as our Analyst Day was that the spending would start in Q4 and in fact would likely be a little bit heavier in Q4 than it has played out. What has happened is John Aldeborgh and I have actually spent considerable time in Asia recently and have gone to our customers and confirmed some of the ship dates. And as it turns out, what has happened is that some of the spending that we had expected to happen in Q4 will actually happen in Q1. So at this point in time, as we said in the call, Craig, we expect actually to see a material improvement in our Q1 revenues, gross margins and EPS. So at this point, it's really a timing issue more than anything else. Nothing has fundamentally changed in our story in terms of market share gains, gross margins. The long-term business model remains intact.

Sure. And just to understand that memory timing shift that you're talking about, Mary, is all of the memory that you have previously expected in the fourth quarter falling into the first quarter with some of it fall into a subsequent quarter in 2017?

Yes. I think what we'll see is that we'll see -- we'll certainly see memory spending not only in Q1 but also into Q2. We've talked for a long time about how we expect to see an uptick in memory, perhaps a little bit of DRAM but more likely in the 3D NAND area moving into 2017.

Okay. The next question is more of a second half question, just on the spares and services business, put like that, take that to $34.5 million in the third quarter, not unusual for it to do that periodically. Is it going to retain that level in the fourth quarter, or should we expect it to normalize back at something closer to $30 million to $31 million?

Yes. I think we expect it to normalize back, Craig, to the $30 million to $31 million. It does get a little choppy sometimes in quarters depending on the used tool volume that's in there, this quarter, just the way things played out with some of the upgrades and tools and stuff have popped it up a little bit higher.

Okay. Thanks, Kevin. And then lastly, another higher level question, just from your vantage point looking at the broader ion implant market, any thoughts on where we're tracking for this year relative to what I think was kind of an $850 million midpoint previously and preliminary thoughts on next year's TAM? Can it get into that mid-900 range? Or is it likely to be something less than that from the intensity that you're seeing within the customer base? Thank you.

Craig, our view is for 2016, it's coming in right around $850 million. And we do see that next year, it gets back up into the $950 million range and then will grow a little bit more beyond that.

And share thoughts for this year, Doug?

Share thoughts for this year, we expect 2016 will be pretty much flat with 2015, so right around in that 18% range. And then for next year, it grows to 22% to 28% on the higher TAM.

And what would account for the variance from the prior 20% to 25% target versus something that's closer to 18% this year? Where is their variance in the plan that the company would have had at Analyst Day and midyear?

It's all in memory, as Mary discussed. So just the memory systems that we had anticipated would be in the fourth quarter that are coming in, in the first quarter shift the annual TAM. The business, the total units, all of that stuff hasn't changed at all. It's just where the fence post is strong.

And I'm sorry, one more question. Doug, you mentioned the share number for next year, what was that number you provided?

22% to 28%.

22% to 28%, okay.

Your next question comes from Edwin Mok with Needham & Company. our line is now open.

Guys, this is Arthur on for Edwin. Thanks for taking our questions. I really appreciate the visibility that you offered going into 1Q. You mentioned you saw some increase in mature foundry/logic in 1Q on top of some forecasted increase in memory spending. On the mature foundry/logic side, what are you seeing in the market that is driving that growth in 1Q? Is it the continued trends in image sensor, RF and silicon carbide power devices?

Yes. So I think the commentary was more that's Q4 -- or that Q3 was very much weighted 100% to foundry and logic. As we move into Q4 and into Q1, we'll begin to see the memory side come back much stronger. Right now, we see the mature logic and foundry market as very stable. It's in a good place, and it will bounce around quarter-to-quarter as is the nature of that particular segment. But it's being driven by image sensor power device, RF. It's very dependent on which customers are expanding and for kind of what end products that they're selling into. So there is some dependency in that market on product cycles for their customers, whether it's automotive or phones or telecom type stuff.

Great. Thanks for that color. And so you guys have highlighted good progress with the Purion M platform over the past year. I think you recently announced a multisystem order for manufacturing devices for the wireless communications market. Just wanted to get a sense of, longer term, how do you envision the mix shifting within the Purion product family.

Well, I think the Purion product family, if we look, Purion H is very much targeted -- well, first of all the products are targeted across all of the segments. But Purion H is very much memory-focused. That's where the volume is. It's where the highest use of high current tends to be. We see a lot of activity on Purion XE, again, from memory, especially in the NAND side. But there's a lot of use of high energy in image sensors as well as several of the specialty devices. So it's -- so there's been an uptick in terms of high energy use in the foundry market, for example.

And then Purion M, we're seeing that kind of across the board. We've got the specialty customers that are doing things like the silicon carbide, where it's 150-millimeter ion implant. And we've got what was recently announced yesterday was the RF customers. That's a standard type of product going into that segment. That same version of Purion M would sell into the DRAM and NAND markets. And then the image sensor market likes Purion M quite a bit as a result of the low metals contamination. So we see it spread. It's very much a product that fits very well across all the segments.

Your next question comes from Patrick Ho with Stifel, Nicolaus. Your line is now open.

Mary, given the consolidated customer base that you see today, so it's completely understandable on the timing moves from customer to customer. However, a lot of your customers tend to be both NAND as well as DRAM. Was there one specific market segment among -- between those two that caused some of the, I guess, the timing delays?

No, I don't think so, Patrick. We are participating on both sides of the equation, so to speak, in terms of DRAM and the 3D NAND. So it was really just a matter of how the customers laid out when they want the shipments, in particular, for these systems. There wasn't -- and there wasn't any one thing that caused or any one project that caused some of the things in Q4 to actually happen in Q1. And I wouldn't call it a push out either. I would just say it's simply timing. As the customer firmed up -- as the customers firmed up their plans, this is just where the equipment sell.

Okay, fair enough. And then maybe going to the foundry side of things. China continues to be a growth opportunity for the entire industry, particularly not only the mature technology side, but on the 8-inch wafer side, where you've talked about in the past the Purion penetration is there. Can you give a little bit of color of, one, the China opportunity for Axcelis as a whole, and maybe also on infrastructure that may be needed to be built to capitalize upon that opportunity?

Okay. So Axcelis actually has a pretty good position in China. We've actually gone back and taken a look at some statistics on our revenues in China. And I would say on an annual basis, we actually have about 25% of our systems revenues come from China. So again, we already have a pretty good installed base and infrastructure there.

I think the way we're looking at China is there are a couple of different segments. There is fabs that are basically totally Chinese-funded. There are the fabs, who I'll call them domestic fabs, like the SMICs and then there are the fabs that have corporate parents that are outside of the country. And so we've taken a look at what the requirements for all of those. And I would say that based on that, we expect to participate in most of the market opportunity that is there. A lot of these customers have actually recently taken at least one of our Purion products. And that's one of our major strategies, is to lead with one and then go in and follow with additional types, not only repeat orders for the one that's there but then additional types of Purion products.

So I think -- again, I think we've got a lot of the seeds sown there. We've got good relationships. The thing that we do need to do -- and I'm sure you've heard this from our peer companies and our competitors, is it's a huge geography. And so with some of these new fabs, there is startup required in terms of new offices, laying in inventory, making sure that we have the appropriate number of field service and applications people available to actually serve these new facilities. So that's an analysis that we're going through right now, we've actually been going through and continue to go through. And our goal is to make sure that we make the appropriate investments to be able to participate in that growth.

Great. And maybe a final question for Kevin in terms of gross margins. Now in the past, we've seen the effects of evaluation units on specific quarters when they get recognized. As we look forward, how many more of these at least over, say, like the next 12 to 18 months, can we expect where there may be a quarterly, I don't want to say blip, where you may experience one of those quarter impacts? Are we only expecting one or two more of this type of quarters? Or is it going to be something that still has, I guess, legs or extendability going into say even 2018?

I would definitely say, Patrick, through 2017, based on the current evals we've got out there and what I know that's coming at us, we'll continue to see recognition, and then there are some that are going to spill into 2018. I guess the only thing I would say though was as the volume increases, the impact of one of these tools hitting in a particular quarter will have a less of an impact, obviously, on a bigger quarter than it does on the current quarters.

And the other thing is too -- unlike some of the first evals, we're getting a little bit better for these tools. The costs are a little bit better. So they're never going to be a great story, but the impact of them should lessen as we move forward. But to your original question, we're going to see this in '17 and out into at least the first half of '18 in terms of closure.

But (multiple speakers) --

Yes. So right, I mean, we're still saying by end of 2017, we expect to exit 2017 with gross margins in 40% range. So we're not backing off our gross margin targets. These things are baked in there, I guess, is the best way to put it.

(Operator Instructions) Our next question comes from the line of Mark Miller with Benchmark. Your line is now open.

I'm sorry, I was also on a call that was running over, but could you give the bookings and the system backlog again?

So the bookings were $23.1 million. And the -- so that was a book-to-bill of 0.86, and the backlog was $20.6 million for the quarter.

Okay. In terms of the expected share gain next year, since NAND flash is going to be probably the strongest area of capital equipment increases, are you expecting to put most of the share gain in that area? And also will it turn more for Purion H type systems relative to Purion M, you've done better, I think, with your Purion M at certain memory manufacturers than Purion H. I'm just curious what's going to -- what product is going to drive most of the share gain?

Mark, it's going to actually be a pretty good mix. At the beginning part of the year, we would expect, with some of these new projects kicking off, that it's a little more weighted towards high energy, towards the Purion XE as that tends to be some of the first things that they move in from an implant standpoint. And then we would expect to see the Purion H. Purion M, well, it kind of cuts across the board. So we'll see probably a little more activity from that like we have recently in the mature foundry and logic market.

Okay. Just one final thing. What was the cash flow from operations?

We finished at $72.3 million, $72.4 million, up from $67 million. So we generated about $5 million in cash in the quarter.

We have a follow-up question from Edwin Mok with Needham & Company. Your line is now open.

Hey guys, sorry, I joined a little late. Just -- and if this question was asked, I apologize. But I think let's talk a little bit about timing of shipment, that give you good visibility in 1Q but those shipment will happen in fourth quarter. Are we talking about one or two tools, especially like given the high ASP of Purion XE? Or are we talking about multiple tools? And that's timing is starting in 1Q, should we expect a more front-end-loaded first quarter, meaning largely these things are shipped in January, early February time frame?

Yes, I would call it multiple tools. And Q1 actually does have more systems that would be front-end-loaded versus some of the recent quarters that we've talked about, where I know people have been asking about cash, and we've talked about how the quarters have been more back-end-loaded.

So you missed the commentary that it really is simply a matter of timing, and December 31 comes when it does, and some will ship before that and some will ship after that. But there's really been no change in terms of our expectation on the number of and types of tools that we will be getting between Q4 and Q1. It's just simply a timing issue.

Okay. That's fair. And then maybe just kind of follow-up question around that. You guys had pretty strong logic sales this quarter, and actually even like earlier on the year. Obviously, it takes some time for customer to digest capacity, right? Do you have visibility into like additional projects that you believe could materialize in '17 that will benefit you on that side, even China or in some other areas? Can you give some color on that?

Yes. So Edwin, the mature market in the past, you always heard me refer to it as the whack-a-mole market. Clearly, with the IoT activity over this last year and a half, it's become much more than that. So it's a little more predictable.

The thing that really drives that at this point is product cycles for their customers. So whether it's automotive, phones, the various devices they're selling into. So that drives when customers are going to buy. And we're in touch with each of those as they're planning.

And then the foundry market to support the guys that are fabless or fab-light, and those guys are investing pretty consistently in and that's a heavy amount of investment in China right now. And so we're very active there. And again, visibility is interesting in that market because you have visibility that there's going to be a need for a new modem device that's coming and you've got two or three potential customers that are looking at it, which you don't know until you get closer is which one of them is going to get it, and usually, that's when you'll see the P&L.

Great. Last question I have. On the share gain targets you guys laid out for '17, obviously, you guys have been really strong in high energy, but in high current and medium current, those are the areas that we have gained share, right? How much confident that, that customer will give you -- in the position where you guys are splitting -- in the customer position where you guys are splitting some of these between you and your competitor, how much confident that your customer will give you that business given your competitor can -- is a much bigger company, so they have potentially can use commercial terms and stuff like that to try to limit their share loss, if you will.

So Edwin, it comes down to this innovation benefit that we've talked about a number of times. What the customers are gaining out of this is they're clearly gaining better technology. They're gaining competition between Axcelis and our larger competitor, our largest competitor. And all of that really drives advancements in productivity, process performance, lower cost of ownership. And those are things that, over the last couple of years since we've had the Purion H out in the fields have been clearly demonstrated. There are solid numbers behind the fact -- behind this innovation benefit.

So customers understand the disadvantages there. And I think that those who have tried it have enjoyed it. We're beginning right now to penetrate where we have evaluations, and we actually have a number of ongoing engagements right now from a demo and applications perspective. Those customers are all seeing our major competitor respond in one way or the other. And that's a great thing. So that's really what's driving all of this. At the end of the day, it comes down to this innovation benefit.

We have a follow-up question from the line of Mark Miller with Benchmark. Your line is now open.

Just wondering what causes you to have greater confidence now than you had previously when you thought you would be picking up the share earlier this year. What gives you confidence now that your Q1 forecast is going to be as stated?

Well, we always knew that the projects that customers, in particular, our memory customers are starting to spend on, we knew that they were going to happen. Again, the question was always around the timing. And as I said, a little bit more of it is going to happen in Q1 versus Q4. We've spent a lot of time now with those customers talking about exactly when they would like the shipments and when those shipments -- our expectation for when those shipments will go out. I would say there tends to be, at this point in time, since they've made the commitment to make the investment, there tends to be a lot more certainty around the date. And there's certainly a lot more certainty around the dates with customers buying for memory than there is this mature process technology segment.

So we feel pretty good. And in fact, we've already got orders for some of the units out into the first quarter. So those -- things are starting to really solidify, and so we're gaining confidence daily.

It wouldn't have anything to do with some of the challenges with maybe trying to bring up next-generation 3D NAND flash, people might be having more challenges than they thought?

No, the customers are doing fine on their end. It's more of the commodities game. When is the right time to start adding capacity based on what the pricing is for DRAM or NAND.

Spot flash pricing just went to a multiyear week high last week, so maybe they're waiting for that, so.

Yes.

This concludes the Q&A portion of the call. I will now turn the call back over to Mary Puma, who'll make a few closing remarks.

Thank you. I want to thank you for your continued support. I'd also like to thank those of you who attended our Investor Day in New York City last month. Materials from that event as well as our current investor presentation are available on our website. We're going to be presenting at the Midtown CAP Summit in New York on December 8 and the Needham growth conference in January. We also expect to be on the road during the quarter doing some NDRs. We hope to see you soon. Thank you.

This concludes the presentation. Thank you for your participation in today's conference. You may now disconnect. Good day.