日月光投資控股 (ASX) 2018 Q2 法說會逐字稿

Hello.

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

Welcome to our second quarter 2018 earnings release.

All participants consent to having their voice and questions broadcast via participation of this event.

Please refer to Page 2 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 today's event, I will be going over a few key accounting treatments and then the financial results.

Afterwards, we will have a Q&A session with Dr. Tien Wu, our COO; and Joseph Tung, our CFO.

Following the event, Justin Yang will be stepping in this time for Eddie Chang.

He will be available to address the media in Mandarin Chinese.

This is, of course, the first earnings release for the new holding company.

It took a monumental effort by our collective teams within ASE and SPIL entities to help make this happen.

A special thanks goes out to all of the people involved.

With the formation of the combined entity between ASE Group and SPIL, there are a few less common accounting treatments, which probably require more in-depth explanation.

We believe it is important for our shareholders to have the opportunity to understand the impact of these accounting treatments.

And as such, we have prepared a couple of slides to help with this.

Page 3. Purchase price allocation.

First of which is a more advanced accounting topic, purchase price allocation, otherwise known as PPA.

PPA is the allocation of our purchase price of SPIL into the various assets and liabilities acquired from SPIL.

The purchase price allocation process increases the book value of assets acquired to an appraised fair market value level.

It is this write-up of assets that generates increases to ongoing depreciation, amortization and rental expenses.

For example, if SPIL had a piece of fully functional equipment that was fully depreciated, 0 book value, such equipment would be written up to an appraised fair market value.

Once the equipment has book value again, it reinitiates the accounting depreciation process.

This appraisal process is performed on all identifiable assets from SPIL.

After all available assets of SPIL are reappraised up to their corresponding fair market values, the remaining amount of the purchase price is then recorded as goodwill.

The end result is that purchase price allocation provides a more properly classified and valued balance sheet.

So for us, the total purchase price of TWD 163 billion is allocated into 3 parts.

First, TWD 78 billion of the original book value of SPIL.

Next, TWD 51 billion of additionally identifiable assets that are marked up to fair market value.

And finally, TWD 39 billion of goodwill.

From the TWD 51 billion of additionally identifiable assets, we generate additional PPA depreciation, amortization and rental expenses.

We have included here the quarterly depreciation amortization schedule of PPA for the next few years.

The amount of PPA depreciation and amortization included in this table will be recorded in operating expenses, TWD 253 million per quarter.

You should also note there is a significant drop off in 2020.

I promise you it gets easier from here.

Page 4. Revaluation gain.

Prior to the close of the transaction, ASE Group owned approximately 1/3 of SPIL.

The average adjusted cost of the shares was TWD 43.9 per share for a total adjusted cost of TWD 45.5 billion.

On April 30, 2018, our shares are treated somewhat similarly to other shareholders of SPIL.

They are revalued to TWD 51.2 per share, or a total value of TWD 53.1 billion.

To the extent of the revaluation, we have to record a gain for the difference.

The revaluation gain on SPIL shares in total was TWD 7.6 billion.

The gain is a onetime noncash gain and recorded in nonoperating income.

Finally, we would like to note that because ASE Holdings was jointly formed on April 30 during the middle of a quarter, ASE Holdings' second quarter results include a full period of ASE and a stub period of SPIL.

Specifically, for April, 33% of SPIL's net income is recorded in net operating investment income.

But however, for May and June, the consolidated results of SPIL are fully included.

We have included this set of results and labeled them as legal entity basis.

We have also provided a set of results, which combines SPIL and ASE on a retroactive basis, simulating the inclusion of PPA and interest expenses while excluding the aforementioned revaluation gain and transaction fees and expenses.

We have labeled this set of results as pro forma basis.

For the second quarter, we are very busy getting ready to be ASE Holding.

We encountered a lot of macro political noise with potential trade tariffs between this country and that.

We saw the potential failure of a major Chinese handset maker come and go.

We also saw virtual currencies dropped to near-term lows, slowing potential mining activity.

And despite it all, we saw a smooth second quarter ramp-up of business.

We believe the industry and demand for electronics remain healthy despite the macro uncertainties.

With that, let's start the financial review.

For those of you familiar with ASE Group's previous earnings release, you may notice that we have attempted to simplify our presentation.

Our results will contain the results of 2 business units, ATM and EMS.

Combining our businesses of assembly, tests and materials into a unified entity aligns with how our customers increasingly demand a unified turnkey solution.

Removal of the IC from IC ATM signifies further the broadening of our assembly, test and material services beyond just integrated circuits.

So Page 5 here.

For the legal entity with 3 full months of ASE Group and May and June of SPIL, the legal entity recorded fully diluted EPS of TWD 2.69 and basic EPS of TWD 2.70.

Sales was TWD 84.5 billion with a gross profit of TWD 13.7 billion and a gross margin of 16.2%.

Operating profit was TWD 5.4 billion.

And after the nonoperating revaluation gain of TWD 7.6 billion and tax of TWD 1.3 billion, net profit was TWD 11.5 billion.

For lack of comparability reasons, there's not a lot of productive commentary we can add to the legal entity results.

Please feel free to go through them at your leisure.

If you should have any specific questions regarding the legal entity results, you may address them to us during the Q&A or you can contact us after the earnings release.

Page -- skip to Page 7. So the EMS P&L here.

During the quarter, we saw a number of minor revenue revisions from a number of customers.

This resulted in slightly lower-than-expected revenues here.

EMS revenues of TWD 30.5 billion were up 6% quarter-over-quarter and 8% year-over-year.

Our gross profit improved TWD 0.2 billion.

We came into the quarter believing that we could exceed our first quarter gross margin, but given the slight revenue short-off -- shortfall, our EMS business unit's gross margin remained at 9.4%.

EMS gross profit was also up 6% quarter-over-quarter, but was down 8% year-over-year.

The change here is primarily the difference in product mix.

EMS's operating profit was down 14% quarter-over-quarter and 29% year-over-year.

In addition to product mix, our EMS factories are in the process of ramp-up expansion.

The decline in operating income is in large part related to extra costs incurred to ramp EMS capacity.

Page 8. Here, we have the pro forma P&L for the consolidated holding company.

We've tried to prepare a fairly simplistic transparent pro forma P&L.

To generate the historic periods, we have added the historical P&Ls of each of the 2 entities on a retroactive basis.

Then we added PPA and interest expenses related to the transaction, as if those expenses existed in such periods.

We also removed relevant transaction fees and expenses.

And further for the second quarter, we excluded the revaluation gain of TWD 7.6 billion.

Aside from these, we made no further adjustments.

On a pro forma basis, for the second quarter, we had net revenues of TWD 91.8 billion.

This represents a 9% increase quarter-over-quarter and a 6% increase year-over-year.

Gross profits were up 21% quarter-over-quarter.

The quarter-over-quarter gross profit improvement was driven almost entirely by the ATM business unit.

While on a year-over-year basis, consolidated gross profit remained relatively unchanged with an increase from the ATM business offset by a similar decrease from the EMS business.

Gross margin improved 1.6 percentage points on a quarter-over-quarter basis to 16.1%.

This was driven by improved seasonal loading from our ATM business, offset in part by lower margins from our EMS business.

Gross margin declined year-over-year by 0.9 percentage points as a result of lower EMS gross margins and currency impact from our ATM business unit.

Operating expenses were TWD 9 billion, up TWD 0.9 billion sequentially.

OpEx percentage was 9.9%.

We expect our OpEx percentage to start declining during the next few quarters.

Operating profit improved TWD 1.6 billion quarter-over-quarter and stayed relatively flattish year-over-year at TWD 5.7 billion.

Operating margin improved 1.4 percentage points quarter-over-quarter and declined 0.4 percentage points year-over-year.

Nonoperating expense on a pro forma basis was TWD 0.5 billion.

Interest expense within nonop was TWD 1.1 billion, offset by net FX income and financial instruments.

Tax expense for the quarter was TWD 1.5 billion on a pro forma basis.

The pro forma basis effective tax rate was unusually high because of ECB conversion impact.

On a pro forma basis for the second half of 2018, we expect our ongoing effective tax rate to be approximately 20%.

Net income for the second quarter was TWD 3.5 billion.

This represents a 392% sequential improvement or TWD 2.8 billion.

On a year-over-year basis, net income declined TWD 4.3 billion from 2017 levels because during the second quarter of 2017, we recorded a TWD 4.2 billion real estate disposal gain in nonoperating income.

With such disposal gain excluded, year-over-year net income was roughly flat.

We believe that the recording of PPA may give a fair assessment for the revaluation of assets post acquisition from a balance sheet perspective.

However, we believe that PPA-generated depreciation and amortization expenses create long-lived noncash distortions to income -- to the income statement on an ongoing basis.

We believe that our results should also be evaluated without such noncash PPA expenses.

As such, we have provided here key P&L line items without the inclusion of PPA.

Without PPA expenses, consolidated profit was TWD 15.9 billion with 17.4% gross margin.

Our operating profit was TWD 7.2 billion with an operating margin of 7.8%.

Net profit was TWD 5 billion with net margin of 5.4%.

EPS, excluding PPA expenses, was TWD 1.17.

Page 9. Here is our ATM pro forma income statement.

You can see here that, as the second quarter ramped up, we started picking up operating leverage.

It's exciting to see the scale of the combined entity and the potential drop-through it can generate.

For the second quarter, revenues for our ATM business were TWD 61.8 billion, up TWD 5.8 billion from the previous quarter and TWD 2.3 billion from the same period last year.

This represents a 10% increase from a quarter-over-quarter perspective and a 4% increase from a year-over-year perspective.

Gross profits within ATM were up 25% or TWD 2.4 billion quarter-over-quarter and 3% or TWD 0.3 billion year-over-year to TWD 11.9 billion.

Gross margin for ATM was up 2.2 percentage points quarterly due to higher seasonal loading.

Gross margin for ATM was down 0.3 percentage points on a year-over-year perspective, was primarily due to currency impact.

We believe currency impact had greater than a 0.8 percentage point impact annually.

Operating income improved 55% or TWD 1.8 billion from the prior quarter and 7% or TWD 0.3 billion from the year-ago quarter.

Operating margin improved 2.3 percentage points quarterly and 0.2 percentage points year-over-year.

Net income improved 374% or 0 point -- or TWD 2.8 billion to TWD 3.5 billion on a quarter-over-quarter basis.

On a year-over-year basis, net income declined TWD 4.3 billion due to the investment disposal gain related to real estate recorded last year.

Year-over-year, net income remained flat at TWD 3.5 billion without consideration of the real estate gain.

Net margin finished the quarter at 5.7%, that is up 4.4 percentage points quarter-over-quarter and down 0.2 percentage points without the consideration of real estate gain last year.

Without currency impact, we would also have posted a significant improvement on net margin.

Without the P&L impact of PPA depreciation and amortization, ATM gross profit would have been 21.2% and operating profit would have hit 10.3%.

Page 10, ATM operations.

Here is a quarterly graphical presentation of our ATM pro forma P&L.

During the second quarter, our ATM revenue improved 10% sequentially and was up 4% year-over-year to TWD 61.8 billion.

On a U.S. dollar basis, ATM revenue was up 6% year-over-year.

Given that this is the first time we've presented this pro forma chart, it is worth noting that Q1 2018 year-over-year decline was principally caused by currency.

On a U.S. dollar basis, Q1 2018 year-over-year revenue grew by 3%.

From a capacity perspective, for the quarter, we had a net addition of 470 wirebonders for a total of 25,216 wirebonders and a net addition of 336 testers for a total of 4,726 testers.

8-inch wafer processing capacity stayed flat at 240,000 wafers per month.

12-inch wafer processing capacity, including bump, fan-out and copper pillar remained at 320,000 wafers per month.

Page 11.

Sequentially for the second quarter, our communications segment stayed at 53%.

Our computing segment grew 2 percentage points, driven by processing and memory devices.

Our automotive, consumer and other segment decreased 2 percentage points to 31%.

This segment remains healthy, though.

We believe the decline is cyclical in nature.

Page 12, ATM revenue by revenue type.

You can look at this.

I think we've -- we're now including tests and materials into this allocation as it's a part of ATM.

All right.

Page 13.

Here, you can see our EMS graphical presentation.

Since we went over the P&L earlier.

We'll skip on.

Page 14.

Here, you will note our product segment within our EMS business.

Our computing and industrial product segments each rose 2 percentage points.

Computing was driven by server-related products, and industrial products were driven by smart handheld products.

Our communications segment declined 4 percentage points.

We expect the decline to be seasonal in nature and expect somewhat of a recovery starting next quarter.

Page 15.

Balance sheet.

At the end of the quarter, we had cash and cash equivalents and current financial assets of TWD 85 billion.

Our interest-bearing debt increased from TWD 74.5 billion to TWD 216.6 billion at the end of the quarter.

The increase in debt is related to the completion of our SPIL transaction.

Total unused credit lines amounted to TWD 135.6 billion.

Our EBITDA for the quarter was TWD 24.9 billion.

Page 16.

On a pro forma basis, machinery and equipment capital expenditures for the second quarter totaled USD 356 million, of which $211 million were used in packaging operations, $117 million in testing operations, $19 million in EMS operations and $9 million in interconnect materials operations.

Those are all U.S. dollars.

We do expect our 2018 capital expenditures to be below total holding company level, depreciation and amortization.

Depreciation and amortization excluding incremental D&A added by PPA is currently running slightly below TWD 1.5 billion per year.

In U.S. dollar terms, EBITDA for the quarter was USD 647 million.

With the retirements of our convertible bonds, we initially believed we would be able to start giving our outlook in a more typical transparent manner.

However, our nonoperating revaluation gain booked during the current quarter was above the 10% of pretax income threshold as established by the Taiwan SEC.

During such scenarios, Taiwan-listed companies are not allowed to directly provide guidance for that quarter and the 3 quarters thereafter.

We, unfortunately, have to start out being somewhat indirect in our outlook.

Despite the jockeying between the various sovereign entities, we believe that the consumer will be able to find a product to purchase.

As has always been the case, the consumer has the power to choose.

They are free to choose a different product, a more expensive one from a different vendor, even in a different color.

The underlying demand remains unchanged.

This is what our customers see, and this is why our customer order flow remained strong.

In today's environment, supply chains are incredibly flexible.

Logistical inefficiencies and bottlenecks are resolved mid product cycle.

We are part of that flexibility.

We are part of that scale.

We are part of that capability.

However, within what we do, we are the most flexible, we have the most scale and we are the most capable.

In short, ASE will be part of every solution.

So next slide.

So our outlook for ATM on a pro forma basis in U.S. dollar terms.

ATM third quarter capacity should increase 3% to 4% quarter-over-quarter while our utilization rate should increase 2% to 3% quarter-over-quarter.

On a pro forma basis, ATM third quarter gross margin should approach third quarter 2017 levels.

For EMS, in U.S. dollar terms, EMS third quarter business should be similar to fourth quarter 2017 levels.

EMS third quarter operating margin should approach 7 -- second quarter 2017 levels.

With that, we can move towards the Q&A section.

(Operator Instructions)

Do we have questions from the floor?

Name and company, sir.

Okay.

Randy Abrams, Crédit Suisse.

For the ATM business, if I did the math right, it looks like you're guiding a few point improvement for third quarter gross margin.

Could you talk about the drivers of the improvement in margin and maybe a bigger picture combined company how you expect leverage, like leverage to improve margins from the scale of the business and kind of what you see as like a target model for gross and operating margin for the ATM business?

I think for the overall ATM business in the third quarter, I think we will continue to see some margin improvement, largely because of the higher loading.

In terms of the leverage there or the synergy that's being created after the merger, I think aside from the restrictions from China MOFCOM, also, purposely, we want to -- we don't want to rush into any integration.

I think the first priority is really the -- stabilize the customers' reaction to the merger and also stabilizing a lot of the employees' concern, if there's any.

And we're seeing very good results on that.

So far, we haven't seen any customer move out of orders.

We're seeing the 2 companies already starting some of the initiatives in terms of our R&D efforts, and we're expecting the working relationship continue to be a smooth one.

I think the results of -- particularly in the second quarter, I think is that the merger so far is being -- going very smoothly.

And we're seeing very strong results in second quarter, and we expect this to continue into third quarter as well and remaining of the year.

Okay.

Second question, just on the guidance.

It's 3 parts here.

But the capacity, 3% to 4% growth, where are you adding capacity in the business?

And then for -- if you put the 2 together, capacity and utilization, if you could give a flavor, utilization where you're at from combined company second and third quarter.

And then for that type of growth, what are you seeing in terms of the markets?

You mentioned the tariffs, and it's not really impacting the demand.

But if you could give kind of a flavor how you see demand from the different end markets.

We do not see too much of the impact from the tariff.

Having said that, we actually do not know what will be happening for the next few quarters.

So we just have to play by ear.

So far, all of the customer demand is to come in pretty strong.

Our loading will be a little bit more tighter comparing to Q2.

We do have capacity increase in wirebond, test, bumping and fan-out.

We are going to see some new product ramp starting Q3, and we're pretty excited about that.

In terms of the segments, we're seeing the general strength from all segments.

Okay.

Last question, if I could ask on the SiP business, SiP on the EMS.

If you could give a flavor where that is for contribution from EMS now and outlook from that if you expect from all the movement in products that can stay stable this year and then how the -- kind of an update on how the outlook for that business looks over the next year.

I think in terms of SiP business, I think we're making quite a bit of progress this year in terms of expanding our customer base as well as taking on new projects.

I think, right now, we are seeing -- we're having multiple customers and multiple projects undergoing with the existing projects also in a steady state.

I think in terms of the overall percentage, it will stay relatively flat from last year, similar level of last year, about 15% overall.

Okay.

And I guess just looking out next, it was a growth driver for a while, and then we went through a period of rebalancing.

And so now we're in a period of stability.

I guess your expectation, do you think stability next few years or you're starting to see waves of products that, that can start to be growth driver again?

I think we're seeing both now.

The existing product, we do have stability.

The end market always have uncertainty, but within our visibility, we have quite a stable production as well as the manufacturing.

The -- what we are expecting is, by the end of this year, we would like to report some news.

And I think I made a statement this year, this will be a year where we start ramping up the SiP in terms of the customer portfolio and also the -- some meaningful dollar gain.

It is too early to talk about exactly what are the customer portfolio and also what is the dollar gain.

But hopefully, by the end of this year, we will give you a much more definitive number.

But this year, we will definitely have new customer ramp-up.

This is Rick from Daiwa Securities.

So just some number I want to ask for the modeling purpose.

On the pro forma -- or you're holding level -- your company holding -- your holding company level, what's your -- the total number of wirebonders and tester in the bumping capacity for second quarter on a pro forma level?

At the end of second quarter, we have total wirebonder of 25,216 units.

In terms of tester, we have 4,726.

In terms of bumping capacity for 8-inch, we have close to a little over 240,000 wafers a month.

For 12-inch, about 323,000 wafers a month.

Okay.

So across the board, you're going to increase the capacity by about 2% to 3%, am I correct?

Correct.

Okay.

And then the second question is about your [co-walls] development.

Can you update us what's your progress of your co-walls ? And are you going to see more design wins in the next couple of quarters?

I'm not sure we can make a direct comparison to the [co-walls].

But if you're talking about the high-density fan-out, you're talking about the package configuration similar to info of co-walls , then the answer is yes.

And there has been designing going through it.

We have been shipping production in a steady state, and the -- we are seeing some of the customer adoption rate in those 2 type of configuration without giving this exact 1:1 comparison.

But in the similar configuration, similar format in a similar resolution requirement, yes.

Okay.

Just a little follow-up.

Apart from the graphic application you mentioned in last quarter result, any more new applications in this regard?

The OPTO Photonics.

Roland Shu from Citigroup.

First question is, TSMC has seen better high-end smartphone demand in second half than 3 months ago.

And the UMC, however, is seeing on the high inventory label for the slower smartphone digestion.

So what's your view for the smartphone demand in second half?

We don't really have a visibility of the smartphone demand for second half, nor are we in the position to comment about a smartphone.

So -- but you said for all segment in second half will be in your core in general chain, but that in the -- for which sector, communication, consumer of computer?

Which segment is performing better?

The comment we can make is based on our customer portfolio.

We are seeing steady demand for the second half, and that is across all sector, including the smartphone.

But that's strictly speaking from the ASE customer base.

That's communications sector, how do you compare the customer demand now and compare with 3 months ago?

Is this improving?

Or is this deterioration?

It's improving, but there is a shift.

I think the first half, we have a stronger uptick from one of the camp, and I think there will be a shift in the second half.

Okay.

Yes, I think -- yes, because we always check that in the past 2 years.

So I look at in 2015 to 2017, both ASE and the SPIL's revenue growth were behind peers such as Amkor or the other China maze in a big way.

So now since the SPIL transaction has been done, so what's your plan or strategy to resume the growth after this acquisition?

And what kind of product or technology will drive the growth going forward?

We have to focus.

There needs to be a balance in terms of revenue growth versus margin expansion.

For the margin expansion, you really have to tie that to the technology advancement.

Over the last few years, we have some rebalancing of our product portfolio, mainly from the revenue balance to the margin expansion perspective.

You have also seen the -- a steady improvement as was investment in the more advanced technology, such as 2.5D.

And this year, in the -- we have a very high expectation about our Deca fan-out start ramping up in Q3.

We also have the embedded substrate.

We will see the ramp-up, as was new product portfolio for customer in the SiP arena.

This new customer, new design, new advanced technology will represent a more steady growth as well as margin expansion.

In terms of the legacy product, we will be facing some pressure from our competitor, and we do understand that.

So it's a balance between how do you protect the legacy product of the existing product portfolio or margin as well as accelerating on the advanced technology in a new customer, in a new configuration to expand your margin.

And that will be the focus for the next few years.

So do we have any view for this advanced technology, like 2.5D fan-out, embedded subject, the contribution this year or next year?

We will have a better view by the end of this year.

Okay.

So may we have OpEx, operating expense guidance or outlook for the -- going forward?

How are we going to model this operating expense?

We were expecting -- I think, earlier in the comments, was 9% for the rest of this year, and...

At the holding level.

Yes, at the holding level.

And then next year, let's get a little bit more experience first, right?

The next question comes from the line.

Gokul, are you there?

Name and company, please.

Gokul Hariharan, JPMorgan.

First question on the CapEx side.

Could you talk a little bit about what is your plan from CapEx?

When I look at the first half of the year, it looks like CapEx is running slightly higher than the levels that you ran in first half last year.

Could you give some -- could you give us some idea about what are you thinking about CapEx in second half of the year as well as -- I think before the acquisition, it still did have some low utilization quarters.

So should we think about lower level of CapEx as we go into the next several quarters?

That's my first question.

I think for this year's CapEx, it will be higher than last year's, and -- but it will still be below our depreciation, although the depreciation, I mean excluding PPA, the additional depreciation that we'll put in.

I think from the numbers that we're running for the whole year, I think the depreciation level will be capped at around TWD 45 billion.

So the total CapEx will be below that number.

Also, in terms of the CapEx breakdown, I think roughly 65% or 2/3 of it will be for assembly and 25% plus will be for test.

I think this year, we are putting a bit more investment into our test kit capacity.

For material, it will be around 2% and then the rest for EMS.

I think it's a pretty well-balanced CapEx plan for the year.

Okay.

The other question I had was on your view on EMS profitability as well as maybe some more specifics in terms of how you think about profitability in the SiP projects that you're starting to ramp, some of the newer projects.

I think, in terms of this year, we're going through some, first of all, the rebalancing of our business.

And also in terms of some specific product lines, we're going through some economic business model or profile changes.

Therefore, those have some impact on the margin.

And also, we are ramping up some of our newer facilities.

There will be some initial costs that will be incurred in this year.

So the overall margin, we're going through some pressure this year.

Although the -- I think the overall target is still -- try to move the gross profit margin back to double digit or 10% level and hoping the operating margin can also see some gain back to the last year's level.

Okay.

On just narrowing down on the SiP projects.

I think you guys talked about expansion in the newer customer.

Could you talk a little bit about what level of diversification you can achieve in terms of the lines on when the customer and the SiP projects?

And secondly, when you have these new projects coming in, what does it do to the overall SiP profitability?

Are they more ATM-like projects, which would improve the margins?

Or are they going to be more like EMS?

Like maybe you could give some qualitative color on that.

Right now, majority will be ATM.

So all the new ones that are coming in should be more ATM-related?

That is correct.

Okay, that's great.

And could you also address in terms of the diversification?

When you have multi-use customers, how long should we expect it to take for you to have a bit more balance in terms of the customer mix compared to where we've had over the last couple of years?

We're hoping to report a more definite number by the end of this year because we're in the process of ramping up.

But without giving the too-specific number, I think the dollar amount will be in the 100 million range.

I'm Eric Chao from Morgan Stanley.

So first of all, I would like to ask a question on cryptocurrency.

So I think from last earnings call, you guys mentioned that, for ASE itself, the cryptocurrency exposure is going to be at mid-single digit.

Is there a change to that like after bitcoin price dropping to $5,000 at one point?

I mean, do you have a revision on that guidance?

The percentage has not changed, still mid-single digit.

Still mid-single digit, all right.

So my second part of the question is that, on the U.S. iPhone, with the new management on board, do you see any changing overall ASE industrial strategy and overall direction on the EMS and SiP business?

No.

No, okay.

Last question, real quick.

So like the Chinese player has been really aggressive on expanding all their capacities.

Can you like help me understand, where do you pinpoint your biggest entry barrier in terms of like more advanced packaging technologies, such as like bumping, flip-chip?

Like what kind of entry barrier do you have comparing to the Chinese players?

And how many years do you think they are behind ASE?

It's a very subjective statement in terms of how many years are we -- or leading our competitor even though we believe that we are leading.

The bumping, there are different kind of bumping.

Fan-out, there is different kind of fan-outs.

In terms of the technical depths, design capability.

And also, you are supporting partners as well as your -- the ecosystem customers.

If you really want to evaluate the overall leadership, you have to look at the overall picture in totality.

There is a revenue leadership.

There's a margin leadership.

But more importantly, how much money does one company invest in the R&D and advanced technology on a consistent basis going forward 20 years?

On top of that, how many sophisticated experienced engineers do you have that understand how to co-design, co-work, co-calibrate with your global supply chain as well as your partner?

I think if you really want to make a statement about leadership, you should never look at capacity as well as funding per se because for semiconductor, the capacity and funding are a very, very small part of the competitiveness.

Hold on, we have an operator.

Operator has a message.

(Operator Instructions)

We have a question on the floor.

Sebastian from CLSA Securities.

I have quite some questions.

So the first one is on the PPA, we have -- just to clarify, so the OpEx portion will be like TWD 253 million, so which means the -- on the COGS side, you'll be the -- use the TWD 1.4 billion to minus that, then we get the COGS numbers, right?

Yes.

Okay.

Second question is, if I look at your second quarter margin, it seems like SPIL, the portion of the gross margin, probably improved a little bit compared to their first quarter at 16% level along with yours, like 2 percentage point improvement.

But if you look at SPIL itself, the past 3 years in margin almost declined by close to 10 percentage point.

So I wondered how -- what's your plan on fixing that?

Because originally, SPIL probably have a higher margin than your IC ATM, but now they have below your -- they are below your margins.

And how do you see -- can fix that problem?

Or it is not a problem that can be fixed?

Well, first of all, we're supposed to run independently.

So I'm sure the SPIL management is very, very conscious about the margin situation.

And actually, I'm pretty happy to report that, in second quarter, I think they pretty much hit the -- close to the lowest point in terms of gross margin in first quarter this year because of the -- some of the other shifts that they experienced in first quarter.

I think the second quarter for SPIL is -- I think they rebounded pretty nicely, and I think the -- margin wise, it's very, very close to the corporate average in terms of ATM.

Right.

But I remember that the guidance you have for the second quarter, I mean in last conference, and also compared the pro forma gross margin would increase margin about 2 percentage point last quarter.

It seems like SPIL's portion also improved by about 2 percentage point, so which means that it probably improved from 16 to 18, but I think that's still quite low.

Isn't it?

No, it's not.

Like I said, it's very, very close to the overall average margin for the ATM.

Okay.

Third question is regarding your third quarter margin guidance for ATM.

So the guidance almost implied close to 3 percentage point improvement while revenue is about a 5% to 7% increase.

And in the past, when your revenue grew about the similar range, like mid to high single digit, your gross margin probably improved about 2 percentage point.

So it seems like this time around the margin improvement is larger than before.

Can you explain why?

It's in the drop-through, faster.

It's in the drop-through, drop-through.

I think as we continue to increase the loading, I think the drop-through gets better.

And therefore, the margin improvement is bigger.

Also, the product mix, specifically testing.

Okay.

So if you look at the third quarter, in general, like apart from the product mix and the loading, and you and SPIL, I mean, which one of -- like as a stand-alone SPIL have a higher margin upside improvement in 3Q?

I don't think we're going to comment on that.

Okay, got it.

And fourth question is on the EMS.

So is -- will be 3Q this year will be similar to 4Q last year, but usually, your 4Q was your -- has been the peak of the internal seasonality.

So it seems like the third quarter will have a very strong EMS quarter.

So the question is that, it seems like this, the increase matter to you, is more significant than typical seasonality.

And can you specify the reason?

And also after this very high level in 3Q, do you further expect the typical seasonality to occur in 4Q?

I think the overall third quarter changes, like I mentioned earlier on, one of the factors that affects the post of margin as well as the top line is really the changing -- or the changed product economic or business profile.

What I mean by that is for such products, the overall pricing cost, volume, unit volumes, and business terms, are all going through some changes, and therefore, have some impact on the margin itself and those are on the top line.

And I think the -- in terms of the overall seasonality, I think we have mentioned we'll continue to see quarter-to-quarter growth even into fourth quarter.

Okay.

And Joseph, you earlier mentioned about the operating margin pro forma base.

Operating margin target is 10%.

Is that excluding PPA?

We didn't have an operating margin.

Yes, we didn't have an operating margin.

No?

I thought like just -- Joseph just mentioned like OPM target to go back to double digit.

No, no, I was referring to EMS.

In the gross profit margin, our target or the -- we're going to strive very, very hard to bring the gross profit margin back to double digit or 10%.

Okay.

But -- okay, got it.

But based on your third quarter guidance for the EMS OPM, it seems that you should be able to already above 10% gross margin, right?

Not quite.

I think the reason why we actually changed the -- I shouldn't say guidance, I think the outlook that we've given to the audience is because, in the EMS business, I think the operating leverage is less.

So the -- their movement in the gross profit margin translates less impact on the operating margin.

So the operating margin is kind of a more stable measurement to look at the EMS business.

Okay.

And to follow on the crypto part.

I remember last conference, you mentioned the first half will be like low single digit, second half will be high single digit and full year about mid-single digit.

You already mentioned about -- still maintain the guide with mid-single digit.

So is this for full year or second half?

A bit of a correction.

I think the cryptocurrency, we were expecting the business to grow from low single to mid-single.

I think with the current development in the cryptocurrency, I think the -- that ratio actually should remain to be at the low-single level for this year.

Okay.

So which means that the first half and second half will be similar in terms of revenue contribution.

Then because your second half, you have a larger base -- revenue base, which mean the absolute revenue is higher.

That's quite -- actually, quite different from what Tien has said.

So is it different customers?

Or...

I think we're coming from a lower base of the -- I think the first and the second half will be similar levels of business.

It's actually very difficult to tell to, right?

The cryptocurrency market is extremely volatile, so I don't think we want to place any bets.

All right.

The last question from me is that the earlier SPIL's Suzhou plant have a release of 30% stake to Tsinghua Unigroup.

Can you tell us that now is -- might belong to ASE holding company?

So can you elaborate more like any specific plan, what type of the collaboration, what's the roadmap there?

I think the overall -- our China operation, some degree of localization is good for the overall business.

And to achieve that, some engagement or some local partners involvement to help improve the overall working situation in China market could be beneficial to us.

And so far, I think the invested -- the investment has been working very well.

I think the partnership is a smooth one.

Hopefully, well, longer term, that overall relationship can bring us new -- some more new business opportunity in China market.

And also for Tsinghua Unigroup, when their product actually comes out, there will be some back-end demand that we will be looking forward to support.

So that mean the -- I think they already have some product on the larger side.

So what -- you just mentioned a new product coming out, you mean memory?

Right.

A few follow-up questions.

The first one, how are you treating FX in the outlook for third quarter?

Is that factored in?

Are you factoring flat constant currency for both revenue and margin?

Very, very minimum FX impact on the margin for next quarter.

Almost none.

Okay.

To clarify, minimum impact.

Are you factoring, I guess -- because exchange rate, as much as it's moved, should have some impact.

So are you assuming flat currency?

So if exchange rate holds, you might get some benefit?

We're expecting 3% depreciation.

3%, okay.

Second question on automation.

There's been some press stories about ASE taking some pretty big activity in automation.

If you could give a flavor just how much automation, like where you are in that project and the type of benefits you could see from the automation initiative.

We have 3 factories right now, fully automated.

In the next 3 years, we'll add another 11.

We do see benefit in a fully automated factory not in terms of labor, savings, but in terms of the higher reliability and also the higher product mix for specific products such as the automotive.

When the quality requirement and the traceability requirement goes way beyond the consumer products, the fully automated line becomes extremely useful, especially for the customer doing the audit.

They understand where the data comes from, traceability and everything they ask for.

We intend to continue this trend on the high reliability requirement for the difficult-to-manage product, and we will try to incorporate with the fully automated line, yes.

Okay, great.

And a follow-up on the fan-out.

You mentioned Deca is doing pretty well.

If you could give a flavor, just the contribution from fan-out and maybe the latest view, I guess when the volume would get big if mobile application processor or maybe just some of the big end markets where adopt fan-out.

What you see in terms of timing?

Or...

We're very excited about Deca mainly for the adaptability.

In other words, the Deca process has a very well built-in software capability to make sure we maintain a high yield for different type of product mix, different kind of resolution requirement.

It is too early to give you the revenue impact.

What we can say is we are ramping up the Deca, starting in July.

We're in the process of ramping up.

We would like to have the next 6 months to ramp up various product to a various volume that when we look at cost model, the performance, the margin model, are we fully in line with all of the simulation that we have.

So by the end of this year, we have a much more clear picture about the future plan.

Our plan is to continue to expand the Deca line and cover more products and also continue to expand the capacity.

Right now, the Deca line is focused on the 300 millimeter.

So even though the configuration of software is different, but it is still around.

And then the -- after we successfully ramp up the 300 millimeter, we'll start ramping up the 600-millimeter panel.

And that would be the project for 2019, scheduled to ramp by 2020.

And can those panels be used for application processor or some of the high end?

Or is it more for kind of low end?

We would hope so.

Okay.

The last question, it's a follow-up to Sebastian on the -- it's kind of both, the memory and the China.

If you could give a flavor just contribution combined company from your China operations now.

And memory, you had a few years ago with the power ASE, but just how you see about how aggressive you want to be getting back into the memory market now.

The overall China operation in terms of revenue contribution to the group is 15% ATM.

I believe the China memory contribution in that regard is at a minimal level right now.

In terms of the overall group, we are increasing the memory exposure mainly in specialized memory product where ASE's assembly process, material process, test process has special contribution.

But as of today, the memory remains to be a small percentage of our overall product portfolio.

Just one follow-up question.

So last quarter, you said that for 1% of appreciation, while negative impact at your IC ATM, gross margin by 40 bps and group gross margin by 30 bps.

So in 3Q, now you expect 3% of the NT dollar depreciation, but you said probably very limited impact to the gross margin.

So it means that you are sensitively -- is no longer hold?

Or...

I think we're recalibrating it because of the inclusion of large-sized operation to us.

But right now, we are budgeting minimum impact from FX on our margin.

So I think the linkage there is kind of not linked very closely now.

So although we are expecting the currency to depreciate, but maybe 3%, but then we're not actually budgeting any impact on the margin.

One more from the floor.

It's me again.

Just -- because this is the first time that you host a conference after the merger, so previously you couldn't comment too much about your synergy.

So I wondered if you can give us some color about the synergy.

I mean, we understand that, before November 2019, you need to run independently.

But I remember in the MOFCOM document, you also said that there are some flexibility on the capacity trends for allocation and also the R&D discussion at the board level.

So for these 2 -- from these 2 perspectives, can you give us some color about how you could -- is there -- how much room you can -- flexibility, you can do that in the next 16, 17 months?

We cannot comment on the synergy.

Still cannot?

We cannot.

We will comment on synergy after November 24, midnight, 2019.

In terms of the R&D collaboration, there has been ongoing discussion, and the relationship is very, very cordial.

We have warm -- the reception on both end in terms of the R&D collaboration about long term, and that is ongoing right now.

Okay.

Let me put another way.

Is that -- when you -- so earlier, thanks to Rick asked the question about your capacity.

And I believe that you have already seen what's the capacity, what kind of equipment, (inaudible) level that SPIL has.

And along with yours, how much room do you see that?

Like -- or -- do you -- how much room do you -- of flexibility do you have like to transfer or allocate capacity?

At this point in time, very minimum.

We understand where the equipments are, but there is no transfer of business and there is no transfer of equipment because both companies are running independent.

Okay.

Or is it because both of you are running that high UTR in the second half, so there's no need to transfer this to there?

No, even if we're running in low utilization, we're not transferring equipment.

Both company are running independent.

Any more questions?

No?

Well, thank you for attending the second quarter earnings release.

See you next time.