Ken Hsiang - Head of Investor Relations Joseph Tung - Chief Financial Officer

Randy Abrams - Credit Suisse Bill Lu - UBS Securities Rick Hsu - Daiwa Capital Markets Steven Pelayo - HSBC Sebastian Hou - CLSA

Hello. I am Ken Hsiang, the Head of Investor Relations for ASE.

Welcome to ASE Group's Third Quarter 2016 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 a high degree of risk and our actual results may differ materially from these forward-looking statements. For the purposes of this presentation, dollar figures are generally stated in new Taiwan dollars unless otherwise indicated.

For the earnings release, I will be going over the financial results. Joseph Tung, our CFO, will be answering questions during our Q&A session.

Following the event, our VP in charge of Public Relations, Eddie Chang will be addressing the media in Mandarin Chinese. Before we get into the numbers, I would like to give you a quick update on our transaction with the SPIL.

As of September 19, our case was accepted as registered with the TFTC. TFTC indicated they would be holding a hearing on the transaction before its decision on the proposed transaction.

As reported in our last earnings release, we have also filed our case with China's antitrust authority and are awaiting the official registration of our filing. ASE has determined that the proposed transaction does not trigger a U.S.

HSR filing, but notes that even without such HSR filing, the U.S. FTC retains the ability to review such transaction.

The U.S. FTC is investigating the said transaction and ASE is fully cooperating.

We have also completed our purchase price allocation assessment as required by IFRS in which the purchase price is allocated into the various assets and liabilities acquired. The allocation effectively results in an asset revaluation which will incur a non-cash expense of TWD281 million for 4Q 2015 and 1Q 2016 and TWD351 million for 2Q 2016 and each quarter thereafter.

During this time, such amount will be netted against SPIL's earnings reported in our non-operating section. From the business perspective, during the third quarter, we saw strong performances from all of our business units.

For IC ATM business unit, our factories were well loaded with bumping in certain wirebond packaging lines near or at capacity. The IC ATM business strength was driven primarily by smartphone-related products and a seasonal ramp in our SiP-related business.

For IC ATM, the third quarter generally represents the seasonal peak as Android-related devices start ramping down and as other devices go through their product launch cycles. We continue to make progress with our SiP rebalancing efforts.

We are beginning to see some of the impact as we are seeing improved financial results within our wearable SiP product. We will continue to shape our SiP margin profile.

We will also continue to focus on SiP in which we have a technological advantage. For our EMS business unit, our sales came in slightly behind where we were expecting.

Certain product order flow did not materialize to the extent expected. However, that didn't have a negative impact on our results with the EMS business unit delivering healthy gross and net profits.

During the quarter, our EMS unit had more favorable product mix along with an improved margin profile relative to our SiP product. With such, we were able to improve our gross margin and gross profit for the quarter.

Page 2, Group quarter-over-quarter consolidated P&L. On a fully consolidated basis, the third quarter, the Company delivered fully-diluted EPS of TWD0.64 and basic EPS of TWD0.72.

Our packaging and testing businesses were both up 11%. Our direct materials business was up 6%.

Our EMS business grew 25%. We booked other revenues of TWD0.1 billion related to real estate sales versus, sorry, TWD0.3 billion in the second quarter.

Total revenues for the consolidated Group increased to TWD72.8 billion. Gross profit increased 15% from TWD12.3 billion to TWD14.1 billion.

Consolidated gross margins dipped 0.2 percentage points from 19.6% to 19.4% as a result of increased EMS product mix. Operating expenses edged up by TWD0.4 billion.

Our operating expenses as a percentage of sales decreased to 9.2%. Operating profit for the third quarter was TWD7.4 billion, up TWD1.5 billion from TWD5.9 billion in Q2.

Operating margins increased 0.7 percentage points from 9.5% to 10.2%. During the third quarter, we had a net non-operating loss of TWD0.6 billion as versus a net non-operating gain of TWD0.2 billion the previous quarter.

The current quarter's non-operating loss includes the following; ECB loss of TWD0.3 billion; net foreign exchange loss of TWD0.1 billion; our estimation of SPIL's contribution for the current quarter of TWD472 million. This amount consists of our percentage ownership of SPIL's earnings less incremental depreciation and amortization recognized as a result of revaluing SPIL's assets due to the purchase price allocation mentioned previously.

Net interest expense of TWD0.5 billion and the remainder is related to asset disposition gains and losses and other non-operating costs. Pretax profit for Q3 was TWD6.9 billion, up 12% from TWD6.1 billion in Q2.

Income tax expense for Q3 was TWD1 billion, down from TWD1.5 billion in Q2. The effective tax rate normalized as the tax on undistributed earnings was recognized in Q2.

Net income for Q3 was TWD5.5 billion, up TWD1.2 billion from TWD4.3 billion in Q2. Net margin improved to 7.6% from 6.9% in Q2.

Page 3, Group quarterly results on a year-over-year basis. Comparing the current quarter's results versus the same quarter last year, our packaging and test businesses grew 13% and direct materials grew 6%, while EMS declined by 14%.

On a year-over-year basis, our consolidated net revenues were effectively flat. With our product mix being more heavily weighted towards IC ATM, our gross profits were up 9% with gross profit margins improving 1.6 percentage points from the previous year.

Operating profits were up 17% with operating margins improving 1.4 percentage points from 8.8% to 10.2%. Total non-operating gain last year of TWD1.4 billion versus TWD0.6 billion loss this year.

The variance is almost entirely attributable to an unusually large non-operational gain in the third quarter of last year related to our ECB and foreign exchange. And as a result, net profits from a year-over-year perspective were down 14%.

Page 4, IC ATM P&L. Please note the inter-company revenues including the SiP technology business performed by our IC packaging business unit on behalf of our EMS business unit are eliminated during consolidations.

Our IC ATM net revenues improved by TWD4.5 billion or 12% during the third quarter to TWD43 billion. Revenues for our IC packaging, testing and direct materials businesses increased 12%, 11% and 15% respectively.

NT dollar appreciation had a 1.25% unfavorable impact on revenue with average exchange rate of U.S. $32.402 to U.S.

$31.781. Gross profit was up 15% or TWD1.4 billion to TWD11 billion.

We had some unexpected items that impacted our gross margin for the quarter as follows. Typhoons hit Taiwan causing increased labor costs and overall factory efficiency disruptions.

And two, both NT dollar and gold price were stronger than anticipated. We estimate these items impacted our IC ATM gross margin by at least 0.9 percentage points.

Even with such impacts, gross margin improved 0.7 percentage points. The gross margin improvement was the result of relatively lower D&A plus rental and lower - and labor to a smaller extent, offset by relatively higher raw material expenses.

Raw materials were TWD10.0 billion in Q3, up TWD1.6 billion with - and was 23.3% of total net revenues, up 1.3 percentage points compared to last quarter. Labor cost was TWD8 billion in Q3, up TWD0.6 billion and was 18.7% of net revenues, down 0.7 percentage points compared to last quarter.

D&A plus rental was TWD6.5 billion in Q3, up TWD0.1 billion and was 15.2% of total net revenues, down 1.4 percentage points compared to last quarter. Factory supplies were TWD4.2 billion in Q3, up TWD0.4 billion and was 9.7% of total net revenues, down 0.1 percentage points compared to last year.

Utilities were TWD1.5 billion in Q3, up TWD0.2 billion and was 3.6% of total net revenues, up 0.2 percentage points. NT dollar appreciation had a 0.6 percentage point unfavorable impact to gross margins.

Gold price movement during the quarter had a 0.13 percentage point unfavorable impact to IC ATM gross margins. Operating expenses increased from TWD4.6 billion to TWD4.8 billion, while operating expense percentage declined 0.9 percentage points to 11.1% from 12%.

Operating margin for the third quarter was up 1.5 percentage points to 14.4% from 12.9% in Q2. Operating profit was up 25% or TWD1.2 billion to TWD6.2 billion.

Page 5, IC ATM year over year. Here you can see our year-over-year comparison for our IC ATM business.

Our packaging and testing businesses were up 7% and 13% respectively. Our direct materials business was down 1%.

During the current year's quarter, raw materials and factory supplies increased as a percentage of COGS while D&A plus rental and utility decreased. Gross profit was up 3% while gross margin was down 1.2 percentage points.

Gross profit margin decline from a year-over-year comparison was principally the result of NT dollar appreciation and typhoon issues discussed earlier, along with the product mix with this year having higher module and flip chip content. Operating income is up 10% from TWD5.6 billion to TWD6.2 billion.

Page 6, our packaging operations. In Q3, our packaging revenue increased 12% sequentially and 7% year over year to TWD34.8 billion.

Our packaging gross margin of 22.2% increased 0.7 percentage points sequentially and down 2.6 percentage points year over year. The sequential margin improvement was primarily caused by higher loading as compared to the second quarter.

This resulted in higher revenues in a semi-fixed cost structure. This impact was offset by relatively higher raw material costs as a result of product mix and higher costs due to typhoon holidays and NT dollar appreciation.

Raw materials were TWD10.7 billion, up TWD1.5 billion and as a percentage of sales was 30.8%, up 1 percentage point of sales. Labor was TWD6.2 billion, up TWD0.5 billion and, as a percentage of sales, was 17.9%, down 0.7 percentage points of sales.

D&A and rental expenses were TWD4.5 billion, up TWD0.04 billion and as a percentage of sales was 12.9%, down 1.3 percentage points of sales. Factory supplies were TWD3.3 billion, up TWD0.4 billion and as a percentage of sales was 9.6%, up 0.2 percentage points of sales.

Utility was TWD1.1 billion, up TWD0.2 billion and as a percentage of sales was 3.1%, up 0.2 percentage points of sales. During the quarter, capital expenditures were $112 million composed of wafer bump, fan-out and copper pillar equipment of $64 million, common and SiP equipment of $47 million and wirebond-related equipment at $1 million.

During the quarter, we added 10 and retired 25 wirebonders. We exited the quarter with a total of 15,905 wirebonders in operation.

8-inch bumping capacity remained unchanged at 95,000 wafers per month and 12-inch bumping capacity including fan-out and copper pillar increased 11,000 wafers to 111,000 wafers per month. Page 7, testing operations.

Test revenues of TWD7.2 billion were up 11% sequentially and 12.5% year over year. Test gross profit margin of 38.9% was up 2.1 percentage points sequentially and 2.8 percentage points year over year.

The changes in gross margin were principally the result of higher seasonal loading in a semi-fixed cost environment. Overall cost of services for test increased TWD0.3 billion to TWD4.4 billion.

Our testing utilization rate improved to around 80%. CapEx for the test business was $57 million in Q3.

We added 126 and retired 30 testers during the quarter. At the end of Q3, our total tester count stood at 3,725 testers.

Page 8, materials. Revenues for our materials business of TWD2.3 billion were sequentially down 5.4% and up 20% year over year.

During the quarter, TWD805 million was from sales to external customers, up 6% as compared to Q2. Our internal self-sufficiency rate decreased to 27% from 37% by value.

Gross margins were sequentially down 4 percentage points to 14.8%. Page 9, IC ATM market segment.

During the third quarter, the market segment share stayed relatively unchanged representing all market segments performed relatively smoothly. Our communications market segment share increased from 52% to 53%.

Our computing market segment remained flat at 12% and our automotive, consumer and others declined to 35%. Page 10, EMS business unit.

During the third quarter, revenues for our EMS business unit were sequentially up 25% to TWD31.2 billion from TWD24.9 billion. Revenues year over year were down 14% as compared to TWD36.2 billion in Q3 of 2015, primarily as a result of lower SiP revenue during the current quarter.

Gross margins decreased sequentially 0.3 percentage point to 10%. Margins were stronger than anticipated, primarily as a result of product mix and rebalanced SiP-related business.

EMS gross profit increased to TWD3.1 billion. CapEx for our EMS business unit was $10 million during the third quarter.

Page 11, EMS business segment mix. During the third quarter, our communications and consumer product segments increased their segment share, 5 and 2 percentage points respectively while our computing and industrial and automotive EMS segments decreased segment share 4, 1 and 2 percentage points respectively.

The moves are largely in line with our SiP product cycle seasonality. Page 12, balance sheet.

At the end of the quarter, we had cash and cash equivalents and current financial assets of TWD39.6 billion, decreasing from TWD40.5 billion the previous quarter. We also had interest-bearing debt increase from TWD110.4 billion in Q2 to TWD119.9 billion in Q3.

Our investment in SPIL of TWD45.6 billion is recorded in investments equity method. As of September 30, 2016 total unused credit lines amounted to TWD169.5 billion.

EBITDA improved to TWD14.7 billion from TWD14 billion during the current quarter. Page 13, CapEx.

Capital expenditures for the third quarter totaled $184 million, of which $112 million was used for packaging, $57 million for testing, $10 million for EMS, and $5 million for interconnect materials. EBITDA on a US dollar basis was $463 million for the third quarter.

For the industry and us, smartphones continue to be the product that is driving overall unit volume. The mass adoption of smartphones has brought about new industries for servicing such devices such as music streaming and payment processing services.

However, if we look a little bit beyond the horizon, we are excited as even more and smarter electronics blend into the home and automobile. We are also excited by the sheer amount of data that will be involved.

Data will be generated, transferred and processed once and then multiple times over. This means bigger, better and more powerful hardware.

This means chips and subsystems connected in different ways, as with our SiP offering. And it means the need for higher density I/Os necessitating more bumping, copper pillar and fan-out processes.

The future continues to look bright for ASE. For the near term, we continue into the fourth quarter with capacity tight in certain product lines, which is typical for this time of year.

The overall environment still looks reasonably healthy. We are carefully monitoring for signs of overbilled and double booking and, at this point, nothing gives us any specific concern.

We believe that the coming quarter should follow in a seasonal pattern for both IC ATM and EMS business units. Our IC ATM capacity should stay flat.

Our IC ATM blended utilization rate should stay flat to declining 5%. Our IC ATM gross margin should be similar with the prior quarter.

Our EMS capacity should stay flat. And our EMS blended utilization should improve by 10% to 15%.

Our EMS gross margin should be consistent with our gross margin during the first half of 2016. Time for Q&A.

Any questions? No questions?

Randy? Name and company please.

Okay, yes. I'm Randy Abrams from Credit Suisse.

First question on the CapEx. Year to date, it's about $550 million.

Just want to see your full-year CapEx now. It looks like it's tracking a little bit below depreciation, but just want to get a sense on fourth quarter.

And if you could give an initial view on 2017 priorities.

I think the overall CapEx for the year remains the same as we earlier reported. I think it will still be between the total D&A and our CapEx last year.

As far as fourth quarter is concerned, I think it will be a notch lower than $184 million we spent in quarter three. For next year, we're still in the reviewing stage and after we have come out with the annual plan for next year, we will have a better picture on what kind of CapEx budget we're going to make.

Okay. I wanted to ask a question on the materials business.

The internal substrate declined from 37% to 27% and the gross margin by a few points. If you could talk about the trends in that business, anything going on to drive those changes?

Well, I think what happened in Q3 is a bit unique. Because in the period we experienced a bit of a hiccup on our yield.

That is already resolved and I think we are back on track. I think the normalized self-supply ratio should be around 33% and above.

Okay. And the last question just on the gross margin guidance.

The EMS margin, it seems like it held up again at about 10%. Were there more of these engineering project fees in there to keep the margin holding up?

Like I think in the second quarter at 10% there was engineering to drive the better margin. But just what drove the better EMS margin.

I think by and large the better than expected margin at EMS is really due to product mix change. But there is some engineering fees that we've collected in the quarter.

But I think aside from product mix change, I think another very important factor is really the - as we see results from our rebalancing the SiP business at the EMS level.

Okay. And the follow-up to that, could you give a view then, the SiP business, how it's looking now for projects?

If you see any new ones? Just whether your core customer or diversification into additional customers, how that's playing out now?

Well, for the existing four projects, what used to be existing four projects we had, we're maintaining two of them, discontinuing one and another one we just kind of exited because we don't feel we have that much value add into it. But we did add one more project and that is starting mass production starting from quarter three.

Going forward, I think the rebalancing effort is continuing. We are positive on the increasing trend of modularization so we believe that there's going - it's going to lead to quite a bit of SiP opportunities going forward.

And we do have multiple engagement with multiple customers at this point in different phases. But I think right now the focus will still be on - we want to stay selective in choosing these projects.

We want to focus on the ones that has a healthy financial profile as well as we do have a technology advantage on it.

One follow up. For the two that you're maintaining, do you have confidence for your expectation to continue those through the next generation?

Are those the type you want to keep with, or when you say rebalance continue, there could be changes with the first two?

Yes. Well I think we will continue to stay very, very closely engaged with the new products coming in.

But like I said, unless the product is financially justifiable and also we do have a technology advantage in it, we want to be cautious in putting in additional investment into it.

Thank you.

I think we'll go with - name and company please, sir.

Yes thanks. Bill Lu from UBS.

Could I start with a clarification? Ken talked about the purchase price allocation.

Was it TWD281 million for 4Q and 1Q of next year? Is that what you said?

4Q and 1Q; 4Q last year, 1Q this year.

And then TWD351 million for the subsequent quarters?

Correct.

How long does that last?

That number will probably last until the combination is completed.

Well, let me elaborate a little bit. I think TWD281 million is because at that point we hold 25% of the - of SPIL.

And the number increased to TWD351 million because we raised our ownership to 33% and above. And that number we should maintain as we continue to hold 32%.

Upon closing of the deal, when we will be holding 100% of the company, then there will be a revaluation. We will redo the PPA and the final results will depend on the then SPIL's financial condition.

Okay, I understand. We're now a few months into the two sides going into this deal.

I think I asked you this last quarter as well, but in the last several months, as you've talked to the customers, can you let us know what feedback you're getting as far as people looking at your combined market share? Do you think the pricing environment has got a little bit better because of this potential move and what other feedback are you getting from customers?

No, I don't think it has any - the pricing has anything to do with this transaction itself. I think with or without the transaction, customer does have options.

And there's plenty of alternative capacity around. So everybody is still on the same page and competing on, not just pricing, but also on technology, quality and so on and so forth.

So I don't think that's - in terms of customers' reaction to it, I think there's a mix of different customers. And some of them are very, very supportive to it.

Some of them may have some hesitations. But all in all I think this - the end game is that we will continue to have a healthy dose of competition among the two companies, although there will be collaborations on resources, and therefore improving our overall efficiency, which is to the benefit of our customers, eventually.

Great. On the SiP business, going into next year there are some talks of leading handset makers going to substrate like PCB, which I understand is kind of a modularization of the HDI board.

Do you consider that to be an opportunity for SiP? And, if so, what is your advantage over the other competitors?

Actually yes. When the substrate gets more complicated like that, I believe that actually is a trend towards SiP, right?

It gives system houses more of a capability to put multiple devices together. And substrate is just one of the tools, so to say, in part of the SiP-ization of electronics.

I think that's the very reason why we have this TDK venture with - which we want to develop, what we call the embedded substrate. Which is going to be a very essential part of the overall SiP effort.

Yes, that was my follow-up question is, to capture that business, do you have all the necessary pieces now in terms of the substrate, the EMS; what else do you need?

What else do we need? I think the whole technology is evolving, right?

We just need to keep up with the technology development and really have a very, very close engagement with our potential customers. We need to put the technology roadmap in sync with them.

We also need to have a better feel of what the going trend is. The whole SiP business is a new effort actually, so we are still on a learning curve.

And some of the projects we did pay some prices on it. So I think one of the efforts that we need to continue is really to find the right business model for it.

And that's what the rebalancing, what we call the rebalancing of SiP business is all about.

One last question, on the wearable SiP, you said that the margin has improved from last year but I think that was a pretty low bar. Is it close to the division average now, or can you just give me a sense for where it is?

Well I can't give you a direct comment on it. I'm saying - what we're saying is through different efforts, including the change of business terms, our capacity, so on and so forth, I think we are seeing very, very positive results from the effort.

And that's really showing in the overall EMS margin in the third quarter.

Thank you.

Rick over there. Name and company, sir.

Yes, hi. This is Rick Hsu from Daiwa Securities.

I've got a few questions here. Again a housekeeping question.

Regarding your bumping capacity, I think Ken talked about that number; I missed that part. Can you remind me, the bumping capacity for Q3?

At the end of fourth quarter last year we had 12-inch bump of 82,500 pieces.

So 8-inch -

That's now grown to 111,000. I think by end of the year it will grow to 116,000, 12-inch.

8-inch I think will remain the same.

8-inch is at 95,000.

95,000 as of Q3, right?

Correct for 8-inch. 12-inch is 111,000.

And 8-inch will remain the same toward the end of this year, am I right?

One more time?

8-inch will remain the same toward the end of this year?

Yes. I don't believe we have plans to increase 8-inch right at this time.

Okay, cool. What about the UT, utilization rates for Q3 across the board of the –?

Q3 packaging is around 85%, which is pretty much at capacity. And for tests we are at about 80%.

I think going into Q4 it will be a little bit lower, as we have guided.

Okay. So that will lead to my third question.

If I recall the guidance from TSMC and UMC, I think roughly across the board for the foundries who have already reported, about a flattish Q4. How come there's a gap between your IC ATM guidance and the front-end foundries?

Well they are not the only business source that we have. So we'll have a - there's a diversified customer base and each has different situation or condition.

But all in all I think we're just experiencing a normal seasonality. In our business we typically peak at third quarter.

I would say from the - from what we're seeing for fourth quarter, it's actually a quite healthy fourth quarter for us.

Right. One last question.

Can you give us some more color about your - so like InFO development.

I think the overall fan-out, I think that's what you're referring to?

Right, yes.

I think our strategy is still to focus on what we call mass-market type of solutions and lower-cost solutions. So from that, I think, right now we do have some capacity and some mass production on fan-out, but not yet at a meaningful level.

But going into next year the investment - both investment and revenue growth will be very, very substantial. I think we do see quite a bit of - quite good opportunity in that area.

I think the difference between us and TSMC or InFO also is like what I just said, our focus is really on multiple customer and mass market, with lower-cost solutions, which includes maybe today the 12-inch bumping capacity that we're installing and increasing. And also going forward there will be - the Deca effort will also start to play out.

That will be a panel-based solution which is - we believe it will give a much better cost structure to our customers as well.

Okay, thank you so much.

Thank you.

Our next question will come from Steven Pelayo who is online.

Yes hi. I guess I get three primary long-term questions from investors on ASE and their threats.

So people are nervous about the China threat, nervous about the turnkey threat from the foundries consuming more, keeping more packaging internally. And they're nervous about the combination and then second-sourcing risk.

In fact I think Amkor has suggested they've been gaining share on the second sourcing. So I'm wondering if you could talk a little bit about each one of those potential headwinds and hold our hands a bit here just to suggest if this is actually something we should be worried about.

Well I think the - first of all, the China threat I think China threat is no different from any other threats. This is a competitive industry and we're in it.

And so far I think we've done pretty good in competing with all the major competitors in the industry. I think the name of the game going forward is really economy of scale, the technology investment that we will continue to make.

And I think also one particular strength, or competitiveness, that we have is really the - our footprint in the SiP business. I think although we're in a rebalancing phase, but we still are very, very confident and positive about the going trend in this business.

We do have the unique and also the most advanced foundation for - to capture that business opportunity. China; they have been aggressive.

They have - funding is certainly not a problem for them. But I think we have 30 years, 30-some years' experience behind us.

And the technology, or the work experience and so on and so forth, the overall management know-how, we still have the leading position. And we will continue to leverage on that to expand our competitiveness going forward.

In terms of foundries going into back-end, I think what TSMC is doing - or I shouldn't even say the name - but what the foundry is doing is really to offer a total solution as they go down the mass of their wafer. All the chips that are made needs to be packaged and tested.

And if the most advanced technology in terms of packaging is toward more wafer-level type of processes, I think it's only natural for the foundries to take the lead on that. So once the technology is well defined and matured, I think same thing happened like the wafer probe, or bumping, I think there's eventually going to be eventual - natural division of works among the OSATs and the foundries.

So I don't think we are in such a pure competition mode with the foundries. It's more like a collaboration among the two.

I'm not sure I understand your third question about second-sourcing?

I guess there are some concerns or fears that the combination of ASE and SPIL creates an even larger player out there and takes out maybe somebody's second source and so then they have to go looking elsewhere.

Well I think with or without the transaction, or with or without the combination, I think the competition is always there. As I said, the customer always has options to go somewhere else and there's always plenty of capacity to go about.

And besides, the two of us put together, it's not going to be - size-wise we're not going to be a monster. So I don't think that's going to change the competition landscape that much.

So we will just continue to do whatever we do. We will continue to stay competitive.

And in the short run, some of the customers may have some concerns that they want to diversify it a bit more. Eventually the real swing factor is really how well a job you can do for your customers; cost-wise, quality-wise, technology-wise.

And that's something we will be focusing on. And eventually, as long as we create value to our customers, I don't think that will be a that big of a concern going forward.

I have a little to add -

Right, and just one follow up, if I could.

Steven?

On China, could you help me understand what percentage of revenues it is for you today? Maybe more importantly, specific to Chinese fabless and system houses, what percentage of revenues is that for ASE today?

Right now I think our business from Chinese customers is about 4% of our overall IC ATM business. And in terms of production, our China sites has about 15% of our overall.

Steven just a follow-up on -

That 4% number; can it double or what do you think it can do?

Steven hello? Are you there?

I wanted to -

Yes. I was asking -

Yes. I wanted to reply to one of your previous questions.

We are actually capacity constrained at this time. So that could be a rationale for the comment that you're getting from a competitor of ours.

Fair point. I've dropped off guys.

They cut me off.

No, you're still on. We can hear you.

Oh, sorry guys. I had a follow up there.

China's 4% of IC ATM, your thoughts for a year from today; second half of next year, do you think that this could grow significantly to maybe even approach 10% of revenues?

I don't want to comment on that.

And that's it from me. Okay, fair enough.

Thanks guys.

Thank you. Do we have any more questions in the room?

Thank you. Sebastian Hou from CLSA.

First question is, Joseph, you mentioned about in your EMS - or Joseph or Ken, you mentioned about your EMS business didn't perform as you expected because some product didn't materialize as you had earlier expected. So can you give us some - give me more colors on what happened exactly and what type of products or application was that?

I think what we said is our third quarter EMS revenue came a little bit short from expected. The main reason for that is one particular - or certain products, the order volume was less than expected.

Okay. So that is also the reason leading to your better-than-expected gross margin in EMS segment?

Partially, yes.

Okay.

Although more so on the existing - aside from the SiP business, the so-called higher margin product mix that's getting higher.

So is that product SiP related?

No, like I said, outside of SiP.

Okay. And do you expect that - because you said that it came shorter than expected in 3Q - is that because of the production timing impacts, or just because of the total, lump-sum volume of the order is smaller?

So do you expect that to pick up in 4Q, or no, it's just total order is smaller?

I can only say for Q3, yes, the order volume is less than expected. I'm not going to speculate on quarter four.

Okay. And can you give us more idea or colors about your SiP business outlook for next year?

Earlier you mentioned about you dropped one project because the profit doesn't make sense, but you add a new project. And how do you see this new project will materialize in next year in terms of how that will progress, or progress in the next generation of the - yes, in next generation of phone or et cetera?

I'm leaving this to Ken.

Can you repeat your question?

Yes. In terms of your SiP outlook for next year, you're adding one project, as you mentioned, this year.

So can you give us more colors on how this project will go into next year?

This project?

Yes.

I don't think we're going to comment on specific projects at this time, especially in this context here.

Well, I think so far this particular project is - meets our criteria. And so I think it's going to be maintained.

And how about the other ongoing projects that haven't really contributed to your revenue right now. Do you see more of the SiP product or projects that will contribute to your revenue next year, total?

I think the effort is continuing in terms of engaging with multiple customers and multiple devices. Really we need to see how things go.

Right now I don't think we can make a prediction on that. I think the effort will continue.

Okay. But not much visibility, can I say that?

Like I said, we have multiple customers, multiple projects in different phases. Some are nearer mass production, some are still at the design stage.

But having said that, it really depends on how the market will go and to see how that business will grow to be like.

Thank you.

Any additional questions? No questions online also, so thank you very much for coming to the third quarter.

Thank you.

See you next quarter.