Joseph Tung - Chief Financial Officer

Randy Abrams - Credit Suisse Rick Hsu - Daiwa Securities Roland Shu - Citigroup Szeho Ng - BNP Paribas Gokul Hariharan - JP Morgan Eric Chen - UBS Andrew Chen - Yuanta Investment Consulting

Welcome to the ASE Group Third Quarter Earnings Release. All participants consent to having their voice and questions broadcast via participation of this event, please refer to Page 1 of our presentation which contains our Safe Harbor notice. I would like to remind everyone on this call that the presentation that follows may contain forward-looking statements. These forward-looking statements are subject to high degree of risks, and our actual results may differ materially from these forward-looking statements. Our CFO, Mr. Joseph Tung will be going over the financial results followed by a Q&A session. Following the event, our VP in-charge of Public Relations, Eddie Chang, will be addressing the media in Chinese after the release. Joseph?

Thank you very much everybody. Thank you for coming to our conference earnings release. And before I start with the presentation on our third quarter results, let me give you a little bit of a backdrop on the SPIL tender offer. This is just timeline of events that we've gone through. I think as everybody knows, on August 21, we announced the tender offer of minimum 5% and maximum 20.99% of SPIL. And on September 28 - I think the tender offer was officially launched on the August 24. And up to September 18, we successfully reached the minimum 5%, and on 22, the offer was completed. We had altogether roughly 36.83% of the SPIL shares tendered in, but we owned 24.99%. Therefore the participation was separated. And on October 1, we completed the exchange of the payment for the shares and we become - officially becoming the 24.99% of SPIL’s shareholder. Such investment will be treated as with equity method. It will be booked under our long-term investment and will be entitled for pro-rate share of the earnings of the company and also we will be entitled for the dividends going forward. Okay. With that, let me start with the financial results of our third quarter. I think in quarter three, we had actually a pretty good quarter. I think in terms of overall consolidated revenue, we had about 4% growth, of which, IC ATM-wise, packaging, we had about 3% growth here on a consolidated basis, but if we include the SiP assembly works that we do, which was eliminated in the consolidation, the actual growth of packaging is actually 6% in the quarter, which - and also in testing, we had about 3% growth, whereas direct material, we had 9% decline in terms of sales. It's also worth noting that in the 9%, it is also with the elimination of some of the substrates that we made for USI Shanghai for their Wi-Fi module and that part of the business was also eliminating the consolidation. So if you look at the materials stand-alone operations, that number is actually 7% growth in the quarter. EMS with the ramp-up of some of the new products and SiP products and also from seasonal pickup in the SiP product. We have been doing the growth in EMS of about 5% in the quarter. So all in, we had about 4% growth. And with the increased revenues, the gross profit margin also improved, as you can see here 16.5% to 17.8% in this quarter. And operating expenses increased slightly from 8.8% to 9.1% and therefore the operating margin improvement was about 1.1% from 7.7% to 8.8%. In the quarter, netting out the income tax and all the non-controlling interest, the net income that we achieved was NT$6.3 billion with the 8.7% net margin. In terms of non-op in the quarter, we have about NT$1.4 billion non-operating gain comparing to about NT$10 million loss in the previous quarter. That NT$1.4 billion gain really came from first of all the ECB valuation gain of around NT$1.4 billion and also in the quarter on a consolidated basis, we had a foreign exchange gain of around NT$600 million, netting out the interest expense of NT$500 million, we had a net about NT$1.4 billion gain on the non-operating level. EPS for the quarter, basic EPS was NT$0.83 compared to NT$0.48 in the previous quarter. In terms of EBITDA, our EBITDA margin had some improvement from 19% in the previous quarter to 21.8% and the total amount reached NT$15.9 billion. Comparing to same period last year, revenue grew 9%, I think almost all the growth came from EMS because of the new SiP product that we had been shipping, but in terms of IC ATM, packaging actually had about 8% decline, again without the elimination of the consolidation, the actual drop is about 5%. Tests, we had about 6% drop and direct material sales is about 26% below previous - same quarter previous year, reasons being that we have - we experienced softer demand, particularly in the DRAM - substrate use for DRAM packages. Consequently gross profit margins declined slightly from 21.3% to 17.8%. Operating margin declined from 12.2% to 8.8%, and OpEx, operating expenses as a percentage of sales, also came down slightly from 9.1% to 9% in the quarter. Looking at the IC ATM business. All in all, we had about 6% growth in the quarter, which is a bit higher than what we've guided last quarter, reason being that we are seeing some pull-in in third quarter, particularly in some of the SiP products that we're building. Because of the increase of revenue, gross profit margin also improved from 25.2% to 26.7% this quarter. Operating expenses in the quarter increased slightly from 11.7% to 12.5% this quarter because of the increase of mostly the employees’ bonuses, as well as some of the professional fees that we paid for the transaction that I mentioned earlier on. Operating margins improved from 13.5% to 14.2% in the quarter. I think in terms of gross profit margin, the favorable FX movement also helped in terms for IC ATM, the FX NT$ depreciation actually helped the gross profit margin by 0.94% in the quarter. Comparing to same period last year, the total revenue actually came down about 6%. As you can see, packaging came down 5%, test 6% and direct material sales came down about 10%, again because of the lower DRAM substrate sales. This is the past seven quarters’ performance. You can see that from quarter two this year to quarter three, packaging revenue went up about 6%. So comparing to same period last year, it came down around 6% as well. In the quarter, gross margin improved from 22.7% to 24.8% in the current quarter. In terms of revenue breakdown, you can see that advanced packaging actually came up from the 31% in the previous quarter to 34% in the quarter. In terms of SiP packaging business, it's about 9% of the overall packaging revenue, therefore the flip chip and wafer bumping with the level of CSP represents about the other 25%. Wire bond, although it still had some growth in the third quarter of 4%, in terms of the overall percentage, it came down from 57% to 56% this quarter. And we experienced some decline in the discrete and module business, therefore the percentage came down from 12% to 10% in the quarter. Looking at test operations. We had a mild 3% growth sequentially, and gross profit margins consequently also improved from 35.2% to 36.1%. In the quarter, for test operation we had a total CapEx of around NT$25 million. And in terms of our tester counts, in the third quarter we added 72 bond testers and we retired 25 of them, so the final count of testers was 3,417 units. Utilization was maintained at about high 70% with ASP being stable. I have to come back to packaging business, forgot to mention. In terms of packaging CapEx, we had about NT$83 million spent in the quarter, of which, NT$40 million for advanced wafer - for flip chip and wafer bumping and the rest NT$43 million is for common equipments for the other packaging activities. Our utilization in terms of packaging for advanced packaging capacity was about 80% and for the rest it's about 75% in the quarter. In terms of bonder counts, in the end of the quarter we only added one bonder and retired 46 of them. So the total bonder counts at the end of the quarter was 15,617, and of which, 13,403 are copper available. Material operation in the quarter, the performance was a bit disappointing. In terms of overall revenue, it actually came down 12%. The direct sales actually went up. In terms of internal supply actually came down. Right now, it only supply of roughly 23% of our internal use and because of the lower revenue, the overall margin also came down and because of the not as good product mix - unfavorable product mix change, the profit margin came down from 16.2% to 11.1% in the quarter. We are expecting the substrate for DRAM business will start to pick up in the fourth quarter and then we will see some improvements in our material operations in the fourth quarter. In terms of IC ATM revenue by applications. Testing and communication stayed flat at 55%, PC actually came up about 1% because in the substrate, the DRAM demand was soft but in terms of packaging actually DRAM packaging actually came up in the quarter. And also we had increased some of the storage area and also graphics, so the PC-related business actually in percentage-wise came up from 10% to 11% at the expense of consumer and industrial. Looking at our EMS operations, the revenue had some growth to NT$36.2 billion and with the margin also improved from 6.4% to 8.3%. I think the gross profit margin at 8.3% is above our guidance and reason being that the FX really helped the margin by about 1.3% and also because of the more favorable product mix change, there is another 0.7% improvement, nearly 0.6% improvement in the margin. We are expecting actually with the SiP product ramping up further in the fourth quarter, the EMS margins will actually come back down to the previous quarter's level, which I will give you a reference when I give you some guidance. In terms of revenue breakdown because of the product ramp up, we've seen in communication actually came up to 56%, while the others came down. In terms of balance sheet, the cash and cash equivalents in quarter was at NT$45.6 billion, down from NT$58.9 billion a quarter ago, whereas our overall interest-bearing debt also increased from NT$91.9 billion in previous quarter to this quarter’s NT$124.5 billion. Now if we look at the breakdown in more detail, I think what happened is in the quarter, we have - adding to the cash and cash equivalents balance of NT$58.9 billion in the quarter, we added roughly NT$10.5 billion of operating cash flow. That balance was used to pay the - paid for the tender offer and after that payment, we have net of NT$34.4 billion cash. And in the quarter, the increased interest-bearing debt, which includes roughly NT$6 billion from ECB issuance and another NT$26.6 billion borrowing from banks, combined NT$32.6 billion which used to pay off - to pay for our cash dividend as well as our CapEx with a net of 11.2%. So adding NT$34.4 billion with the 11.2% is NT$11.2 billion. We have a total balance of NT$45.6 billion in the quarter. Current ratio is maintained at 1.31 and net debt-to-equity increased to 0.48 from 0.22 to in the previous quarter. Okay. In the quarter, our EBITDA was about NT$500 million, whereas the CapEx is about NT$140 million in the quarter. Of that, NT$140 million, like I mentioned earlier on, packaging is about NT$83 million and NT$25 million for tests. We spent about NT$24 million for EMS and then the rest for the material. I think the full-year CapEx will be just down from previously NT$700 million to NT$800 million to NT$600 million to NT$650 million in this year. With that, I'll give you a little color about how things will look in the fourth quarter. I think given the current business outlook and the exchange rate assumptions, we are expecting fourth quarter to be as follows; IC ATM capacity should stay flat, and blended utilization should be down 4% to 6%, sequentially. Whereas gross profit margin in IC ATM should resemble first quarter 2015’s levels. EMS capacity should say flat and blended loading should be up mid-to-teen percentage sequentially. And EMS gross margins should approach second quarter 2015 level. Now with that, I'll be opening the floor for questions. Q - Unidentified Analyst: Hi there.

Hi.

Can I just start with the guidance? I'm not really clear on the EMS gross margin. If you look at 2Q to 3Q, you’ve got more SiP, currency helped, and therefore margin went up. 3Q to 4Q, you’re going to have even more SiP, currency should still help, and margins going to go down to 2Q level. Can you just help me understand that?

Yes, I think third quarter is a bit unique. I think third quarter we have two things happening at the same time. One is of course the FX does help the margins, and the other factor of that is we actually had some push-outs in some of the SiP products because of the - such product is going through transition and the real volume - actually the volume in the third quarter was actually less than in the second quarter. Because of these for one particular project is really EMS is providing its logistic support, therefore the margin on that is actually lower than the average. So will that increase in fourth quarter? That will actually put the margin down a bit in the quarter.

So that product is coming back in the fourth quarter, the same generation or the next generation?

The advanced generation.

Okay. And margin is still not as good at the…

That’s because for that particular product EMS is only providing logistic support.

Okay, got it. And then maybe if we just look at SiP more broadly, can you talk about the outlook for 2016 in terms of broadening out the customer base, new projects, and also it seems like when I talk to EMS companies now, more and more people are talking about getting into SiP. You guys were early, so what the competitive landscape looking now a year from now?

I think it's only natural that the customer will eventually look for secondary and third source and that is bound to happen but we will continue to be the primary supplier of these products.

We will now begin the question-and-answer session. [Operator Instructions] Thank you.

Although I'm not - we don't know how much growth there will be because we are still in our budgeting cycle. And also, the SiP products that mostly assisted by the sale. So the market aspect of it is they are different from just being leading components. So we will have to see how things would shape up, maybe we will have a clearer picture in the coming quarters. In terms of the - first of all, we are expanding our customer base. I think we will see some meaningful numbers coming in next year. We have been engaging with quite a bit of customers in different product categories and with, I think, this year, we are already seeing some volume ship out in some of those products and we are expecting the numbers to start to grow in next year.

I guess squeezing one last question. It's been a few weeks since the tender offer was completed. Can you give us a sense for what kind of conversations you’ve had with SPIL and also what’s the feedback coming from the customers?

I really don't have anything to report at this point. We will continue to see for dialog hoping to form any simple form of collaboration, but so far I don't have anything to report.

Randy. When you get the mike, announce your name and company.

Thank you. It’s Randy Abrams from Credit Suisse.

Hi Randy.

First question on the 4% to 6% decline in utilization. Normally fourth quarter, you have the flagship launch, you get contributions. So could you talk about where the weakness is coming from, and is there any SiP that would book, like utilization is down 4% to 6% but will you have some incremental product flowing through that may not affect utilization but could help sales so normally going pass some of the bookings?

Well, I think the overall softness in the fourth quarter really came from two directions actually. One is on the traditional IC ATM. We did see some pull-in in the third quarter, and also the - this is kind of - on the IC ATM side of it, there is some pull-in in the SiP product because of the preparation for the product launch for the new generation product launch but on the EMS side, the final product, there was actually some push-outs for quarter because of the timing gap. So from an IC ATM standpoint, there was some pull-in in third quarter and therefore that had some impact on the fourth quarter. On top of that, we do see that inventory still being digested in the quarter with happen to see end of it. So the overall softness still persists in fourth quarter, and therefore it has some negative impact on the overall revenue on the IC ATM side.

Okay. For those inventory adjustments, is it pretty even across end-markets or it's one segment or concentrated, and what’s the early view on where we’re at into first quarter if it could be above seasonal when we’re coming off of a full base?

Not to give any guidance on first quarter, but I think what we are seeing now first quarter of next year seems to be pretty normal quarter for us. There is going to be seasonal decline terms of the overall business.

Okay. Could you talk on the SiP, it seems like with the ramp you were back at a 6% to 7% margin for the year, and that’s for gross margin. Is that the way we should think about that business for next year, just given it still have a lot from a big customer or there are things that may change the profitability of profile, I guess better or worse?

Yes, I think last quarter I mentioned for one particular SiP project, we are still in the investment stage. So for that partner, this is still not exactly at the targeted loading level. So the main focus on that particular business is moving to streamline the cost structure a bit and we have been doing quite a bit of effort on that. So we’re seeing that the cost is being better aligned. But still I think that particular business will still have some drag on our overall EMS margin. So next year - again, I can't really give you a projection so to speak, but I think what we are reaching now is something that should be sustainable.

And the final question, the CapEx cut you made. If you could talk about the areas where the initial budget where the CapEx cuts are coming, and I guess if should think of this is as a low base next year anytime we’re back to depreciation levels, get back to what we originally started this year?

Well, I think the - I typically divide our CapEx into three categories. One is what we call the maintenance CapEx, is basically the - every year if we assume a 10% ASP drop, you lose 10% of the capacity and we have to make it up. So that 10% lost capacity if you want to make it up is roughly NT$500 million a year. And if there is growth, then it’s a dollar-for-dollar investment to support the growth in the business. And the third category is really the projects that we've been doing. For this year, in particular, the number seems to be within that range. I think NT$500 million for maintenance because there is really no growth in the traditional business, hence for the project we have about NT$130 million to NT$150 million investment this year. So for next year, it really depends on how the market will shape up and we would put the CapEx into our budget for next year.

Yes. Hi, Joseph and Ken. This is Rick on Daiwa. So two questions. I think the first question is talking about your EMS capacity for the last - for loading it’s going to be up mid-teens for Q4. Is that fair to assume when it comes to revenue base, it can’t be even higher than 15% because of higher ASP?

I wouldn’t be very technical about this. This is really just a reference for you to come up with your own thoughts on how EMS revenues would look like for next quarter. This just gives you a way to come up with the numbers. I think what I'm trying to say here is capacity stays flat and loading goes up, revenue should go up accordingly.

Okay, fair enough. Okay. So for your utilization rate for Q3, can you elaborate a little bit more? I know you're talking about your advanced versus the traditional, but in terms of tests came in bumpy and utilization rate for Q3, can you give some numbers?

Okay. In terms of wire-bonding, Q3, we are at the mid-70s, advanced at low-80s, testing we are about mid-70s.

I'm sorry, low 80s for…

Test. I'm sorry low-80s for advanced.

For advanced, okay. Okay. One last question I think because the recent issue between your company and [indiscernible]. Are you guys concerned about your customers trying to diversify the orders away to your competitors?

Well, I think people ask that same question to the other company as well, and they seem to haven't seeing anything happening on that front and so we are not saying anything abnormal happening now either, and I think things are going on track and whether we are looking at is really the normal business that we are going through.

Thank you very much.

Thank you.

Hi Joseph. First question is about what made EMS and ERT [ph]?

Please name.

This is Roland Shu from Citigroup. First question is for your EMS revenue with ERT [ph] in 4Q? Thank you.

In 4Q?

Yes. Because in September actually you have that multi-sales for EMS, and just I wondering will this momentum continue in 4Q and a possibility in ERT [ph] in 4Q? Thank you.

You mean on the marketing basis?

Yes.

I think it's pretty linear in the EMS business and what we are seeing is a normal pattern although with a different scale in this year. We are saying that there will be quite a bit of growth in EMS in the fourth quarter, as we mentioned here in the guidance.

Yes, definitely going to grow in 4Q. So can we assume that for the marketing circle be continued growth through end of this year?

Yes.

Okay, thank you. And how about for first quarter next year, since for our EMS business actually did climb about 25% in the first quarter this year and also first quarter last year.

Yes.

So how about for next year?

As I mentioned, we will see normal seasonality going into the first quarter next year.

So this means that 25% decline normal seasonality in EMS?

I don't have the numbers now. I'm sorry.

Okay, yes, thank you. Second question, if you look at this year and last year, I think for your - the logic outset total revenue actually under-grow - underperformed on foundries revenue growth. My personal opinion [ph] is that given your sales up for foundry actually since the company migrate to the DTH technology, so in the meantime they are able to increase the wafer ASP, therefore also like for your DTH application packaging. I think the unit price probably would be still [indiscernible] and I will think this is why you have the different thinking about that for the like-for-like ASPs change for OSAT in the past years?

I think the incurrence of technology in the OSAT world, I think the lifespan is much shorter, i.e. for each particle knob, you may have a much longer time to enjoy maybe what we call the premium pricing. But this is really not the case for ASAT [ph]. I think for OSAT, even if you bring out a copper, this will lapse the overall the pricing. I think really soon the competition will come in and then the customer would drive the price down a bit. For this year, we are expensing very similar price expense on quality basis. So no, I don't think we like-to-like with foundries.

From your ASP point of view for like-to-like ASP, did you see the ASP increase in the past years, 7% to 8% will be - it will decline a lot though.

It will go through a normally 10%, 15% drop on the relative basis.

10% to 15%?

Yes. That’s the typical pattern. Anything that's maturing and then it stop, see price drops. So the main - one of the main focuses that we have is continue to grow our efficiency even for the new products, like I mentioned for the SiP products. We also need to work very hard on bringing the cost down.

Okay. Thank you. Yes, and also TSMC, next year is going to make its InFO for 20 customers. And what this is impact to ASE and what you’re contesting on TSMC’s INFO technology?

Well, we do have our own fan-out technology as well, some in ‘16 some being developed. And I think that in terms fan-out, we are aiming at different target markets. I think the foundry really targeting on the very most advanced or high performance type of chips with limited customers. We are - our technology is really aimed at more broader customer base with may be low-to-mid end technology requirement. But it also - well, even for InFo itself, I think at the end of the day, when they start mass production there will still be some natural division of works between us and the foundries and how much business we would get we don't know, but that will happen with [indiscernible] and this will happen with wafer bumping as well. I don't think the foundries will ever get into –well I shouldn't say for them, but logically this is really not their core business. So I don't think we will have a head-to-head competition with foundries going forward. It will still be a complementary type of relationship with them. There will be some competition but I think most of it still be cooperations.

Okay. When do you think this collaboration or cooperation will happen?

According to our schedule is second half of next year.

Okay, thank you. My last question is then for one memory in Taiwan, I think invest in 2.5D, even 3D [indiscernible] they are able to make production next year. So what’s your view on this total addressable market for 2.5D, 3D or PSD [ph] and what’s your progress on this project? Thank you.

Do you want give it a shot? Well, I'm not that technical, so I'll leave this question to Tianyu Wu next time.

Okay.

But I think what we are seeing what foundry is doing is really the tip of the overall industry, and I don't think it will have a huge impact on us in terms of the overall can, and also that part of the business is really additional business from our perspective because most of the processes, be wafer level process and I think it’s only natural that the wafer foundries either way in terms of the technology development of that. What we need to do is really to put ourselves in sync with that technology roadmap and be ready to do whatever will we given to do once the divisional works is formalized.

Actually the next question will be coming from caller Szeho Ng from BNP.

Hi Good evening. After the investment in SPIL, would there be any change in your dividend policy for next year?

There is no plan of changing our dividend policy. I think in the past two or three years, we have been paying cash dividends to the level of 65% to 70%. I think we are not changing it at this point.

Okay, that's great. And then second question on the wafer bumping capacity. Could you remind us what's the latest there for 12-inch and 8-inch?

Wafer bumping for 18-inch, we have about 95K a month and for 12-inch we increased it slightly from 80K to 82.5K.

All right. And any expansion plan for next year for wafer bumping?

I think we will continue to increase in smaller steps but I think we will need to look at the overall situation to decide how much we’ll be putting in next year after the budgeting cycle.

I see, all right. Yes, because simply from the installed capacity ASE right now is smaller than the other tiny competitors. So just want to - like to see any plans to be more aggressive on that area?

Well, I think the overall capacity increase is really depending on the dynamics of our business composition and we will make our necessary investment accordingly.

Okay, all right. Okay. Thanks very much.

Thank you.

Well, do we still have one? Name and company.

Hi. Gokul Hariharan from JP Morgan. My first question is on SiP. You mentioned that one of the projects is still below your desired level of profitability or even potentially loss-making. Should we have to wait for a new model cycle for that to get better or are you seeing that getting better any time in the next three to six months?

I don't have any comment on that. I think this is something - it’s really something that's not want to comment basically. What we can do is really to look at the current situation and streamline the operation better and also to find ways to reduce our cost. I think we have been making progress on that in terms of cost savings, in terms of bringing up the yields, so that we can bring on the overall cost of it. In terms of actual business terms, I think we are working with the customers very closely, since we now actually have a better picture of the cost side of it. So in terms of business terms, when it comes to the next generation, I think we will have a better reference points to tell what's the - to decide what's really the right terms that we should be negotiating. And I think things are working towards that direction and we remain confident that eventually this will become most probably best for us as well because really if you put things into perspective, this could be a next billing unit business as well. So I kept calling it we are still at the investment stage and this is really what we are talking about here.

So on the SiP side, previously you have mentioned competitive landscape in your current projects is time Japanese vendors. As you go into next year, is it going to change significantly or it's going to be still the same names?

It's hard to say, but I certainly cannot preclude that there will be other competitors coming in. I think as I mentioned, customer will look for second and third source. I think our main goal is really to the stay as the primary supplier and get the first-mover advantage for the product that's coming on-stream.

Okay. Thank you.

Eric Chen from UBS. In term of the SiP, the profit about the new content, the SiP business and how about the competitive in Q4, we see the competitive decline or the competitive still impact [indiscernible] EMS business?

I really don't have comments on that.

Okay. So the cost efficiencies mainly from that ERA [ph], right?

From ERA [ph] and from the overall operations. We want to streamline the operations.

Okay. And you also mentioned you probably will negotiate with [indiscernible] next year.

I'm not saying that, I'm saying that for each generation there will be round of negotiations on the basic terms. I'm saying that now that we have a better picture on what the real cost will be, but when you're going through the first generation there is a lot of uncertainties involved, now after you’ve gone through a product, that we have a better idea of how the cost is associated with it. And so we have more data points coming out with the more suitable or accurate terms that we need to pursue.

Okay. So far do we have any visibility in term of the competitive usage and for the next generation product?

No

No? So, that means that varies some high [ph].

No, comments on that.

No comment even for the idle capacity?

I have no comments on a particular product.

Okay. Thank you. And by the way, for the - you talked about the EMS and you probably have full project with your major client and all of them you added primarily the [indiscernible].

No, I think three out of four were primary. One is we actually are latecomer. I think the - for that particular project, we were brought in to prepare ourselves for the next or even young generation, when it goes into a more sophisticated type of product.

Okay, thank you. And my last question regarding the cash position. Would you mind to go through your cash position gain in term of the ECB and the tender offer and how can you give it on consistent cash dividends, so we would either get idea?

Well, in terms of cash from third quarter - second quarter to third quarter, the cash balance - let me give you those numbers. At the end of the second quarter, we have NT$58.9 billion cash and cash equivalents. And at the end of third quarter, after we paid the NT$35 billion for the transaction, our cash balance is still at NT$45.6 billion. And if you look at our third quarter, you can see that EBITDA was about NT$500 million and the CapEx was only about NT$140 million. So we continue to generate free cash flow. In fact in third quarter, the cash flow that we generated is about NT$12 billion, except because of exchange movement, the number was reduced to about NT$10.5 billion. So going forward, I think not only that we can continue to support our operation. I don't think it really has been much of a pressure on us in terms of maintaining our dividends as well.

Okay, that’s very helpful. And I’m sorry, one more question. Regarding to your strategy in China, JECT [ph] very aggressive and then Nantong for 2Q was very aggressive and so do you have any of the big picture or the big plan given the strategy going forward, how you play the game and in China and overall?

Well, the competition is not just coming from China, I think from all over the place. I think everybody is - we have been in the industry for the past 30 years and we've seen all sorts of competition coming from all directions. I think China is getting more aggressive in terms of also on the OSAT. So regardless competition is given, and I think what we need to do is continue to drive up on technology, also to drive down on cost and try to continue to enhance our competitiveness going forward. But of course we have been saying that the Taiwan OSAT industry as a whole, should find ways to cooperate in the phase - to face the upcoming competition from China because it is really that are supported by the government. So the competition from that is different from just competing with the individual companies in the world. So I think Taiwan - we believe that Taiwan OSAT collectively will do hold the majority share of the world markets, but individually we're still smaller in scale. So I think one form or another cooperation among the OSAT is really very important to face the upcoming competition going forward.

Okay.

I think from our own efforts for the cooperation or collaboration is very necessary.

Okay. Thank you very much.

Thank you.

The first follow-up, I wanted to expand on the question about the debt position. That debt-to-equity that you quoted is higher but you're also now carrying the SPIL long-term investment. So I'm curious if you need to do equity financing, if you want to bring down that debt-to-equity, what factoring in that SPIL investments do you feel pretty comfortable with the amount of the cash and investment resource?

Yes, I think given the strong cash flow what we have been generating and also I don't know whether this is good or bad, I think in the next few quarters, I don't think the CapEx will be very, very substantial. So we believe we will have quite a bit of free cash flow coming in and we will use that to pay down debt.

But the timeline on the fan-out, you're going to target more of the mass-market solution for fan-out, how do you see that in terms of adoption or customer interest relative to flip chip? Do you see over next couple of years, flip chip will transition fan-out to the mainstream part of the market. Just curious how you see this mass market solutions?

What I know is that we are making investments in the fan-out, and I think aside from just investing into capacity for existing technology, I think we are also finding other solutions with better efficiency or better cost structure. And we are doing quite a bit of searching on that. But I think things will start to happen, we’ll sort of see some of our own capacity on some of our own solutions coming on stream. Maybe in the second half of next year, we will see some of very good results on that.

One housekeeping on this, so if you could give the contribution percent of total sales for ATM and consolidated for the third quarter - for the SiP?

SiP in terms of packaging is about 9% in the quarter. They will be the same in fourth quarter. SiP in terms of overall, it will be - it's 22% in the quarter going to above 25% for fourth quarter. It will be in range from 25% to 30% in the quarter, pretty much in line with what we projected at the beginning of the year. I think this year’s SiP overall revenue, we will achieve while we plan out to double next year's number.

And then the last question I have just on the materials business. It looks like it’s a lot quite a bit this quarter on memory. How is the medium-term, is there a temporary pullback in memory? I know we are back to the levels of that, or do you see opportunities to new substrate like more flip chip or other things to rebound this business?

Well, I don't think this is an industry phenomena, it's just our own selective customers going through some hesitance, and we are seeing that coming back in the fourth quarter. So it's a temporary thing.

Okay.

Hi, this is Andrew Chen from Yuanta right here. I have a pretty simple just top-down big question. This year has been a pretty interesting year. If you actually look at the M&A activities going on in the industry, you had JCET was flat and you also have Amcor on the J-device [ph]. Obviously I think for most industries, you can almost say that horizontal consolidation over time should be pretty positive for the industry participants, especially the leader. Do you agree that this is the right time to go and do you also think that ASP want to be part of it in terms of the horizontal consolidation trend?

We certainly think that with the competition landscape changing every day, we are seeing that - from a customer side, we are seeing quite a bit of consolidation as well, like Avago buying Broadcom, Intel and Altera and so on so forth. So that - our customer becomes bigger and given more leverage on less. We are seeing - at the same time, we are Chinese are very aggressive buying of flagship and buying of AMD’s back-end positions and we would expect to see more coming actually. So yes, I think some form or some level of consolidations is needed as we want to stay competitive going forward. I think it's really not a question of - I think what we are trying to avoid is one minus one. So if you look at Taiwan OSAT industry, one is that we need to expand our scale. I think individually, even ASE is the world's largest OSAT player but still it has less than 20% of the overall OSAT market comparing to TSMC. Their position is on the world difference. So we do need to form some sort of alliance or some sort of cooperations, so that we can better utilize our collective resources. And I think also very important item that is really to emerging SiP markets and that requires huge resource to build that businesses up, and we do see that as really the next big thing in the OSAT world and just a company of our size or any other company size it's really resource constrained to try to develop that part of the business in any meaningful way. So yes, I do believe that horizontal consolidation is something that's really necessary.

Just a quick follow-up. So it does seem that there is definitely a sense of urgency, so we should be a bit more aggressive, I guess, going down the line.

In what sense?

In terms of following this industry trend of horizontal consolidation.

Yes, I do agree with you.

Okay. And just a follow-up on the one plus one less than two. I think a lot of the market has been focusing on this, I just finding intriguing. In fact the way I look at it is actually the available flip chip capacity is where the advanced packaging capacities out there right now are quite limited, and it depends on the few players. So that’s also kind of the reason why some of the key players are consolidating. Probably for some customers, there is limited availability of sources they can go to. So of course, [indiscernible] probably is susceptible to share loss potentially during some integration, say for instance, during the time for JCET with OSAT. Come on, if you just look at the flip chip, do you really see that big of a high-risk in terms of that diversion?

No, I don’t - I think the real risk is one minus one. I don't think it really meaningful to discuss whether one plus one is greater or less than two. It is really - if we don't are working together, I think we are continuing to compete with each other, at the end of the day will be one minus one.

Thanks for your kind answer. Hopefully that message get conveyed some time down the line.

No questions? Well, thank you for attending the third quarter earnings release. See you in the fourth quarter.

Thank you.