AXT Inc (AXTI) 2017 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q2 2017 ATX (sic) [AXT], Inc. Earnings Conference Call. (Operator Instructions) As a reminder, this conference is being recorded.

I would like to introduce your host for today's conference, Gary Fischer, Chief Financial Officer. You may begin.

Thank you, and good afternoon to everyone. Before we begin, as usual, I would like to remind you that during the course of this conference call, including comments made in response to your questions, we will provide projections or make other forward-looking statements regarding, among other things, the future financial performance of the company and our ability to control costs and improve efficiencies, increase orders in succeeding quarters, improve our competitive position in the market, our ability to meet market demands for our products as well as other market conditions and trends, including expected growth in the markets we serve.

We wish to caution you that such statements deal with future events, are based on management's current expectations and are subject to risks and uncertainties that could cause actual events or results to differ materially. These uncertainties and risks include, but are not limited to, overall conditions in the markets in which the company competes, global financial conditions and uncertainties, market acceptance and demand for the company's products and the impact of delays by our customers on the timing of sales of products.

In addition to the factors that may be discussed in this call, we refer you to the company's periodic reports filed with the Securities and Exchange Commission and available online by link from our website for additional information on risk factors that could cause actual results to differ materially from our current expectations.

This conference call will be available on our website at axt.com through July 26, 2018. Also, before we begin, I want to note that shortly following the close of market today, we issued a press release reporting financial results for the second quarter of 2017. This information is available on the Investor Relations portion of our website at axt.com.

Now turning to a review of our second quarter results. With that in mind, total revenue for the second quarter of 2017 was $23.6 million compared with $20.6 million in the first quarter of 2017. Of our total revenue, substrates sales increased to $19.1 million compared with $16.6 million in the prior quarter. Revenue from our raw material joint ventures was $4.4 million in Q2 compared with $4.0 million in Q1.

In the second quarter of 2017, revenue from North America was 7%, Asia Pacific, 68%, and Europe was 25%.

Two customers generated more than 10% of revenue in Q2 and the top 5 customers generated approximately 37% of total revenue, reflecting again our diversification of both products and customers.

Gross margin in the second quarter was 30.8% compared with 30.5% in the prior quarter. It's important to note that gross margin from the substrate business was higher than the corporate level, offset by gross margin from the raw material business, which included a lower cost or market write-down of raw gallium from one of our consolidated joint venture companies. This reduced our total gross margin by almost 1 percentage point. As we look forward to Q3, our expected increase in volume will benefit the gross margin, specifically as unit volume grows, the manufacturing overhead costs are spread over more units and as we all know, that helps the gross margin. I'll provide more color on this in our Q3 guidance in a bit.

Total operating expenses in the second quarter were $5.0 million compared with $4.9 million in Q1. Actually the spread was only $44,000, Q1 operating expense was $4.917 million in rounds down, Q2 was $4.961 million in rounds up. We continue to run relatively flat quarter-to-quarter for OpEx.

Total stock compensation was $310,000 in the second quarter.

Operating profit for the second quarter of 2017 was $2.3 million compared with $1.4 million in the previous quarter. Interest and other income for the second quarter was a net charge of $176,000. This net number consists of 4 categories: number one, net interest earned of $114k; number two, foreign exchange loss of $90k; number three, equity accounting on our unconsolidated joint venture companies, a loss of $188k; and number four, other items equaling a loss of $12k. In addition, the tax provision for the second quarter was $321,000 compared with $159,000 in Q1.

For Q2 of 2017, we have a net profit of $1.9 million or $0.05 per share. By comparison, we had a net profit of $665,000 or $0.02 per share in the first quarter.

The diluted share count in Q2 is 39.7 million shares. Cash and cash equivalents and investments closed at $87.5 million as of June 30, up $767,000 from $86.8 million as of March 31.

Depreciation and amortization in the second quarter was $1.1 million and CapEx was $1.9 million.

Accounts receivable net of reserves were $18.3 million at June 30, compared with $17.6 million at March 31. Net inventory at June 30 was $40.6 million compared with $39.2 million in inventory as of March 31.

Ending inventory consisted of approximately 49% in raw materials, 46% in WIP and 5% in finished goods.

Okay. This concludes the financial review. I'll turn the call now over to Dr. Morris Young for a review of our business. Morris?

Thank you, Gary, and good afternoon, everybody. 2017 is shaping up to be a solid year of growth for AXT. While much attention lately has been placed on certain applications for common semiconductor substrates, we are seeing an expanding universe of emerging applications and technologies across our portfolio that are driving growth in every one of our substrate product categories. As a result, we achieved record revenue in the phosphide substrates in Q2 and posted solid growth in semi-insulating gallium arsenide, semi-conducting gallium arsenide and germanium substrates. Our customer and revenue base continue to diversify, giving us a broad-based opportunity for continued growth.

This is particularly evident in indium phosphide, where key applications are gaining momentum. And we are also receiving money spot orders from smaller accounts serving a wide variety of emerging applications.

These small orders do help the revenue, but more importantly, they indicate an increasing interest in indium phosphide. We believe (inaudible) there are many new development programs underway, either at the research level or at the development stage level.

Although each may be a small volume today, some will likely grow into true production-level volumes in the future.

One of the positive elements in our business model is that the product cycle lifetime is very long for our substrates. One market segment that has received a lot of attention is the passive optical network applications in the fiber-to-the-home and office.

In Q2, the PON market continued to be a strong revenue generator. As many of you know, the PON market had their low in 2016 as a result of an inventory rebalancing. In the last 2 quarters, we've seen our revenue from this market bounce back, though not at the level we experienced in 2016.

China is certainly a key region, and its recovery will provide further opportunity for revenue growth for AXT. But AXT supplies into a number of regions and customers worldwide, and we see healthy opportunity in the global PON market for many years to come.

In the meantime, demand for indium phosphide from silicon photonics applications continue to grow, providing AXT with another major driver for our revenue.

Silicon photonics is a breakthrough technology in data center connectivity that enables high-speed data transfer from low power consumption over optical fiber. It is quickly the research and development phase to commercial development, with leading global companies such as Intel, HP, Cisco, IBM, Broadcom, Infinera, Mellanox and many others driving its adoption.

In less than 18 months, we have seen this application go from contributing virtually no revenue to now contributing more than 25% of our indium phosphide revenue. Further, with the number of silicon photonics product coming to the market this year, the growth rate is likely to accelerate.

In addition to data center connectivity, silicon photonics is being leveraged in a wide variety of applications in telecommunications, Internet of Things, metrology, consumer-electronics and display, health care, sensing, autonomous cars, security and others. As I said, we believe some of these applications will generate good volume for us in the future.

And finally, a third area of focus for indium phosphide is telecommunications, where the current infrastructure upgrade cycle and preparation for 5G are providing opportunities in short haul, long haul and metro development -- deployment. We believe that this application will be a market driver and contributor to our revenue growth for many years to come.

In total, we believe that we are in the early stage of indium phosphide growth, and that 2017 and '18 will prove to be pivotal years.

Overall, the optical applications are expanding, and the small number of companies supplying indium phosphide wafers should benefit from this expansion.

We're excited that the investment we have made into product development, manufacturing efficiencies and customer support are positioning us to optimize what is shaping up to be a very exciting market in the coming years.

Now turning to gallium arsenide. Shells was up in Q2, and I expect to be up by a healthy percentage again in Q3. As many of you know, this area of our business has continued to provide a solid, profitable and stable revenue base for quite some time now.

Over the wireless side, traditional applications, including power amplifiers for RF devices that did not move to SOI. We also believe that our substrates are going into cell phone Wi-Fi receivers and transmitters.

Now on the LED side of our gallium arsenide substrate business. Traditional applications include backlighting, signage and display and among others. All of these applications are expected to continue to contribute to our gallium arsenide substrate revenue for the foreseeable future.

In addition, one of the unique things about gallium arsenide is that [it governs] the spectrum of light that is unseen by the human eyes, making it ideally suited for application that call for infrared light. We're excited to see a host of new such applications coming to the market that are providing other wave of growth for these substrates.

For example, we're currently experiencing growing customer demand for a variety of other high-end applications for infrared light, in which low-EPD requirements limited number of competitors that can meet the stringent specifications. These applications include virtual and augmented reality, retinal recognition and automotive sensing, among others.

In addition, amongst the most well-publicized market development these days -- right now is 3D sensing. We view this as a major opportunity for high-end substrate manufacturers in our industry, because the strict technical requirement of 3D sensing provide a strong barrier to entry to new or low-end players. Today, only 3 competitors, including AXT, are able to provide low-EPD substrates in sufficient volume for production [band]. And we expect that market growth in the coming year will provide an exciting opportunity for all.

In total, gallium arsenide is experiencing a resurgence in demand, driven by a combination of true applications and new emerging technologies. Among other substrate material, gallium arsenide has the advantage of being a highly reliable, proven industrialized material with a healthy supply chain.

As pioneers of low now industry-standard VGF technology and as a leader in low-EPD substrate, we are in a good position to benefit from this trend.

Now turning to germanium substrates. Sales increased in the quarter, reflecting a recent improvement in the worldwide satellite market, particularly in China. The predominant applications remain in communications, positioning and navigation and remote sensing. We believe that strengthening market conditions are likely to provide opportunity for AXT's revenue growth over the coming year.

Finally, with regard to our raw material business. The market continues to struggle with overcapacity, particularly in raw gallium. As Gary mentioned, during Q2, one of our gallium joint venture took a inventory write-down, which was reflected in our financials.

Many of the markets in which our joint venture supplies are improving. Our joint venture companies are an important part of our supply chain strategy, and we do expect future improvement. Over time, we're hopeful that this may serve as rebalancing the economic supply and demand.

Before I close, I wanted to take a moment to give you an update on a couple of operational components of our business.

First, I am pleased to report that we have identified a new site for our gallium arsenide manufacturing facility, and we do expect to finalize that agreement in the very near future.

We'll give you the full details when we are able. But what I can tell you now is that it is reasonably close to our current location and it has 140,000 square feet of manufacturing space already built, plus a additional building that can be used as dormitory and offices. Existing buildings will ease the logistic requirements for our transition. As discussed, we're planning a staged transition to avoid any line down situations and provide product continuity.

I also want to note that we have recently hired 2 direct sales professionals. One will be based in Europe, and the other in the United States. We're now actively working to further build our direct strategic partnership with key customer accounts in these regions.

These individuals combine good technical skills with a strong customer presence. We believe that additional direct coverage will allow us to deepen our relationships with certain current customers and to participate in the ever-broadening universe of new customers and applications for our products.

Now in closing, it's an exciting time for our business. For the last couple of years, we have anticipated the potential growth of silicon photonics, and we're now excited to see it now making a material and growing contribution to our indium phosphide revenue. Coupled with the sizeable opportunity in fiber-to-the-home and office and a number of smaller emerging applications, we remain confident that indium phosphide will continue to be a major driver for our business for many years to come.

In addition, with a number of new application for gallium arsenide, all directed within our core competency, we are seeing renewed growth in the area of our business that some thought was in the twilight.

As I have said many times, gallium arsenide is not going away. This ongoing diversification of our revenue base expand our business potential, reduces our dependency on any one product or application. Operationally, our recent investment in R&D and customer support are allowing us to provide consistently high-quality products for some of the -- today's most technical difficult applications, and give us -- our customers the technical support they need to make their product initiatives a success.

We're actively planning for growth and business expansion, and we're making good progress with the relocation of our gallium arsenide facility. As always, we'll remain firmly committed to continuing in our business and financial model.

This concludes my prepared comments. I will now turn the call back to Gary for our third quarter guidance. Gary?

Thank you, Morris. Once again, as Morris said, Q3 is shaping up nicely, reflecting an increase in each of our substrate categories. Because of this, we believe that total revenues in Q3 will be in the range of $26.5 million to $27.5 million.

As I mentioned earlier, we're seeing a meaningful positive impact by increased volume because our manufacturing overhead costs are spreading over more units, and this helps gross margin. We estimate that in Q3, this is expected to provide a 1% to 2% increase from our Q2 gross margin level.

In addition, we have worked diligently with 1 of our 3 consolidated joint venture companies to increase the utilization of a very important raw material that they supply for our manufacturing process. As a result, our cost savings from this effort in Q3 is expected to provide an additional 1% to 2% benefit to our gross margin over Q2. And this illustrates a tangible benefit derived from our supply chain strategy. As a result, in Q3, we are expecting to be in the range of $0.08 to $0.10 per share based on a share count of 39.8 million diluted common shares outstanding.

Okay. So this concludes our prepared comments. Morris and I will be glad to answer your questions now. Operator?

(Operator Instructions) And our first question comes from the line of Joe Maxa from Dougherty & Company.

Can you give us a little color on what maybe the breakout of the upside in the guidance? Is it -- I mean, you talked about all 3 segments growing, but is it indium phosphide more so than gallium arsenide? Or would you break that out pretty equally?

Well, they're all -- gallium arsenide is going up, LED gallium arsenide, semi-insulating gallium arsenide, indium phosphide's going up and germanium's going up. And it's a significant movement. It's -- they're all pretty much double-digit in growth, yes. So -- and indium phosphide is growing probably the most, if you keep gallium arsenide in 2 different buckets, between LED and wireless, indium phosphide will be the biggest dollar growth, yes. Yes. So it's reflective of the color Morris tried to give and it -- as we like to say, there's a story behind every number. So Morris gave you the -- some of the story of what's going on in the marketplace, and the beauty of numbers is they roll out and illustrate what was just said.

Understood. And the indium phosphide, it sounds like the key driver of that is the silicon photonics?

Yes, that's correct.

Okay. I wanted to ask a little bit more on the 3D sensing side, what's your -- what you're seeing out there and when -- and it sounds like you don't have an order yet, which is understandable, but I'm wondering what's your thought process is as far as preparing for that?

You're right, we don't have an order in-hand. We want one. We're working very diligently, we're -- very hard to work with our customers to remove any barriers in front of placing an order. We have worked with our customers, and the feedback so far is our material works very fine in the line. We have one of the lowest EPD requirement sufficiently fulfilling the requirement, but somehow, we just don't have the order yet. And I believe this is an early stage in development. So we believe the battle or the longer term that's a lot of opportunity for us, and we're not also limiting ourselves, although everybody knows there's one customer which is dominant. But there are many other smaller customers that are actively looking at the material. And we are really putting our money where our mouth is, we're doing the development in terms of technical reducing the EPD everybody likes, and we are hiring a sales and marketing person to look at what the customer requirement is -- really is, and we're working in the long term. We believe that we'll benefit from it, but we just don't know when yet.

And are you accelerating the plans to open the new facility in part to take advantage of the 3D sensing opportunity?

Yes, absolutely.

The fact that it has existing footage already constructed was a motivator in the selection of this site.

And when would you expect to be running that? Maybe not for 3D sensing, but just be up and running in that facility, assuming you close it relatively quickly?

We do expect it to start operation in stages. I think the first stage is probably doing some of the crystal growth part, which we do expect it to be completed in 3 to 4 months because of an existing building we already have. And wafer processing probably takes a little bit longer because it evolves with building out the clean room, and so we do expect samples from the clean room environment with the processing part by the end of the first quarter of 2018. But our -- one of the beauty of our business is such that we believe some of our customers will accept a 2-stage move. In other words, we will qualify the crystal growth side first and still doing the polishing on our Beijing facility. And later on when the wafer processing is completed, we will then qualify the wafer processing as well. This has advantage as you know, that our business is very, very, I shouldn't say robust, but it's very good. So I think that will give our customers a more safety net to get more capacity as their demand increases.

All right, just one last one for me. The Q3 guidance you gave, is that a number -- I mean, I know you're not giving forward guidance, but is that a reasonable number to think about going forward? Those levels? Or do you expect to see growth from that?

Well, we don't think it's a blip, and we would expect it to be sustainable and be added to, so absolutely, yes. Absolutely.

And all these, we're not counting the 3D sensing yet.

And our next question comes from the line of Edwin Mok from Needham & Company.

So first question just kind of circle back on indium phosphide, I guess 2 things I'm trying to ask. First is, Morris, you mentioned there are some small orders coming that kind of helped the quarter. Is that a trend that you continue to see going to this quarter and give you confidence about this double-digit growth this quarter? And are those small order coming all over the world or especially concentrated around China for the PON market? Or can you give some color on that?

Yes. First answer is it's actually coming from all over the world. We've seen some new activities in Europe. I mean, there are companies setting up, actually, new facilities setting up from Europe. We definitely see it very active in China. The United States, it's probably mostly existing customers, but we are getting into them. We are telling them we are really a good, reliable, high-volume, good quality material provider, and we're gaining market here. But I think the smaller application -- I mean, so most of these smaller accounts are probably not addressing the PONs market, I think the PONs market is probably a lot more -- a little bit more mature. I think some of these, we actually don't know what they are doing actually, but we are seeing they are taking hundreds of wafers instead of thousands, and they are not as consistent as some of the larger customers, they give monthly orders. But we definitely see -- we visited them, and they are very, very enthusiastic about the development on the indium phosphide businesses.

I see. Okay, that's helpful. I remember a few quarters when you guys talked about this indium phosphide being 30% of your sales, or something around that magnitude. Is that fair to think of the business at that size now?

Can you repeat that question?

I remember you guys talk about indium phosphide being around 30% of your sales. Something in that magnitude, right? Is it fair to assume we're on that range?

Yes. It's still close to that range. The one thing that's made a little bit of difference is the other 2 products line sets have grown also. So I mean, when we were saying around 30%, we were seeing good growth in indium phosphide but relatively flat in the other 2 product lines. Now the market has been -- the demand for all the products has been very good. So it's a horse race.

I see. For the third quarter, you guys talked about the PON market kind of improved a little modest [throughout] 2016, right? For the third quarter, are you still counting on the PON market to remain at that (inaudible)?

Yes. But as we stated in our prepared comments, the PONs market level is not as high as some of the 2015 numbers. It has a appreciable drop from the 2015 level. But as you recall, if we say 2015 was 100, okay? By the second half of 2016, it went down to 20. Now it's bouncing back to something like 65, okay? We're counting on this PONs market to be around 60 to 70. It has not returned to 100 yet because -- well, I guess the 2015 level has quite a bit of this building inventory and there's a shortage in material. But I do believe the PONs market will recover and also, I believe that potentially the new PON, which is the 10G PON, will then accelerate the adoption of the market or the demand for this PON market again.

Yes, definitely. I'm -- yes, nice broadening of demand -- end market demand, which is great beyond just the PON market. Can I ask you guys about a question around seasonality? I think if I go back and look at historically, 4Q, your substrate business also have raw material typically is down sequentially, maybe as much as double-digit, right? This year, so it seems that the market is really, really strong right now. Do you expect -- do you think that you might be able to outperform that seasonal pullback in the fourth quarter while you suffer a smaller pullback in the fourth quarter?

Well, we don't have numbers yet from our field guys on Q4. However, we have color, and in an anecdotal sense, Q4 is looking good. We think it's quite possible it won't be down. Or if it's down, it will be just down a tad, but it could also be up a tad. So we don't know. But what you're speaking to, Edwin, is the fact that yes, if you look back historically, Q4 has generally been down from Q3 and from seasonality reasons, but those reasons still exist. However, the strength and the sort of dynamism in the marketplace right now is going to offset that.

Yes. And also Q4, we have 2 events you have to consider. One is the year-end. Most customer, if they are not that busy, they want to reduce their inventory. And second thing is perhaps particular to AXT is we have all manufacturing in China, where October 1 is their holidays, and they usually take 10 days off. So with these 2 in mind, it's difficult for first quarter to grow. But you're right, the demand from what we see now is very strong.

So growth is not out of the question, it's just not clear yet.

That's extremely helpful color. Gary, I have a question around raw material. You mentioned that there is a write-down that impacted gross margin by 1 point. And I think the raw material business based on your file report suggest it's still not profitable. Is there what you kind of think about how -- or if this cost saving that you talked about is going to start bringing that back to be a profitable business? Or is there more work going to be done business for that business -- for those [JV businesses]?

Well, the 7 companies that we use the equity method of accounting for, they finally dropped below a significant hit. We were hopeful that their charge to our income statement in Q2 would be less than negative $200,000. And we were accurate on that, it came in at minus $188,000. So that's a huge step forward from the previous 3 or 4 quarters, especially Q1, which was over $900,000, because we had a write-down that quarter. So those 7 companies, we're hoping that they can get to breakeven in aggregate by the end of this year, or maybe just that the charge or the loss that we see continues to go downhill. Then there's the 3 consolidated subsidiaries, 2 of those companies are actually pretty strong, and they are beneficial for us. And then the one that's the raw gallium company is still struggling. So sort of what's happening is the 2 healthy ones are already pretty much offsetting the raw gallium company. So the 10 joint ventures in aggregate for the -- starting in 2016 and up until now, have been problematic. But if you go back historically, they've been accretive and beneficial. And I'm very pleased with the identification, which we credit Morris for in this case, but who really worked with this particular subsidiary to increase their output so that we can buy this important raw material at their manufacturing cost because of the way the accounting of this works. So -- and it's beneficial for us.

And our next question comes from the line of Hamed Khorsand from BWS Financial.

So first off I wanted to start off with -- was last quarter, you were talking about broader customer base with indium phosphide. This quarter, you're talking about the same. But the unit volumes are increasing. Now, did you see a return of orders from your [larger customers]? And are you seeing that going into Q3? Or are you depending on the smaller customers to make up the difference?

Actually, I think it's from both. The larger customers, they are increasing the demand, the volume, but not adding one. I mean, some of them than the other. But we do notice the other very distinct phenomenon is that there's a lot of smaller orders, and they do contribute to the revenue. But that -- those are less predictable, because they are -- they don't give us a clear path of how many they want to order next quarter or next month. So in our projection, we're probably mainly looking for the existing long-term customers, they are giving us their color or guidance on how much their demand will go for the next quarter or quarter after that.

Okay. And then on the pricing realm, given that you're seeing so much demand, why is it that you still don't have pricing power?

Because we're in the semiconductor business.

Well, I guess, specifically, you're saying, why don't we raise prices? I guess, to a certain degree, we can, but customers, obviously, is very reluctant, and we haven't got to that stage yet, okay? And so the first order of benefit to us is that we can benefit from this spread out our cost -- fixed cost, so we'll gain from that. We're already thankful and good. And then, as our capacity start to get constrained and we're getting way over our expected revenue, because don't forget, we have to deliver this set revenue for our Wall Street friends. So we're very cognizant of the fact that we need to get orders. But once we got that, and as we go over that, then, we could be selective. Some of the lower-margin ones, we would drop -- start to drop, and we'll not pursue those lower-end market. That, in a way, is a increasing price. And lastly, I would also comment, look, we had been saying there's other way to increase prices. That with some of these customers, they start to migrate, for instance, the indium phosphide, they migrate from 2-inch to 3-inch. They put a little more stringent specification onto it. And every time they do all these, then the price is different. So we can gain margin that way. I mean, other matter is 3D sensing. Look, this is a brand-new market. And although we are not happy to report we don't have an order, but this market, once we get in, should give us healthy margin, because they are very difficult to meet the specifications. So that is another way to improve your price and gross margins.

Yes, I think we can improve margins by features and benefits, diameter sizes and specifications. But we're both over age 50, and in my entire career, I've never been -- to be able to just raise the prices. It just -- so I wouldn't bake any of that into anybody's forecast because I think it doesn't happen. So...

Back in 2000, I increased the 6-inch price on [450] to [480].

There you go, there's an exception to every rule. So...

Yes. And then 2 quarters later, it's us.

Last question is did you add any crystal growth capacity in the quarter? And do you plan on doing it in Q3?

Yes. And one of the strengths of AXT is that we have been already saying to our investors as well as our customers is that we know VGF from the inside out. So we are very dynamic in terms of adding capacity for crystal growth because we do the assembly of the crystal growth equipment ourselves.

And our next question comes from the line of Tom Sepenzis from Northland Capital Markets.

Just wondering if there was any risk at all in the current inventory? I know you just took a write-down in the June quarter, but what kind of risk are we looking at in the current inventory that might get written down in the next 6 to 12 months? Or is there any?

Of course, it's very difficult to predict the future, you know how the drill is. If we have certain inventory, especially the gallium manufacturer, let's say 10 tons of this inventory, and it carries at this current market price, and I believe right now is CNY 820 or CNY 830, and it writes to that level. And if next quarter it dropped down to CNY 700, you have to write it down. But so far, we believe that it's at a very, very low value already. So I believe anything, it could go up, but again, we don't want to predict the future.

Yes, so we wouldn't expect any LCM, lower cost of market, that would touch the substrate inventory. This is specifically a problem because of the raw materials businesses, and it's uniquely focused -- almost singularly focused around raw gallium. Even our purified gallium company, we don't have write-down problems, it's the raw gallium in China that has been problematic. So it's -- it's not broad in scope, it's narrow, and we can identify it, but we can't predict. It's not likely it will go down further, but we can't predict.

Great. And then in terms of operating expenses, obviously, you're not expecting a huge shift here in Q3, but just directionally looking out a few quarters with the higher revenue, do you plan to start spending more on R&D to get back to levels that you had maybe in 2016 or earlier for anything or should we expect that they'll stay pretty low here?

I think they'll tick up some, but probably not because of reinvesting in R&D, which I'll comment on in a second. But we've added 2 new sales guys, and we're still thinking of adding a couple more professional people, so there's some labor there. As a side note, one of the reasons the R&D number is lower is because we had some things that were charged to R&D because they were truly R&D, but then they went into production. So now, the effort that goes towards those -- that product line, which happens to be in one of the subsidiaries, is going through cost of goods sold. So it's really a change in the business model that it was a development program, so it should be charged by GAAP rules to R&D, but when it went into production, now if you're fine-tuning it, you don't -- it's difficult to capture just the development cost, so it's all in the cost of goods sold.

When did that shift occur? Was that in the March quarter?

Yes, it was.

And our next question comes from the line of Richard Shannon from Craig-Hallum.

I guess, I just got a few left here. First one is on gross margins. Gary, you mentioned something about a benefit in the third quarter coming from a new raw material or something like that of 100 to 200 basis points. I wonder if you can detail that more about where that's coming from? And as well, is that as a sustaining or a onetime benefit?

The -- it's an important ingredient into one of the substrates for single crystal growth. And Morris' idea was to push this company to produce more because we do buy also from an independent third-party. But because of the opportunity to buy more from one of our subsidiary companies, we would then have that cost of goods sold via their manufacturing cost because of how the elimination of rules work for GAAP accounting. So it's a strategy that we can build on. It's too soon to say that we can repeat it over and over and over, because we need to see what their capacity levels are. And -- but Morris and I are both going to China in August, so it's a topic that we'll be discussing with this company, yes.

Okay. Fair enough. A question on OpEx. As you make this facility move, it sounds like you'll still finalize this plan shortly. Are there any duplicative expenses here as you're running both facilities at the same time that are meaningful? And if, so how long would they last?

Yes. I think we probably would operate in our 2 facilities for maybe 1 year, 1.5 years, would you say?

Yes.

But however, I don't think the expenses that we will incur more is the extra equipment we're going to buy in anticipation for the larger volume as well as for the ease of moving to a new facility. We already told our purchasing people to buy 20% more material that you should see the depreciation coming online once we got those installed, they'll start to produce the material. But if our business were to pick up, then we should be able to sort of absorb that as our demand goes higher in 3D sensing or some of the silicon -- indium phosphide business were to increase in volume. As far as manpower is concerned, I think for the transitional period of time, we probably should have a little bit more expenses -- the added expenses or manpower such as cafeteria workers from both sides and direct labor from the line.

So it's a very important and intelligent question that -- and we have looked at it in various ways. And you kind of have to break it into categories. So, obviously, the single biggest dollar expenditure is for the land and the buildings. And then that gets depreciated. But the good news is that that's depreciated over 27.5 years. So it's -- because of the useful life span that we're allowed to use, even though it's the largest single dollar category, it doesn't really have a super big impact over time. And then, as we vacate in Beijing, we will furlough some of that space, and then eventually, we hope to monetize it. So once we do the furlough step, we won't be depreciating anymore because it's not in use. So that's the facility. Then you have the equipment. We are adding equipment and that's going to increase depreciation expense. But that's going to happen mostly because the volume is going up. It's not happening because of 2 sites. And so, in that case then, we wouldn't do it unless we thought that the business needs to have it done and could afford to pay for it. So -- and thirdly, then you have what I would call manpower and administrative costs, and that's clearly going to be an increase. But the labor rate in the new location, first of all, it's in China, of course, but it's even less than Beijing, because Beijing will be the highest labor rate, and we're in a different city now for the new location, which we haven't announced what it is yet. But -- so that labor rate is in China, and it's not in Beijing, so it's not really going to be too noticeable to an outside observer in the financials. We have a little more hotel expense, driving back and forth. So there -- you can't say it's not going to be more expensive, but I think you can say in the scope of the financial statement, it will be barely noticeable, if at all.

Okay, that is helpful. Another question, I guess more for Morris. You're pretty clear that you don't currently have an order for 3D sensing, you're working hard towards it. Do you want to take a guess as to when you might see one? Could it be -- to have an order by the end of this year? Or is that still hard to tell? Or can you give us a sense of what that visibility and pipeline looks like to get that first order?

Yes. We are actively working with our customer and customer's customer and customer's, hopefully, customer's customer's customer, and when we're moving some of this requirement that they wanted from us. And so far as we know, the feedback is we have very good material property. They love our EPD. We have some other issues with the customer, perhaps including moving up the site. And -- but I don't know whether that's a absolute no, because we understand our material -- merit of our material property may get us in, but we just don't know yet. We do have some secondary customers we are working with, although they are taking very small volume right now in the hundreds of wafers we're sampling, but we are also hoping that they can be developed into substantial customers. So to answer your question, we are keeping in touch with the customer, we are working diligently and hopefully, we can remove any entry barriers to get into the first round. But we don't know when that's going to be.

Okay, fair enough. That's helpful to understand that. My last question, probably for you Morris, germanium is doing pretty well here and you talked about a favorable satellite cycle. I guess my question is what's your experience as to how long these satellite cycles can last? And do you see this cycle being any different than what you've seen in the past?

Yes. But of course, this cycle could be different. I mean, there's a lot of interest in building -- that the last cycle was the GPS. And now, China wants to build their own (inaudible), and that's a big driver we know. And that's like the U.S. GPS. and I know our substrate is also seeing the Russian satellite as well as European satellite systems, okay? And looking forward, I do believe there's other potential that is one left, and that program actually is huge. It could be 10x this. But we're not counting on that. So I think we are both [feeling some] for the germanium business because we see the germanium pool from the satellite business is fairly strong. But what do we know? Maybe a quarter, 2 quarters, or maybe 3. But beyond that, it's very difficult to predict.

And I'm showing no further questions at this time. I will [turn the] call back over to Dr. Morris Young, Chief Executive Officer, for closing remarks.

Okay. Thank you for participating in our conference call. And during the third quarter, we will be participating in the Second Annual BWS Financial Summer (inaudible) Roadshow Extravaganza in New York and the 2017 Dougherty & Company Institutional Investor Conference in Minneapolis. We hope to see many of you there. As always, please feel free to contact me or Gary Fischer directly if you would like to meet with us. We look forward to speaking with you in the near future.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program, and you may now disconnect. Everyone, have a great day.