AXT Inc (AXTI) 2017 Q4 法說會逐字稿

Good afternoon, everyone, and welcome to the AXT's Fourth Quarter and Fiscal Year 2017 Financial Conference Call. Leading the call today is Dr. Morris Young, Chief Executive Officer; and Gary Fischer, Chief Financial Officer. My name is Amanda, and I will be your coordinator today. (Operator Instructions) And as a reminder, this conference is being recorded.

I would now like to turn the call over to Leslie Green, Investor Relations of AXT.

Thank you, Amanda, and good afternoon, everyone. Before we begin, 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, improve efficiency, increase orders in succeeding quarters, improve our competitive position in the market, our schedule and timeliness regarding the relocation of our facility, 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 some statements that deal with future events are based on management's current expectations and are subject to risks and uncertainties that could cause actual results or events to differ materially.

These uncertainties and risks include, but are not limited to, overall conditions in the market in which the company competes, the global financial conditions and uncertainties, increased environmental regulations in China, 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 expectation. This conference call will be available on our website at axt.com through February 21, 2019.

Also, before we begin, I want to note that shortly following the close of the market today, we issued a press release reporting our financial results for the fourth quarter and fiscal year 2017. This information is available on the Investor Relations portion of our website, at axt.com.

I would now like to turn the call over to Gary Fischer, for a review of our fourth quarter and fiscal year results. Gary?

Thank you, Leslie, and good afternoon to everyone. Total revenue for the fourth quarter of 2017 was $26.3 million, which is within the range of our guidance. This compares with $28.2 million in the third quarter of 2017 and $20.3 million for the fourth quarter of 2016. This is a 30% increase in quarterly revenue as compared with the fourth quarter of 2016.

Of our total revenue, substrates sales were $20.5 million compared with $22.4 million in the prior quarter. Revenue from our raw material joint ventures was $5.8 million in Q4. Q3 was also $5.8 million. In the fourth quarter of 2017, revenue from North America was 9%, Asia Pacific was 67% and Europe was 24%.

Two customers generated more than 10% of revenue in Q4. And the top 5 customers generated approximately 36% of total revenue, reflecting again our diversification of both products and customers.

Gross margin in the fourth quarter was 37.2% compared with 39.5% in the prior quarter. Total operating expenses were $6.1 million in the fourth quarter of 2017 compared with $5.9 million in the third quarter. Total operating expenses in Q4 included onetime expenses of approximately $400,000, primarily related to severance payments.

Looking ahead with the transition of our new factory now well underway, we are gaining a bit more clarity on our headcount requirements and other relocation-related expenses, including training and travel. As such, we are anticipating that our quarterly OpEx in 2018 will remain at approximately this Q4 level.

Total stock compensation expense for the fourth quarter of 2017 was $444,000. Operating profit for the fourth quarter was $3.7 million compared with $5.2 million in the previous quarter. Interest and other income for the fourth quarter was a net charge of 330k. This net number consists of 4 categories. Number one is the interest earned of 127k. Number two is equity accounting on our unconsolidated joint venture companies, which is a loss of 307k. The third was foreign exchange loss of 188k, and the fourth was other items that equaled a gain of 38k. The tax provision for the fourth quarter was 131k compared with 181k in Q3.

For Q4 of 2017, we had a net profit of $3.1 million or $0.08 per share. By comparison, we had a net profit of $4.4 million or $0.11 per share in the third quarter and $2.2 million in Q4 2016. This is a 42% increase over Q4 of 2016. The diluted share count in Q4 was 40.45 million shares.

On to some balance sheet comments. Cash and cash equivalents and investments closed at $77.0 million as of December 31. September 30 was $78.3 million, so cash was down $1.3 million. Depreciation and amortization in the fourth quarter was $1.1 million and capital expenditures were $4.7 million.

Accounts receivable net of reserves were $22.3 million at December 31 compared with $20.9 million at September 30. Net inventory at December 31 was $45.8 million compared with $40.8 million in inventory as of September 30. Ending inventory consisted of approximately 51% in raw materials, 44% in work-in-progress, and only 5% in finished goods. Both WIP and raw materials increased, and this is intentional as we see raw material prices increasing and as we build inventory during the relocation.

This concludes the discussion of our quarterly financial results. Let me now briefly highlight the fiscal year. For the fiscal year 2017, which is our calendar year, revenue was $98.7 million, up from $81.3 million in fiscal year 2016. This is a 21% increase over 2016.

Gross margin for fiscal year 2017 improved nicely to 34.9% of revenue, up from 32.4% of revenue for fiscal year 2016. Our profitability in fiscal year 2017 also increased meaningfully. We achieved net income of $10.1 million or $0.26 per share, compared with $5.6 million or $0.17 per share for fiscal year 2016. This is an 80% increase in net profit as compared to fiscal year 2016.

Okay. This concludes the financial review. I will now turn it over to Dr. Morris Young for a review of our business.

Thank you, Gary, and good afternoon, everybody. Our Q4 results capped off a strong year of execution for AXT. We delivered healthy growth in 2017, including record indium phosphide revenue and solid performance across our product portfolio.

In addition, we achieved improvement across our key financial metrics, driving an 80% increase in our net profitability over the prior year. We also made good progress on the relocation of our gallium arsenide manufacturing facility, and we are on track with our stated goal to provide qualification wafers in Q1.

As we move into 2018, we are excited to see a resurgence of demand for compound semiconductor substrates in new applications across our product portfolio that has the potential to reshape the technology landscape over the coming decades.

As such, we continue to invest in the advancement of our products, our customer support capability and believe that we are positioning the company well for continued growth and new opportunities in 2018 and beyond.

For example, indium phosphide was a strong driver for our business in 2017 and is expected to be a significant contributor in 2018. This is largely due to the strength of our technology and market position as well as growth and demand from both established and emerging applications.

The fast-growing applications currently for AXT is data center connectivity, driven by indium phosphide-based silicon photonics. These technologies solves the key data center networking challenges, including latency, bandwidth, power dissipation and cyclical integrity.

Leading industry analysts believe that silicon photonics is reaching a tipping point of widespread adoption. As evidenced, optical transceivers and other component based on this technology are now in volume production by a growing ecosystem of providers.

Hyperscale data centers such as Facebook, Amazon and Azure have also publicly announced its adoption. In addition, data center expansion in other parts of the world such as China, Japan and U.K. could provide additional demand in 2018 and beyond.

In particular, China now has more than 700 million Internet users and companies such as Tencent and Baidu continue to build hyperscale data centers to accommodate growing Internet traffic.

During 2017, data center connectivity surpassed passive optical networks as the largest driver of our indium phosphide revenue. Prior to this, the EPON and GPON markets were key contributors for indium phosphide demand in applications such as fiber-to-the-home and fiber-to-the-office.

Throughout the past year, a slowdown in purchasing from carriers as well as the inventory buildup in the supply chain constrained market growth. However, we have been encouraged by recent indications from our customer, such as conditions are improving in key regions such as China.

This strengthening of demand may drive improvement in the power market in 2018 and give AXT another avenue for growth. We'll be watching this closely as AXT supplies into a number of regions and customers worldwide, and is well-positioned to benefit as market improves.

And finally, a third area of focus for indium phosphide is telecommunications, where the current infrastructure upgrade cycles and preparation for 5G are providing opportunities in short haul, long haul and metro deployments.

Though still a relatively small portion of our indium phosphide revenue, we believe that this application could be a contributor to our revenue growth for years to come.

In total, we believe that we are in early stage of a multiyear opportunity for our indium phosphide business. This is a highly-specialized material in which AXT has made considerable market progress, both in terms of outstanding technical properties of our material as well as our customer traction.

As a result, we're currently increasing our indium phosphide capacity in a meaningful way in order to meet the expected increasing demand for our product in 2018 and beyond.

Now turning to gallium arsenide. Over the last 2.5 decades, the application that have driven our gallium arsenide business have certainly evolved. And yet, the material is as relevant today to the unfolding technology landscape as it was when it was first commercialized, our VGF gallium arsenide substrate in the late 1980s.

Application such as 3D sensing, augmented and virtual reality, 5G wireless, lidar for autonomous cars, retinal recognition, drone-based Internet connectivity and many others are emerging and will require the performance characteristics and adaptability that gallium arsenide uniquely offers.

Further, the stringent technical specification of these high-end applications continue to severely limit the number of company that can produce substrates and enough volume to meet global demand.

As a pioneer in VGF gallium arsenide substrates, AXT continue to be bring innovation and deep expertise in materials science to our markets, helping our customers to unlock the potential of new technology by overcoming technical barriers.

Our work in gallium arsenide thin-film solar cells for the automotive market is a perfect example. Working with our customers, AXT engineers developed substrate that feature unique shape and specialized properties for this application.

Audi and BMW are now using gallium arsenide-based solar panels on certain new models to provide power to the vehicle's climate control system fan without draining the battery.

In addition, Audi and Alta Devices are also integrating solar cells into panoramic glass roofs of Audi models to generate solar energy that extend the range of Audi electric vehicles.

We are also excited about the potential of 3D sensing technology for our business. While AXT is not currently participating in the initial development -- deployment of 3D sensing technologies for Apple iPhone X, we believe 3D sensing technology present a significant long-term opportunity for AXT.

In addition to Apple, we are hearing from a number of sources that 3D sensing capabilities are on the product roadmaps of several other device manufacturers. And we believe we will see more widespread adoption by 2019.

The strict technical requirements of 3D sensing provide a strong barrier-to-entry to new lower-end players. Today, only 3 competitors, including AXT, are able to provide low EPD substrate in significant volume for production there. We understand that our product meet the technical qualifications and perform well for these applications.

With the relocation of our gallium arsenide manufacturing facility showing good progress, we are on track to have the first 3D sensing substrates from our new site for customer qualification this quarter.

Once submitted, delays of qualification trial process will depend on the sense of urgency within the supply chain. We believe that we could see a modest amount of incremental opportunity from this application in 2018, but with more meaningful contribution in 2019.

It is, however, very important to note that the mobile device application is likely just the beginning of many applications that we're using 3D sensing technology in the coming years. Some analysts have stated that by the year 2022, the volume of gallium arsenide use for 3D sensing in autonomous vehicles will be equal to or greater than what's used in mobile phones. We are hearing that development work is already underway with these applications. We'll continue to monitor this and other new applications closely, and believe we are well-positioned to participate in their adoption.

In total, we are enjoying stable revenue from our traditional markets for gallium arsenide, including power amplifiers for mobile phones as well as LED lightings, signage and display.

In addition, we're encouraged to see our substrate selling into a wide variety of nascent applications that have the potential to become meaningful to our business over the coming years, underscoring the long-term nature of our product portfolio.

Now turning to germanium substrate. This segment of our business was again strong in Q4, as the satellite industry continued its potential positive trend.

We typically see a fairly constant demand for germanium substrate used in satellite solar cells. But in 2017, our germanium substrate revenue grew approximately 30%. Though we are not expecting this level of growth in 2018, but we do believe that positive market conditions are likely to provide continued opportunity for growth over the coming quarters.

Finally, with regard to our partially owned raw material companies in China, recently, we have begun to see an increasing raw material price, particularly gallium, largely the result of overall improvement in commodity pricing.

Increase today remained modest. And on a whole, have served to bring suppliers to close to breakeven levels, following a very significant decline in pricing over the last 2 years.

We do believe that as the demand for compound semiconductor substrate continue to increase, our holdings will provide important leverage for our operations as well as opportunity with positive contribution to our overall financial model.

Now before I close, I want to give you an update on the relocation of our gallium arsenide manufacturing to our new facilities in the city of Dingxing, China.

With the continued solid execution of our team, we are on schedule with our plans and pleased with our progress to date. This new technically advanced facility gives us the opportunity to plan our business for our next stage of growth and to support long-term requirements of our customers.

We're excited for this facility to become a showcase for our capability for customers and investors alike. As we have discussed, we are executing a staged relocation of gallium arsenide throughout 2018, which means that we will produce substrates from both our current site in Beijing and the new Dingxing site for a period of time, and then gradually increase production volume at the new site.

In closing, this year, we will celebrate 32 years in business, including 20 years as a public company. Over the time, AXT has grown and evolved significantly.

Our business focus, operational discipline developed over time, have allowed us to hold our grounds against much larger competitors to weather significant technology transitions in our industry and build a sustainable business model of growth and profitability.

But it is the enduring entrepreneurial spirit of AXT that has driven us to embrace new applications for our technology and our product, to recognize future growth opportunities and develop the core competencies to supporting and to identify the strategy early of vertically integrating our supply chain.

I am very grateful to the entire team at AXT for their talent, their dedication and their tireless effort on behalf of our company, our customers and our shareholders. We are very proud of the unique company we have built together, and we are excited about our future.

We see great upside potential among applications across the product portfolio that we have already qualified into and supplied. In addition, new applications and opportunities from incremental applications such as 3D sensing hold promises for 2018 and beyond. This is an exciting time for our business, and we look forward to reporting to you on our progress.

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

Thank you, Morris. As we discussed, we're excited to see the upside potential across our portfolio, and we believe we are well-positioned in many applications that will drive our business growth in 2018 and beyond.

Coming off of a solid Q4, we expect to see revenue in Q1 of between $26.0 million to $27.0 million. If we take the midpoint of our Q1 guidance, this would be a 29% increase over Q1 of 2017. Given the expected Q1 OpEx levels that I discussed earlier, we believe our profit per share in Q1 will be in the range of $0.07 to $0.09, based on 40.44 million diluted common shares outstanding.

This concludes our prepared comments. But Morris and I would be glad to answer your questions now. Operator?

(Operator Instructions) Our first question comes from the line of Hamed Khorsand of BWS.

So I just want to discuss this. As far as this increase in OpEx, is this all related to the new Beijing facility? And is that what you're talking about as far as growth initiatives go, is you're ramping this facility up right now?

Well, Hamed, this is Gary. So in the Q4 of 2017, no, it's not all related to the relocation because there's about $400,000 of severance-related expenses. However, as we look on the horizon going forward, with -- setting aside the onetime things I just mentioned, we do think our OpEx will be around the $6 million level, and mostly that increase is related to the expansion of the buildings in the -- the building that's going on with new employees, with training and things like that.

Okay. Because I don't want to -- that's about $1.1 million more than where you guys were at the beginning of 2017.

I agree. I understand. So I think what's happened over the last 12 months is the market opportunities that Morris described had become more and more tangible, more and more understandable. And so we think that it's important to invest throughout the income statement, but that's going to mean building up a little bit more on OpEx as well.

Okay. And then on the last call, I mean, it sounded like you guys were still questioning the PON market as far as it contributing to any demand. What are you seeing as far as the end markets there, especially with China and the Lunar New Year in Q1?

Yes. We think the PON market is having a good recovery. Although we like to wait and see how long it will continue, but we think the recovery is fairly solid as of now.

Okay. And then my last question, just you usually talk about Q1 being sequentially up from Q4, and this time, your guidance suggests it's going to be flat. What's driving that? What are you seeing that -- your questioning that there would be a sequential increase in Q1?

Well, I guess, we -- I never thought of it. But we do have our own sort of unique way of coming up with this guidance. We take it from our sales representative as well as looking at our booking as well as projecting what the future growth is going to be like and they come out to be exactly like you (inaudible). So -- but I think, specifically, maybe, I think indium phosphide sort of continue the growth trend, germanium is a little bit better and the gallium arsenide is a little bit weaker. I think that's probably, overall, it's a flat.

Yes. And this is Gary. If you look at the historical trends, you're correct. We usually come down in Q4 from Q3, so there's some seasonality there. Q1, sometimes it's close to Q4, sometimes it's a little bit above Q4. But it's still a solid quarter. We have positive trends. We think things are okay. So -- and we don't give guidance beyond Q1, but I think you can tell from the tone of the content, we're very optimistic about the future quarters. So I think it's okay. I think it's a solid quarter. And you have to make your own judgment about it. But we're not -- we don't think there's anything wrong or anything sort of going on that we don't understand. This is just how the numbers came out, and you know us, we just sort of say it like it is.

Our next question comes from the line of Jon Fisher of Dougherty & Company.

Just going through the segments that just the way you talked about, germanium and InP and in gallium arsenide kind of came across as probably more neutral on your commentary than the other 2. And I guess, I would push a little bit more as a follow-up to the prior question. Kind of LED and OSRAM had been a really strong source of demand through 2017. Has there been any sort of meaningful slowdown in demand from OSRAM or in lighting or in LED? Or what are you seeing in gallium arsenide that kind of makes your tone, when you talk about that substrate, maybe a little more neutral than the more bullish tone you had for InP and germanium?

Well, first of all, we don't really comment on specific customers, that can get us into a lot of trouble. So we don't talk about OSRAM or anyone else. I think gallium arsenide, yes, we had pretty good growth in Q3 and Q4, right? And I think it's holding its own right now. But maybe one specific area we see a little bit weakness is we hear some customers because the 3D sensing market is so strong, people are giving up on the pHEMT business in preference to do the VCSEL business, such that it impacted our demand for the pHEMT gallium arsenide substrates. So -- but we also have a variety of applications. So it's a cumulation of all these customers giving us these results or predictions. But we don't see any particular area which is weakening significantly going into the future.

Okay. And when you compare sequential performance, Q4 '16 and Q4 '17, the deceleration from Q3 was greater this year than it was in '16. When you look at the 3 primary substrates, were any of the substrates up sequentially on a quarterly basis? Or were all 3 of them are down? And was there any one in particular that was down most significantly?

There's no super big trend in any direction. Since we were down in aggregate, $1.8 million, everything was generally down except raw materials. But again...

Indium phosphide was up.

The numbers rolled out, we think, very strongly. Our outlook for gallium arsenide is very positive. And again, this is the way the numbers came, and we just put them out there. But our expectations and enthusiasm are not diminished for any of these products. It's very solid.

Okay. And then from a CapEx standpoint, kind of what would be the outlook for CapEx in 2018? A step up from 2017 levels or kind of in line with 2017 levels? Or would there be a step down from '17 levels?

I think by the end of 2018, it will be probably more than 2017. I don't know how much more yet though, because both years have the building stuff going on. So -- and we're not done with that yet. So -- but generally, what we say for CapEx is for the equipment side of it, it runs $4 million to $6 million a year. In the last couple of years, that's probably been on the low side of the $4 million to $6 million range. Right now we're doing a little bit more in equipment as more -- for indium phosphide, we're buying more furnaces. So both on the equipment side and then on the facilitization side, it would be up a little bit.

Don't forget, we also said in order for us to build a new facility, we're going to build -- we're going to purchase new equipment in the order of 25% to 30% more new equipment to be put into the new facility and get them running before we start to move the old equipment into the new one. So that's why you see -- you will see a spike in the equipment purchases in the last quarter. I think there was $4.5 million, (inaudible) which is $1 million, right. So yes, it's going to be more...

Some of that is building, some of that's equipment. But in general, it's meeting our expectations. So we are on track with what we expect.

And then the last question, just from a geographic standpoint, when you look at revenues, are you comfortable with the performance of all 3 of the main geographies in Q4, and kind of whatever your inferred outlook is for Q1? Or is there any weakness in any one of the particular main geographies?

Well, the geographic trend for our product is pretty dominated by the Asia-Pacific region, and that's not going to change. Does it mean that we don't think we can improve in any of these 3 regions? No. I think we could improve in all 3 of them, North America, Europe and Asia Pacific. We’re not disappointed in any of the regions, and this is pretty much how it's been. We have made some changes in Europe. We have our sales representative for over 20 years retired, and so we hired a direct employee whose off to a good start, and we hope that he can do more business development in the European front.

Our next question comes from the line of Dave Kang of B. Riley FBR.

This is actually Lee Krowl filling in for Dave. First question, appreciate the visibility on the gallium arsenide potential qualification timing. Just kind of curious in terms of customer reception, it sounds capacity for 3D sensing is actually tied across the universe of gallium arsenide suppliers. So I'm just curious if the small revenue you expect in 2018 is perhaps conservative just given the fact that there is some new demand out there from other OEMs as well as the core adding new products? So just kind of curious when the customers get qualified and maybe if there's a potential ramp for something higher than just marginal revenue in 2018?

Well, it could happen. We were told that -- well, we have multiple fronts, by the way, not only that particular company. Other customers are running qualification. But since we -- they're not running production right now, it's difficult for us to put an exact number on it. But overall, we are excited about it. But when we build our business model, our budget, we don't put that much of a big expectation out of it because it's something which -- once we qualify, you can run big, you can end small. It could take a quarter or 2 before they become comfortable. And as I also said, this is not only one customer. We have multiple customers. But one of them probably is the biggest that everybody knows, but there are other smaller customers running qualification as well.

Got it. And then just on the indium phosphide front, you guys mentioned that you are doing capacity expansion. Just kind of curious what end markets would drive that capacity expansion. I would assume you have some sort of visibility to invest. Curious if that's driven by an improvement in GPON? Or if that's all primarily datacom?

I think it's all of the above.

All of the above.

All of the above. The data center business is showing very strong prediction for growth. GPON, EPON is coming up as well.

The whole silicon photonics advancement, which is widely understood now, I think, and expected to continue for multiple years, it's going to help us a lot and probably more steady than the EPON, GPON market, but both are strong. That's what we're seeing. So.

Okay. And then just last question on gross margin. You guys mentioned better pricing and you built some inventory as a reflection of better pricing. Just kind of question on 2018 gross margin levels, assuming pricing stays level or improves more for your outlook, how does that drive gross margin? And then, just given the growth in indium phosphide, how does that also fit into your kind of outlook year-over-year for gross margin?

Well, the gross margin is the most volatile line on our income statement, we say that repeatedly to your community. So if you look at the 4 quarters of 2017, first 2 quarters, it was like well 30.5%, in that range. Second 2 quarters, it's over 35%, 39% and then 37%. So we continue to say that the safest way to probably, for anyone running a model, would be maybe somewhere in the mid-30s. Our goals are greater than that. But I think it's more -- it's probably appropriate to be realistic and conservative. So that's kind of what we would guide. Let me clarify one thing. As the raw material prices increase, which Morris mentioned, we have a hedging sort of business model strategy because it will actually bring our substrate gross margin down a little bit because we may have to pay more for the cost of raw materials in the open market. But it will help our 10 joint venture companies, many of them because they'll be able to charge a higher ASP. So we have 3 companies that we consolidate and then 7 down below that we use the equity method of. And you may not have watched it closely, but I know David has, David Kang has and other analysts. But that line down below has been negative for all of '16 and all of '17. And we've been saying that we're hoping it will come back to be neutral. And we still think that's a very realistic hope. So raw material prices are increasing and that's going to help those companies improve.

Our next question comes from the line of René Saner of Octavian.

Just to follow-up on the raw material price situation. I mean, can you talk to the drivers? I mean, the price for gallium, obviously, seems sliding for many years, and now towards beginning of December, we saw price feed rise but not only in gallium, also in indium and germanium. So you're mostly active in gallium. And looking back to the financials a couple of years ago, you had very significant contribution from both the direct consolidated investments well as your equity consolidation investments launched the gallium price, which is above $150, so we are close to $200 now. So maybe you could talk us through the drivers between the price increases? And what it could mean for the numbers in 2018? And what you embedded in your Q1 guidance in terms of raw material price expectations?

Okay. Well, this is Gary. So what would it mean? Well, what I've been telling you guys for about 6 months is that my goal and Morris' goal is to get the raw material companies back to at least breakeven in aggregate. And for the 7 companies that we use the equity method for, for Q4, there was still a charge of $307,000. So that's down nicely from the beginning of the year, which was $900,000 charge. So we're going in the right direction. I think for people running models, it's probably reasonable to put that for the year of '18, 0, maybe there's upside, but we tend to be more conservative, so we don't want to people to get too far ahead in terms of modeling. But the overall impact, in aggregate, in the history of the company, the partially owned companies, there are 10 of them, that we have partial ownership of, they have been accretive and it's been helpful not only financially, but it's given us visibility into the supply chain. It's given us shorter lead time, better control, especially if there's shortages. So then for 2016 and '17, it's been a little bit tough. But now we think that, that toughness, we hope, is behind us and will at least be into sort of a neutral zone. But we can't see farther ahead. You'd be just as good at guessing as we are. Even though we have a lot of knowledge about China, you have the access to all these world charts and things like that.

All right. And what do you think are the recent drivers for the price increase of all raw materials as you had elaborated before?

Say that again, please?

What do you think are the drivers for the, I mean, the price have been staying low for a very long time, and now it's risen -- gallium has risen 70% since midyear?

So he's asking what were the drivers for the prices to be low for so long and why kind of come back up now?

Okay. So the reason for it to go down so much was too much capacity. As you know, that -- during the last cycle, back in 2010, the gallium price went wild. It went to above almost $1,000 a kilogram. So everybody in China start to build capacity. So way too much capacity, that drives the price down. Now why is it coming back up? Now I think, during the last 2 years, just about everybody even in China is losing money in producing gallium. So some of them has probably threw up the towel and quit the business. And also I think the other driver potentially up a little bit in pricing, is they say solar demand coming up in using those gallium arsenide. As far as gallium is used for LED production as well as making the fluorescent light bulb for the powder. So gallium has many, many multiple applications, which, in technology field, has seen increasing demand, that drives the price up.

Our next question is from the line of Gus Richard of Northland.

I just had 2 quick questions. First, on what is the impact of mix on gallium arsenide versus indium phosphide in terms of margin contribution? In other words, which product has higher margins?

Well, we don't give hard numbers. We'd like to help you by being able to do that, but we found out that customers and competitors were listening to these calls. But we have said publicly is that indium phosphide is higher than gallium arsenide, and that's well known. So if we have a choice of selling $100 worth of gallium arsenide versus $100 of indium phosphide, we'd sell both.

We'd sell both.

Right, understood. And then in the second half of the year, you had a nice increase in gross margin. Was that driven by mix? Or is that higher capacity utilization because you're preparing to move locations?

Yes to both. Mix was better. So for example, Q3 and in Q4, indium phosphide is basically at an all-time high for those 2 quarters, the second half of the year. And the volumes secondly for production was high, just because we were both producing to ship and we're also producing for a little bit of a bank account for categories to have extra inventory. So both of those contributed.

Okay. And then the -- in the near term could be a little bit of outgoing tide in terms of margin just because your utilization rates will be falling a little bit as your new factory comes up?

No. Go ahead, Morris.

No. I think, let's not -- I think in the moving of the factory, which is in the beginning, it's going to last for a whole year. But I don't want you to have the wrong impression that people have taken more material just help our business and in the future, once we move and they're going to deplete the inventory, I don't think that...

Yes, that's not happening.

That's not happening.

Yes, that's not what I was trying to suggest. I was thinking that you have 2 facilities and they wouldn't be fully utilized as you're moving from 1 foot to the other.

Well, here's the interesting thing. And we've discussed this with people that drilled down into the details. So when we buy new equipment, that's basically a push, because of expectations on volume and near-term outlook, which, as you can see from the tone of today's message, we're very optimistic and excited about that. So we're buying more equipment because we would no matter what location we were in. And then that equipment would be depreciated over 5 years. The facility depreciation rate is 27.5 years. So having an extra facility, it doesn't move the needle as much as people might expect. And once you understand what the depreciation life cycle is, then people realize, okay, then it's not huge. So no, we don't expect -- the reason we -- I know you haven't been following the company too much, but the reason we keep trying to set people's expectations for margin in the mid-30%s is because our margin is volatile. It changes sometimes, but it's not because we have some secret awareness that there's going to be the tide going out. In fact, our guidance for Q1 is modest. We don't give guidance beyond that with hard numbers, but you need to read between the lines and get the message because we feel very optimistic about future quarters. We don't see the tide going out.

Yes. I was just speaking in terms of margins. I appreciate your patience and explanation. Just one final one for me. Can you talk a little bit about tax rate, given the changes in the tax law for the upcoming year?

Yes. Basically, we don't think the tax law in the United States, the change, will have -- currently, we don't think there's any big significant difference. The reason is that we have, from historical times, when the company was started, we built up a big NOL carryforward. So we don't have a huge tax liability in the United States and having the rate drop probably won't have a big impact on us. So we generally pay a modest amount of taxes, and most of them are in China.

Got it. So thinking about a similar rate as compared to '17?

Well, let's take a look at '17. So give me a second. So yes, '16 was $733,000, '17 was $792,000. So I would guess we'll be north of $792,000 because our volume's going to be up. We're paying -- we have more employees, maybe more different kinds of taxes, but it's not going to -- it's not going to spike up to $3 million, something like that.

Our next question comes from the line of Edwin Mok of Needham.

I have 3 quick questions. First, did you disclose how much indium phosphide grew this year? Or this last quarter? Or did it reach 30% of sales in 2017? Any more color you can provide around that?

We didn't disclose a hard percent. The last time we've given sort of percent guidance, we said maybe several quarters back that indium phosphide was approximately 30% of revenue. And that's still okay.

I think, in general, sure, we're very feeling strongly about indium phosphide. But we do have a record year for indium phosphide. But on the other hand, gallium arsenide, also germanium grew. Germanium grew almost 30%, as we said in our prepared comments. So -- but perhaps why we're more excited about the indium phosphide growth is it has more legs to count because we are building upon an existing business which has the potential to grow further. And as we said, PONs market is recovering. I think that's a very excited market because in the past 1.5 years, it was mainly the data center growth opportunity which helped us indium phosphide grow. But if PON was to come back, that should layer on top of it. So I think it's perhaps not the percentage which give us the excitement. But overall, year-over-year, I think we have over 25 -- 22%, 23% growth in our revenue from 2016 to '17, correct?

21%. Yes. So the whole total revenue.

Okay, that's actually helpful. Can I jump in quickly on GaAs? Morris, you said it seems like there's a few emerging application GaAs that is driving some growth here? But still, obviously, RF is still a big driver there. Is that a way to kind -- you can kind of think about which one of those markets you will see a particular strength and then near term will -- you're seeing -- you're particularly optimistic about in the near term? Or does it mostly come from still [called main RF PM] market that -- the traditional market where you're seeing kind of just growing volumes?

Traditional market actually would deliver good revenue growth. The 3D sensing has greatest potential, but because it come from such a small volume, to begin with, so we expect the contribution for this year may be smaller, but we do see gallium arsenide, especially LED, that has a great potential to increase in volume, yes.

Yes. So that's an existing market but it seems to be strengthening, yes. And it's requiring a high-end gallium arsenide which is good for us.

Okay, great. And then last question I have, just I want to understand where we were on the JV problems? I think the last time we talked, there was -- you were thinking of pushing JV to take some step to improve the cost and -- of where those business becomes more profitable. Any way you can kind of think about that or is it still kind of staying within this comp, basically closer breakeven the range?

Yes, generally, they're improving, and maybe we did comment a little bit ago, but you might have missed it. But if you compare '17 beginning to '17 ending, it was $900,000 charge for the 7 equity companies dropping down to $307,000. And because prices are rising right now, we do think that it's very reasonable for that group to at least get to breakeven probably very soon. So beyond that, I wouldn't suggest modeling any uptick yet. I think it's too soon. But it's probably next time we give guidance, we can give a little more color and maybe let people move it up a little bit. But until we have a little more time under our belt, I think breakeven's probably good.

Our next question is from the line of Shawn Boyd of Next Mark Capital.

Just a quick one for me, and that's really the comments about indium phosphide and the strength that you're seeing on silicon photonics for the data centers and then this kind of recovery on the PON market. I know these dates kind of flip-flopped and you've got a greater contribution from silicon photonics here at this point into the indium phosphide, and I'm just wondering, do you anticipate it to stay that way? And in very rough numbers, I'm thinking about the businesses sort of half from silicon photonics and maybe a quarter from PON, but anything you can do to help me on that and really as we go in and we think about '18 now. That's the question.

Well, I'm not trying to avoid the question, but the standard question -- answer is actually we don't really know exactly where our customers are using our substrate for. We just sell indium phosphide substrate whether it's 2-inch, 3-inch, 4-inch, we divide it into different doping levels. But when the customer takes it, when they grow the epi-layer out, they can use it for PON, they can use it for detector or they can use it for silicon photonics. The reason why we know silicon photonics is running very, very strong growth is because there are certain customers we know for sure they're using it for silicon photonics. But the others we don't, we just don't know. So sometimes we add up and say, "Well, silicon photonics, let's say is 25% and PONs market is 15%. Oops, you only come up with 40%, the rest is 60%." So it's -- the others -- they are -- we don't know. So what I will say is that ratio of silicon photonics is bigger than PONs for sure, but what percentage it is, we really don't know. We can't help you on that.

And at this time, I'm showing no further questions. I would like to turn the call back over to Dr. Morris Young for the closing remarks.

Thank you for participating in our conference call. As always, please feel free to contact me, Gary Fischer, or Leslie Green directly if you would like to meet with any of us. We would like to look forward to speak with you in the near future.

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