Good morning, good afternoon, good evening, ladies and gentlemen. Welcome to the conference call. Mr. Lee, please mute your call, and I will be standing by. Thank you.
Okay. Good morning, ladies and gentlemen. Welcome to the ASM Pacific Technology 2016 third quarter results announcement investor conference call. Before we proceed, I would like to note that during this conference call, there may be certain forward-looking statements with respect to ASM Pacific Technology's business and financial conditions.
Such forward-looking statements may involve known and unknown uncertainties and risks which could cause actual results, performance, and events to differ materially from those expressed or implied during this conference call. I would like to start with a brief discussion with our 2016 third quarter results, followed by the Q&A session. Let me hand this over to Mr. Robin Ng, our CFO, for the introduction.
Thank you, Nalan. Good day, ladies and gentlemen. Thank you for joining us. I will first provide you with the summary of results for the third quarter ended September 30, 2016, and followed by a Q&A session. Group bookings of $435.8 million were 14.4% higher than the same period last year, but 22.4% lower than the previous quarter in light of seasonality.
Group billings were $540.6 million, rose 29.9% and 14.4% against the same period of last year and the preceding quarter, respectively. Group net profit was HKD 578 million, representing growth of 220.7% year-on-year and 63.1% quarter-on-quarter. Our quarterly gross margin of 39.4% was the highest over the past 23 quarters, and our net profit for the first nine months this year had already surpassed that of full year 2015.
By geographical distribution, China including Hong Kong, Europe, the Americas, Malaysia, and Taiwan were the top five markets for the group in the third quarter. I will now go into the three business segments, starting with the materials business. Our material business performed well, with the third quarter bookings being 30.5% higher year-on-year and 8.9% quarter-on-quarter.
The favorable booking uptrend serves as a good indicator of the current healthiness of the global semiconductor and electronics industry. Billings were $63.7 million, 18.2% and 3.1% higher than the same period last year and the prior quarter, respectively. The group has invested in molded interconnect substrate, or MIS, since the fourth quarter of last year. As the investment is still in an early phase, it has lowered the overall profitability performance of our materials segment this year.
Without this investment, segment profits for the first nine months of this year would have already surpassed the full year segment profits in 2015 by a double-digit percentage. Bookings for the back-end equipment rose 21.5% year-on-year, but declined 29.4% quarter-on-quarter. Billings for the back-end equipment were $287.3 million, up 53.1% year-on-year and 12.6% quarter-on-quarter.
The strength was primarily from the CIS and the LED markets. We are pleased to see CIS became the third pillar of our back-end equipment business, along with the IC discrete and LED businesses, contributing over 10% of the segment revenue. Gross margin was at 45.3% in Q3, an increase of 6.1% and 1.1% over the same period last year and the preceding quarter, respectively. Our efforts in realigning our manufacturing strategies to drive cost reductions yield positive results for the profitability of this segment.
Bookings for SMT Solutions segment during the third quarter rose 6.9% year-on-year, but declined 21.3% quarter-on-quarter. Billings were $189.6 million. It grew by 21.8% quarter-on-quarter and 8.6% year-on-year. Gross margin of SMT Solutions segment at 39% was slightly lower than the preceding quarter due to product mix. Nevertheless, it was higher than what we had expected at the beginning of Q3.
Due to challenging market conditions, we anticipate that this business segment will experience a small revenue contraction this year as compared with last year. As for the outlook, we expect full year 2016 group billings to see a mid- to high single-digit percentage rise. Net profit of the group this year will likely exceed that achieved last year by a margin, since net profit for the first nine months this year has already surpassed the full year 2015 net profit.
In particular, both the back-end equipment and materials segment are expected to achieve full year growth in revenue and segment profits. Overall, we believe that our strategic transformation will put ASMPT in a competitive position to gain market share and capture future market opportunities. We are confident that the group is on track to achieve long-term growth and deliver higher returns to shareholders. On this note, I thank you for your attention and support. We are now ready to take questions.
Thank you, sir. We will now open for questions. If you'd like to register for questions, please press star one on your telephone. Our first question comes from Steven Pelayo from HSBC. Thank you.
Thanks, guys. Congratulations. Great margin performance, and that's something I want to ask about. As you look into the fourth quarter, it looks like you're suggesting revenues are going to decline somewhere in the 20%-25% or so quarter-on-quarter. How much as you fine-tuned or realigned manufacturing, how much can it help flex when revenues are declining? Can you talk a little bit about margin guidance in the fourth quarter by segment? That's my first question.
Morning, Steven. This is Lee Wai. Yes, we expect the fourth quarter to decline. While I think the actual percentage of decline really depends on the market development. As of this moment, as we have shared, the market sentiment is better than the same period last year. So, although we do expect bookings in Q4 to come down, quarter-on-quarter, we also believe, more likely, it will go up year-on-year.
As many of you have realized, actually, one of the good, I would say, reference to our quarterly billing will be looking at the booking in the previous quarter. So I think that already served overall very nice. So far, if you're looking at them by segment, I would say the material segment is developing pretty well. As we have shared, these are booking, the improvement year-on-year is pretty good.
We are pretty happy to see, actually, a quarter-on-quarter booking improvement. So we do expect the performance of this material segment will sustain in Q4. So I think the fluctuation should be relatively small. Whereas for the back-end equipment, the seasonality effect will come in. We do expect it come down, because we are really towards the end of the year, and this year there will be an early Chinese New Year, so we do not expect too much capacity installation by our customers.
But as of this moment, as I mentioned, because of positive market sentiment, and also we see some customers are really continuing to prepare for their business continued expansion in 2017. So we also do not expect too drastic a contraction or adjustment to this business. So that's the overall situation. These are SMT in the, I would say, normal development.
I would say we're happy to see there's some strength being developed recently in the market, in particular in Europe. So these are the positive factors supporting the industry. With anticipated lower revenue in this quarter, we do expect our gross margin to come down. However, we expect for the back-end equipment business with all these manufacturing strategy realignment, we believe we will continue to manage to stabilize the gross margin.
Of course, it won't be able to sustain at the 45% level, but probably come down by a few percentage points. Whereas for the SMT equipment business, we also expect the gross margin to come down. Actually, the 39% in Q3 is, I would say, a nice development, better than what we had expected three months back. So we do expect with this lower revenue and also the geographical and customer mix, the margin will come down. We still expect it should be in the mid to high 30% level. We do still expect that. With more sustainable revenue for the materials, we do not expect a very material variation in gross margin for the materials business.
Great. Thank you. My second question, and I will get back in the queue, is just looking to next year, specifically in the assembly of the back-end business. It looks like your back-end business is probably tracking towards something around 20% growth or so, maybe a little bit more this year. That is obviously, I think, massively outperforming. Clearly, your served available markets in areas like CIS and LED are helping you. So I am curious, as you look to 2017, do you think that served available market is going to allow you to significantly outperform the overall market again next year?
Well, as of this point in time, we do expect CIS and LED demand will continue. We believe next year there will be a strong demand for smartphones. As a lot of people have been anticipating, major new phones will be coming into the market. We do expect the phone makers will come out with the new generation of phones and also equipped with better cameras. So camera will continuously become the differentiating factor among the different smartphones.
So as of this point in time, we have a good expectation in this area. We do receive a good, I would say, indication from customers. At the same time, we also expect LED to continue to improve. We also see the demands of this point in time are sustainable. For those new capacity we deliver to customer, the utilization rates are high. Customers are continuously giving us an indication they will continue to expand.
Now, we also continue to see with the mid-term consumption, at a pretty high level, in the second half of this year. That actually give us a very good support on our confidence. The IC discrete market is healthy and may continually improve. Also, with the higher demand for smartphones and even LED, we do expect correspondingly there need to be support for ICs and discrete products. Overall, as of this point in time, we are having a view of cautiously optimistic about 2017, in particularly for the back-end equipment business.
Okay, excellent. I will go ahead and get back in the queue.
Thank you.
Thank you, sir. Our next question comes from Mr. Bill Lu from UBS. Thank you.
Yeah, hi. Good morning. A two-part question on the CIS business. The first part is, you've said that CIS is over 10% of segment revenues. Can you give us maybe an outlook for what that might be in 2017? Secondly, it does seem like a lot of the Chinese handset makers are going to adopt better cameras or dual cam. Assuming we see these models come out, say, 2Q of next year, what is the timing for equipment purchases?
Morning, Bill. For the CIS, I think we really developed very well this year. We expect on the full-year basis, there will be a very substantial year-on-year growth for this particular area for us. Looking into 2017, as I mentioned earlier, we do receive a very positive indication from customers. Also based on our own assessment of the market, we expect next year, as I mentioned earlier, there will be many new phones, with very good features coming out in the market. Today, we expect the CIS contribution to our back-end equipment business to continue to go up in 2017. Can you repeat your second part of your question?
Yeah. I'm pretty new to this business. I'm just asking, if you want the product in the market by, say, 2Q, would you have to order equipment, say, in 4Q or 1Q? What is typically the lead time?
Oh, okay. It really depends on customer. Typically, our equipment lead time will be around three to four months, for this type of equipment, and in particular for the active alignment machine. This is more a special machine. However, there's also a gating factor over there. Because this is a very precise machine and also, we need to build these machines in clean rooms.
We are also sharing with our customers, clean room space could be a concern for us. We are telling customers, "Well, if you really need good quality of machine, you must start to order earlier." I think in this respect, we do expect, if the market's strong, customer has to order the machines earlier than the normal lead time. Anyway, we expect most of the machines should be installed in Q2.
Q2 is the peak, early part of Q3, and with the warning we are giving out to customers, we do expect some installations by our customer in the later part of Q1. These are the, I would say, typical peak season for us. However, this year is really a good year for CIS. In general, the seasonality of CIS is higher towards Q4. Most installations should have completed already. But this year, we continue to see installations of CIS machines still in this quarter.
Great. Thank you. That is very helpful. Second question is, there is a lot of talk on next year, the handset makers, or at least one big handset maker, is going to substrate-like PCB. Can you talk about what that means for your SMT business?
Well, we do not really know it yet. If you are talking about those really very tier one, very high-end machines, everybody knows. You do not know too much about it. But, in the SMT market, we are really in the forefront of the technology. We continue to engage with some key phone makers to customize our equipment to a certain extent to cater for their new demand. But I cannot talk too much beyond that.
I guess, maybe not talking about that, this one big customer, but just generally, if you are going to switch to substrate-like PCB, that will require brand new SMT investment, correct?
It really depends on the demand. As I mentioned, we do not really know all the details as of this point in time. Maybe a few months later, it is easier for me to comment on this one.
Fair enough. I ask you again in three months.
Okay.
Thank you, Bill. Our next question comes from Randy from Credit Suisse. Thank you.
Okay. Yes, Sam, thank you. I wanted to ask a follow-up about the high-end platform refresh. in 2014, you had a big order, and I think initially into 2016, you were expecting potential for a platform refresh. I guess now at this stage, if it looks like that is coming for 2017, or do you have any expectation that the tools they purchased a couple of years ago, you would need to see a refresh, and if the magnitude, if it could be kind of similar to what we saw a couple of years back?
Well, I think we have the same expectation like you. To be honest, we are also a little bit disappointed. This year, this refreshment cycle did not really happen. However, next year, it will be three years already, so I think it will time actually continue to, I would say, increase the level of expectation we have. Furthermore, as we discussed just now earlier, there are also some new features, new technology, new materials, new substrates, coming into the market. So we are also consciously and anxiously watching for this development. So, we do believe customers will need a newer generation of equipment, better capability to handle this new technology, new requirements. As of how will be the magnitude, we really don't know at this point in time. Yeah.
Okay. Yeah, I want to also ask on the fan-out and InFO-like packaging. If you could talk about, you have a couple of slides on the tools you sell into that space. Could you talk about, is there a difference in your exposure if it's a wafer-based approach like TSMC did, versus the panel, where it looks like, are you more optimistic on your exposure to panel? I guess for next year, do you expect a meaningful revenue where this could become another 10% contributor within the back-end business?
Actually, we are optimistic for both, actually. We believe wafer level and panel level, they are going to serve slightly the market. In general, the wafer-level fan-out packaging will be able to serve higher-end applications with a finer line width and also finer spacing between the lines. So, in general, people are able to, using the wafer level technology, they can achieve a line width and a spacing of 2 μm by 2 μm . Whereas panel level, in the past, people are more talking about 10 μm line width and 10 μm spacing and above. However, this year we also start to see their customers developing new technologies to build a gap. So their customers talking about coming down to 5 μm by 5 μm. This will be interesting development we are seeing. So we believe, at the end, they will serve different markets.
I think, for those really requiring a very fine line, a high I/O connection, the wafer-level fan-out packaging has its advantage. For those that don't need this kind of accuracy, so the panel level packaging, either the 5 μm kind of accuracy or above 10 μm accuracy, will offer them the cost advantage. Our engagement, as of this point in time, almost covering every, I would say, segment of these applications. The development is interesting, but on the other hand, we don't expect this wafer level packaging or advanced packaging will be able to contribute to more than 10% of our business in 2017. We do expect it will continue to develop.
However, we are pretty optimistic about its longer-term future. I think, we are of the view that the technology has nicely come to a point. It should be able to take over, and it should be able to offer the end customers some real value, such that these customers are willing to pursue this technology. Hopefully, this will, after the CIS, LED will continue to serve as an important revenue driver for us in the next few years.
Okay. Thanks a lot.
Thank you, Randy. Our next question comes from Mr. Michael Chou from Deutsche. Thank you.
Let me take just a follow-up question for Fan Out. You mentioned Fan Out should be a long-term growth driver, but do you expect your panel level Fan Out solution will see significant growth on this versus this year? That is my first question.
Sorry, Michael. The last part of your. You are fading out, so can you repeat the last part of your sentence, your question?
I am sorry. I am wondering, your Fan Out revenue next year, would that grow strongly versus this year? Or it will be a gradual pickup quarter by quarter next year?
Well, difficult to predict. It really depends on the market. On one hand, we are seeing there is some potentially good market sentiments driving the strong demand. As I shared earlier, I think everybody expecting new phones to be available in the market. We expect the competitive landscape will drive even many more new phones to be available in somewhere next year. So that should drive chips demands, capacity expansion in the IC discrete area, CIS area, and also even the SMT. We do expect this. However, to what magnitude will it be explosive kind of a ramp up, we really do not know at this point in time. So I think we have to watch in the next few months how these customers will really place order for the equipment.
The other thing is, do you think the panel level solution is quite okay for mid to low-end product over the next 12 months? Because one back-end company indicating that they would start to ship the panel level solution to their customer in Q2 next year. Do you think that will be cost competitive versus wafer level Fan Out solution?
I think next year, some customers will start to probably supply a small quantity to their end customer for testing. I do not expect a massive adoption next year. I think, judging from our engagement with all these customers, I think it will still take some time. Most of our customers still in the development stage. They do expect they should be able to come out with something sometime next year. But judging from the cycle, after these customers develop their process, they will assemble chips for their end customer to qualify. So it will take some time. So we do not expect a significant production volume next year, so we believe soonest will be 2018.
Okay. Can we say that your panel-level fan-out solution should be one of the cost-competitive solution, or even the most cost-competitive solution in the industry compared to your peers going forward?
Well, I think whether it is a cost-competitive solution actually more depends on our customer. Different customer have different solution, very interestingly. We are engaging with different customers to participate in different programs. Each of their programs, their own merits. I would say they are addressing slightly different segments of the business.
Because we have the widest product portfolio, so that is why we can participate in a different way. At this point in time, I would say customers select ASMPT for their program, not necessarily because our equipment are of the lowest cost. Probably, they like our industry for ASMPT. They can work with ASM, and to really develop a technology to realize their risk. So for example, some customer, we are adapting our equipment to suit their need. So this, I think, is more of the reason why these customers have selected us.
Of course, I think if all these programs are successful, I would say every one of them will be able to offer their end customer certain value in their selected segment of application they participate. Some could be cost efficiency, some could be performance, so not necessarily must be the most cost-effective, but it could be at a certain price level, certain cost level, but they can offer a much more advanced competitive solutions.
Can we say your market share in fan-out solution, especially for panel-level fan-out solution, should be quite good compared to your progress in flip chip before? Can we say that?
Well, at this point in time, it's difficult to talk about market share because the shipment is really small. But in terms of engagement, I think we are really clearly leading. We are engaging all the key programs, almost all the key programs. As I mentioned, different solutions, different customer also selected to engage ASM.
It may not be everybody have a full engagement, but almost, ASM must be able to offer them something in certain process. I think from that perspective, we do expect if the panel level and wafer-level packaging technology take off, I think we probably should be able to enjoy a good market participation and hopefully a good market share.
Oh, okay. That's very helpful. Thank you. The other question is regarding your entry for SMT. As a higher line density for substrate-like PCB, although you mentioned you don't know when you will see the massive upgrade, but if a customer wants to adopt this solution, would they need to upgrade some of their SMT machine from a technology perspective, or they can just use the current machine to do something like PCB?
As I shared with some of you before, earlier, really we do not know too much about those new substrate technology at this point in time. As I share, we do have engagement with some top-tier customers, developing solutions for them. We don't know how much this kind of technology will be adopted on a larger scale. Will there be a modified solution for that? It's difficult for us to tell at this point in time. But in general, customers will take advantage to upgrade their capacity and their capability every few years. That's why, I would say, we are increasing our level of expectation, for a new round of investment for SMT next year.
Oh, okay. For SMT outlook, what is your outlook for closed loop SMT solution and the mid to low-end solution, in terms of revenue contribution next year?
I think we'll still be in the development stage, gaining traction, more and more customers are adopting it. However, in the industry, this kind of new technology will take some time to gain customer acceptance. I would say, it will still take some time. Yeah.
Thank you. Our next question comes from Mr. Ken Hui from Huatai. Thank you.
Thank you for taking my question. My first question is regarding LED. I think, you correctly predicted that the orders will recover when the LED chip prices stabilize. We did see very strong growth for the business for this year. Are you concerned that we are having a high base for this year, and therefore the growth for LED may be slower or may even turn negative into next year? That is my first question.
We don't think so at this point in time. As I mentioned earlier, we have closely monitored customers' capacity utilization. For those new capacity we deliver to them, install for them, they really put into utilization very quickly, so high-capacity utilization at this point in time. We also monitor these customers, their own financial performance.
These customers, I would say, delivering a better financial performance, with the new capacity. They are also citing visions for the improved performance for their own, due to they are selling higher margin products, higher priced products. We believe the LED industry is developing in a healthy trend. People are no longer having a cut-throat price competition. We do expect 2017 will continue to be a healthy market for LED.
Thank you. My second question is relating to the wafer-level packaging. I think, previously, you talked about the difference between RDL-first and RDL last. Is there any particular size that you're focusing your business on, or you are just trying to offer every possible solution to your customer? What would that mean for your opportunities in the market?
Well, we try to address all, I would say. Actually, our solutions are so flexible, we are able to address all. At this point in time, the wafer level, more of RDL last. The panel-level, I would say that those customers in the leading edge of the development at this point in time are basically also RDL last. We also believe, those RDL-first solutions may also, I would say, speed up in the near future. As I mentioned, our solution will be suitable for both. Yeah.
Do you think that you probably are having better advantages in RDL-first?
Well, RDL first actually may be easier for many people to participate. Because it is basically almost like a flip-chip bonder, okay? Of course, we need a high-speed flip-chip bonder, but it does not need a very high accuracy in a certain way. Whereas, RDL last approach is more of a placement machine. I would say as of this point in time, actually, ASM has solution for both.
Since we have dedicated, developed semiconductor equipment for very accurate die placement, either flip chip or die attach placement, and we also have the SMT solution customized for this advanced packaging. Addressing those people, they do not need a very high placement accuracy. For example, the customers talking about, they can do effective patterning, such that they can have a less accurate placement, a less accurate RDL. I think ASM is in a very good position in that, as of this point in time. Yeah.
Thank you very much.
Thank you, sir. We have a following question come from Steven Pelayo from HSBC. Thank you.
Yeah. I'm curious if we could talk a little bit about your SMT market outlook for next year. I guess the presentation that you guys had 90 days ago, your internal market intelligence group was suggesting the SMT market may be a little over HKD 3 billion or so. It looks like you've lowered that in the last 90 days or so to maybe, I don't know, looks like about HKD 2.7 billion. Did something change over the last 90 days to make you guys think the SMT market is smaller next year than you originally thought?
Well, I don't think actually. Well, I would say, 90 days ago, people are still, hopefully expecting the more believable investment, believable capacity from some customers. It did not happen. But other than that, I don't really see there's a major development. I don't think so. Actually, Steven, to be honest, all these are changing very dynamically. It could be in the next 90 days, there will be a new view of it. Really dynamic. Yeah.
Okay, I understand. It's difficult for us to really get kind of good market sizing data on SMT market, so we really rely on your guys' outlook there. My next question, maybe it's for Robin, I guess. Again, you guys have done a great job on the gross margin line, and even operating expenses for that matter in the third quarter. I'm curious once again, if revenues in the fourth quarter were to decline 20% or so, can OpEx decline at kind of half the rate or more, or how much can you flex the OpEx line?
Well, there's less flexibility in OpEx. Typically, it will be slightly lower in Q4 compared to Q3, but the percentages are not really that much. I would say, the OpEx are not really that flexible on a short-term basis. That's why overall, we do expect the profitability will come down quicker than this revenue contraction. Yeah.
Okay, fair enough, guys. Thanks.
Thank you. Our next question comes from Mr. Bill Lu from UBS. Thank you.
Yeah, hi. Following up on Steven's question, if you look at OpEx, the company has expanded into quite a few new categories over the last several years. But your OpEx, both in terms of R&D and SG&A, has increased by 30%-40% in the last five years. Is that now a good level? Or as we look out into 2017, you think it has to go higher because you're still doing a lot of new things?
We believe it's already at the level we like to control. Overall, it was still, we believe, going up on a very moderate rate. In terms of operating expenses, a lot is still human causes related. So, salary increment, that's still bound to happen every year. So we still do expect some of these positive upward trend. On the other hand, we have to exercise our very tight control on the expenses.
So, we should be able to offer a certain extent. On the other hand, we continue to consolidate our SMT business, combining the printing and the placement divisions. We do see this combination have the synergy effect to lower the operating expenses. So I would say, going forward into 2017, I will suggest you model a very moderate OpEx increase.
Would moderate be low single digits, or how do we think about that?
Well, I think for OpEx, if you go up beyond a single digit, I won't say it is moderate anymore, on a year basis.
Fair enough.
Yeah.
Secondly, I don't think you gave an update on this call, maybe I missed it, on the manufacturing facility in Malaysia for SMT. Can you just update us on how that's coming along and when we should expect that to ramp?
Okay. Well, actually, it's developing on track. They already started to ship out the printers from the Malaysia factory. The placement machine, we are expecting the initial or the first machine should be out of the factory within this quarter. Of course, it's a very small quantity at this point in time. But I think the overall preparation is well on track. We should be able to see some positive contribution to the financial result next year.
Great. Thank you.
Thank you. Next following question comes from Randy from Credit Suisse. Thank you.
Yes. Thank you. I wanted to ask a follow-up on the gross margins, where SMT now is running mid to high 30s, and then back-end pushed to mid-40s. Could you give a new look at your target for these segments, where you've had a lot of actions on the change in some of the sourcing, and then in the back-end, the outsourcing, if you see room for further improvement or we're done with those programs and these are the right kind of levels for the high season of the year?
Okay. Well, for the back-end equipment, as you mentioned, we are happy to see, during the good quarters, we already reached the mid 40% level. Overall, for the nine months, we are also above the 40% level. Now, we are still putting in a lot of effort to further improve this area. So one of the major initiations in this area is to go further automation.
As of this point in time, we have installed four robotic lines in our factory in China. Connecting around 60, 70 machining centers. So with that, actually, we are able to run those machining centers with very minimal operator intervention. And also we are able to run this machine 24 hours a day. So that will be significant to our productivity improvement, cost reduction. So, when we have successfully implemented this, it may allow us to cut down the level of outsourcing.
Without significantly increasing our internal manpower. As I shared with you before, there is a cost to pay for the outsourcing. So if we can also retain those profits internally, it will contribute positively to our gross margin. So we do expect next year will be a major investment for us in this area. So hopefully, with all those in 2018 and beyond, we can see further structural improvement in gross margin of the back-end equipment.
Now, in the SMT, just now we mentioned about the Malaysian factory. As I shared before, actually, we are targeting to have the Malaysian factory to share could be up to 50% of the production loading of the SMT business. So when we materialize this, you can imagine that it will bring a very interesting improvement in our gross margin in the SMT business.
So we are putting in these efforts to drive this area. Whereas in the materials business, we continue to adjust our product portfolio. We are putting in this new investment in the MIS molded interconnect substrate, because we expect this is a much higher margin material business for us. So it will be very positive for the long term of this segment of business. Within the leadframe, we are also focusing to address more of the higher margin business. So, we continue to adjust for this. So, we are setting a target to further improve our gross margin as a whole for the group. So hopefully in the next two years, we continue to develop, activate, deliver this performance.
Okay. Ken, I just had two follow-up questions to that. The MIS, I think you mentioned, was a little bit of a drag now because of the cost. If you could talk a little bit how much drag you're seeing now, and then how much that could improve margin or improve your revenue scale once that ramps. And the other question I had on the CapEx, with this automation investment, year to date, I think in the last slide it shows it's underspending a little bit year- to- date at HKD 36 billion versus the HKD 84 billion budget. If you still expect to spend that and if next year may be a higher CapEx year for the automation.
Well, actually, because the MIS is a joint venture between ASMPT and our customer, that is why the loss is also shared in our P&L. Actually, it is not really that big a number, a small number at this point in time, because the scale for that business is still small. It is still an initial investment. When we go further to develop it, expand it, please do expect the loss could even expand before it turns into positive, because we have to put in more investment, we have to scale up the operation. But we believe these investments are worth it, because it really has a promising future going forward. Yes, for the size of this business, we do not really know yet. However, this could create a new segment by itself, because we provide opportunity to customer that the existing frame or substrate cannot serve.
For example, in the RF application, we are seeing that. There are also customers working with us using this technology for some power applications. This also not able to be achieved easily with the leadframe and the substrate. It also offer customers in a certain area to provide a thinner package, that also all those existing substrate or packaging solution cannot offer.
It also offers customers opportunity to lower their substrate cost for certain applications. There are many potential applications. I would say in going forward, when one day this business become a segment portion of our mature business, I will not be surprised at all. Although at this point of time, it is really difficult to quantify because it is really a new business by itself. One second. Yeah. I hear. Okay, you have your second question? Yeah.
Y eah. The second question was just about the CapEx, if you plan to still spend your 84 budget, and if the automation may lead to higher CapEx next year.
Yes. I think, we believe probably, we will underutilize our CapEx budget this year. But we do expect we will spend the money next year, for, as I mentioned, automation. This robotic automation is one of the important investment. But overall, we still maintain the overall strategy of trying to level with our depreciation. We have not totally prepared our CapEx budget for next year. But we will try to use whatever being unused this year for further automation next year.
Okay. Thank you.
Thank you. Our next question come from Mr. Ken Hui .
Thank you for taking my question again. My fourth question is regarding the CIS business. I think within the CIS business, obviously one of the big revenue contributors is actually the active alignment tools, given that it is a very expensive machine. But I think some of your key customer are still trying to adopt their in-house solution. Have you seen any sign that they are also giving up their in-house solution and trying to move to adopt your solution? That is my first question. Thank you.
Well, that has been happening already, I would say. Initially, this is purely an in-house solution market, with only very small number or even a small size of external suppliers. When ASM comes into this market in a big way, first we displace some of those small suppliers, and we also gaining customers' acceptance more and more. Even not totally, I would say, but many of them already switch some of their loading, from using internal equipment to ASM.
We believe this will continue, because an in-house active equipment development always has a disadvantage, because we cannot catch up with the latest technology development in the market as fast as an independent third-party supplier. I think ASM has this advantage, and the size, the scale of ASM offer us this position. That's why when we come into the market, we quickly displace those smaller players. We expect today, strictly speaking, the only competitor for ASM are all those in-house supplier. We expect our market share will continue to get over them.
Okay. Thank you. From a future product roadmap perspective, would you try to focus on cost down the active alignment tools, or would you try to further improve the UPH for the tools?
Actually, I would say three areas. Just now you mentioned two of them, cost down the machine and also to further increase the throughput of the machine. Actually, customer also continue to engage ASM for their new requirement. We are also working continue to work with customer to develop machines or modify the machine, customize the machine, for their new generation of camera sensors. We expect all three development directions will continue to exist concurrently.
Okay. Thank you very much.
Thank you. Once again, ladies and gentlemen, should you have any question, please press star one on your telephone. Ladies and gentlemen, should you have any question, please press star one. Thank you. Gentlemen, there seems to be no further question at this point in time.
Well, I think we've had a good discussion this morning, with very good questions. I'm afraid we have to conclude the conference call now in the interest of time. Thank you very much for joining our conference call today, and we'll talk to you again next time. Thank you. Bye-bye.
Thank you. Bye.
Bye.
Thank you for your participation. This conclude our conference. Thank you.