日月光投資控股 (ASX) 2018 Q3 法說會逐字稿

Hello.

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

Welcome to our third quarter 2018 earnings release.

All participants consent to having their voices 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 -- I'll be going over the financial results.

Afterwards we will have a Q&A session with Joseph Tung, our CFO.

Following the event, our VP in Charge of Public Relations, Eddie Chang, will be available to address the media in Mandarin Chinese.

I would like to remind everyone that because ASE Holding was jointly formed on April 30th, during the middle of our second quarter, ASE Holding's second quarter results include a full quarter of ASE and a stub 2-month period of SPIL.

Specifically for April, only 33% of SPIL's net income was recorded in nonoperating investment income.

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

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

During the third quarter, even though we came in slightly under where we thought we would be, we still saw record revenues in our ATM business.

The business picked up in its usual pattern with the July and August revenue run rate stepping up from our second quarter pace.

However, our September revenue run rate seems to have leveled off before when we would have expected.

When looking at our results from a total quarter perspective, some of our business related to cryptocurrency and communications didn't achieve our forecasted targets.

This order timing discrepancy resulted in a small misalignment of product and the capacity and labor put into place to support such product.

As such, our planned optimization of certain lines were not fully achieved during the third quarter.

However, most high-end packaging and testing capacities were still running relatively tight.

So it may be contrary to some sentiments at this time but things are not doom and gloom just as they are not packed to the brim.

Our EMS business performed generally as we expected.

We went through our typical third quarter ramp up.

During the quarter we did note that things appeared a bit more dynamic than normal with some product timing changes from multiple customers.

However with the way of modern supply chains these dynamics are part of the norm.

With that let's start the financial overview.

On Page 3 and 4 you'll find our legal entity results for the holding company in our ATM business unit.

For the legal entity, we recorded fully diluted EPS of $1.43 and basic EPS of $1.47.

Sales were $107.6 billion with a gross profit of $18.4 billion and gross margin of 17.1%.

Operating profit was $8.4 billion.

It is worth repeating here that ASE Holding's second quarter results only include 33% of SPIL's results recorded in nonoperating investment income during the month of April and the consolidated results of SPIL during May and June.

For the lack of comparability reasons, there is not a lot of productive commentary we can add to the legal entity basis results on these 2 pages.

Please feel free to go through them at your leisure.

And if you should have any specific questions regarding the legal entity results, you may address them to us during the Q&A or contact us thereafter.

Let's move to Page 5. Here we have our pro forma P&L for the consolidated holding company.

To generate the historical pro forma periods we 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.

Further, for the second quarter specifically, we excluded the revaluation gain of $7.6 billion.

Aside from these, we made no further adjustments to the pro forma.

On a small housekeeping point, we have included in our presentation an appendix at the end of your packet which contains our quarterly pro forma P&L results from 2017 to present.

These results are presented without commentary.

On a pro forma basis for the third quarter, we had net revenues of $107.6 billion.

This represents a 17% increase quarter-over-quarter and a 12% increase year-over-year.

Gross profits were up 25% quarter-over-quarter.

The sequential gross profit improvement was driven by stronger business in both our ATM and EMS business units.

While on a year-over-year basis consolidated gross profit increased 5% driven primarily by our EMS business.

Gross profit margin improved 1 percentage point on a quarter-over-quarter basis to 17.1%.

This was driven by improved seasonal loading from our ATM and EMS businesses and the corresponding economies of scale boosts.

Gross profit margin declined year-over-year by 1.1 percentage point principally as a result of higher EMS revenue mix and slightly lower ATM gross profit margins.

Operating expenses were $10 billion, up $1 billion sequentially.

However, because of the higher revenue base, our operating expense percentage was down 0.6 percentage points to 9.3% from 9.9%.

Operating profit improved $2.7 billion quarter-over-quarter and was flat year-over-year at $8.4 billion.

Sequentially, operating margin improved 1.6 percentage points driven by increased loading in the seasonally up period offset by higher operating expenses.

On a year-over-year basis operating margin declined 0.9 percentage points, driven by increased operating expenses in our ATM and EMS businesses and higher EMS product mix.

Nonoperating expenses were $0.3 billion which includes net interest expense of $1 billion.

Tax expense for the quarter was $1.6 billion on a pro forma basis.

Net income for the third quarter was $6.3 billion, representing an improvement of $2.8 billion from the previous quarter.

On a year-over-year basis, net income was roughly flat with 2017 levels at $6.3 billion.

On the bottom of the page, we have again provided here key P&L items without the inclusion of PPA.

Consolidated gross profit excluding PPA expenses would be $19.6 billion with a 18.2% gross margin.

Operating profit would be $9.8 billion with an operating margin of 9.1%.

Net profit would be $7.7 billion with net margin of 7.2%.

EPS excluding PPA expenses would be $1.81.

On Page 6 is our ATM pro forma P&L.

For the third quarter, revenues for ATM business were $66.3 billion, up $4.8 billion from the previous quarter, and up $2.5 billion from the same period last year.

This represents a 7% increase from a quarter-over-quarter perspective and a 4% increase from a year-over-year perspective.

Gross profits within ATM were up 20% or $2.4 billion quarter-over-quarter and 1% or $0.3 billion year-over-year to $14.3 billion.

Gross margin for ATM was up 2.3 percentage points quarterly due to weaker NT$ currency and scale efficiencies achieved.

Gross margin for ATM was down 0.6 percentage points on a year-over-year perspective, primarily from product mix and softer-than-expected order flow and the resulting capacity misalignment.

During the quarter, operating expenses were $7.6 billion, up $0.6 billion from the second quarter primarily due to higher expenses related to organizational changes.

As we complete these changes, we expect for these expenses to persist into but end in the fourth quarter.

During the third quarter, operating income was $6.7 billion, representing an increase of $1.8 billion or 37% quarter-over-quarter.

While operating income declined $0.3 billion or 4% year-over-year

On a quarter-over-quarter basis, operating margin improved 2.2 percentage points from 7.9% to 10.1%, driven entirely by gross margin drop through offset by higher OpEx percentage.

While on a year-over-year basis, operating margin declined 0.9 percentage points year-over-year principally as a result of lower gross margin and the organizational restructuring cost recorded in operating expenses.

Net income improved 79% or $2.5 billion to $6.3 billion on a quarter-over-quarter basis.

While on a year-over-year basis net income was flattish at $6.3 billion.

Net margin finished the quarter at 9.4%, which is up 4.4 percentage points quarter-over-quarter and down 0.5 percentage points versus last year.

Without the P&L impact of PPA depreciation, amortization, ATM gross profit would have been 23.3% an operating profit would have hit 12.3%.

On Page 7 you'll find graphical presentation of our pro forma ATM P&L.

And again, during the third quarter, our ATM revenue improved 7% sequentially and was down 4% year-over-year to $66.3 billion.

On Page 8 is our ATM revenue by market segment.

Sequentially for the third quarter, our market segment movements had our communications segment climbing 1 percentage point to 54%.

Our computing segment declined 2 percentage points, led by cryptocurrency applications.

Our automotive, consumer and other segments increased 1 percentage point to 32%.

These are not particularly significant changes.

On Page 9. You can see here that during the third quarter wire bonding revenue declined 2 percentage points, offset by increases in testing, discrete and others.

We do expect our test business to continue to perform well as we make attempts to gain more market share.

On Page 10.

You can see the results for EMS business.

During the third quarter we had revenues of $42 billion, representing an increase of $11.5 billion or 38% sequentially, and an increase of $8.9 billion or 27% year-over-year.

This was driven by our consumer product segment which usually initiates its seasonal ramp at this time.

Sequentially, our gross profit improved 45% or $1.3 billion to $4.2 billion.

Compared with the previous year, gross profit improved 22%.

Gross profit margin for the EMS business unit came in at 9.9%, representing a 0.5 percentage point increase from the previous quarter.

Compared with the previous year, EMS gross margin declined 0.4 percentage points as a result of generational device differences.

Operating profit for the quarter was $1.7 billion, which is a $0.9 billion improvement sequentially, and $0.3 billion improvement year-over-year.

Our operating margin came in at 4.1%, which is a 1.4 percentage point improvement sequentially and a 0.2 percentage point decline year-over-year.

The operating margin decline is principally the result of higher operating expenses.

On Page 11 you will note that our product segment mix within our EMS business.

Our consumer product segment grouping is performing well and has grown 11 percentage points sequentially.

Outside of our automotive segment, all products segments grew in absolute dollar terms.

The percentage share declines in this quarter are more representative of the strength of our consumer product segment.

We currently expect continued consumer strength into the fourth quarter.

On Page 12, you will find key line items from our balance sheet.

At the end of the quarter, we had cash and cash equivalents and current financial assets of $63.1 billion.

Our interest-bearing debt decreased from $216.6 billion to $208.2 billion.

Total unused credit lines amounted to $165.1 billion.

Our EBITDA for the quarter was $21.6 billion.

I believe many people did not notice that in our press release starting in the previous quarter we have started to include our summary statement of cash flows.

I believe the information will become increasingly important given the impact of PPA on our results.

On Page 13 you will find our pro forma capital expenditures.

On a pro forma basis, machinery and equipment capital expenditures for the third quarter totaled USD 290 million of which USD 128 million were used in packaging, $130 million in testing, $21 million in EMS operations and $2 million in interconnect material operations and others.

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

We believe that capital expenditures held below depreciation and amortization to be representative of an overall modest growth environment.

Depreciation and amortization, including PPA adjustments, is currently running slightly below USD 1.5 billion per year.

The overall geopolitical climate is on everyone's mind.

But ironically the more we think about it as a general impact the less clear it becomes.

Even though there is plenty of chatter about changes in order timing, whether it's pull-in or push-out, what's fundamentally clear is that there is a lack of consistent supply chain strategy across all of our customers.

And if you think about it, this makes sense.

In the face of potential tariffs do electronic producers move their supply chains quickly, slowly or not at all?

While the answer is that it depends upon each of our customers' products competitive position, the end market it competes in and the sensitivity to cost shifts.

What that means is that every customer there can exist seemingly contradictory strategies for any one product.

That effectively means there will be no consistency in response.

In the U.S.-China trade war, our ATM business with the majority of our factory output in Taiwan looks to be only minorly impacted.

We could potentially see benefit from potential outflows of product dropping out from China production.

From a short-term perspective, our EMS business looks to see limited impact given that we offer multiple production line options, either in or out of China.

The equity market's recent unease has pointed to an overall sign of uncertainty.

Whether it's interest rates or trade imbalances, uncertainties throw financial markets into a tailspin.

I point this out not to make light of the jitteriness of the markets.

I point this out to say that in what we do change and unknowing is the norm.

Adaptability gets you in the game and certainty is actually an illusion.

And even though our engineering and advanced processes provide the most consistent output and yield, why our customers come to us is that we are able to provide unprecedented flexibility in the face of momentous uncertainty.

With our outlook, we don't see anything that would indicate the environment is dramatically out of the norm.

Things, of course, are dynamic.

Given that, we will continue to monitor the overall situation and stay vigilant.

On a pro forma basis and in U.S. dollar terms, ATM fourth quarter 2018 business should be similar with second quarter 2018.

ATM fourth quarter 2018 gross margin should be similar to third quarter 2018 levels.

For EMS in U.S. dollar terms, the EMS fourth quarter 2018 business sequential growth rate should be slightly lower than the third quarter's sequential growth rate.

EMS fourth quarter 2018 operating margin should be similar to the third quarter 2018 levels.

Q&A?

(Operator Instructions)

Name and company, sir?

It's Randy Abrams, Credit Suisse.

Yes, the first question, just wanted to go through the guidance you just gave.

The -- it looks like the gross margin is seeing better leverage, holding up relative to third quarter.

But your sales, you're guiding back down to second quarter levels, which I think implies like mid to high single-digit decline.

Could you talk about on sales what's driving that, what you are seeing across end markets for that sales decline?

And then, what's holding up the gross margin at third quarter levels?

So just a quick correction on that.

The revenue at the time referred is down roughly 4%, right.

And there should be some deviation.

I think the margin will pretty much be very similar to what has achieved in the third quarter.

I think the, basically, the result came from a little bit more favorable change of product mix going into the quarter.

As for the EMS business I think the margin -- because of the -- also because of the product mix change at the gross level it will come down a bit but at the operating level, it should stay the same.

For the EMS business, could you talk about or give an update on the SiP business both from a contribution?

And if you could talk about diversification, how much is kind of coming from your core projects and maybe the pipeline or diversifying outside your traditional customer.

Well, as we mentioned last time, we are currently engaged in multiple projects in SiP.

But most of the additional projects are in at the ATM level where EMS has remained to have the core projects [similarly now].

But the momentum of such project is actually getting stronger than previously.

And how is the momentum outside your traditional customer?

Well, it's basically on track.

And as Tim mentioned last time, I think the -- we are looking at over $100 million range.

You talk also about the memory strategy.

I think for SPIL maybe confirm what's in the press.

Looks like there's originally a stake with SPIL in Tsinghua and now with Jinhua, so you could talk about those equity stakes, just the strategy behind that, and what the opportunity is.

And I guess now with the risk with the U.S. looks like they're putting export ban on Jinhua?

I think that Jinhua investment in our facilities in China is mostly as a step for us to somewhat localizing our operation in the region.

It's not so much aimed for the memory business that eventually come from Jinhua's factory going forward.

As well as for the [Shaman] project that SPIL is engaged in.

I think I should say from the start it's never intended to be a captive site for the Jinhua.

But it's more to, the setup is more toward the, serving the potential customers in the South China region.

And right now the project is going on the track.

We put in roughly $110 million investment into the project here, but more than 80% of it is really spent for the construction of the factory.

There is very, very minimal amount invested into machinery at this point.

So I think it has sufficient flexibility to move to something else if the progress in Tsinghua factory is not as planned.

So that the impact on that equity very, very minimum.

Okay.

Just one final question, 2 part on the CapEx with a bit of slowdown in uncertain environment.

If you could give an initial view, how you are looking at CapEx going into 2019.

And in the EMS business where you have more of the capacity in China, how flexible are your tools?

Would you need to spend to avoid tariffs if you had to shift production out of China for EMS?

Do you have flexibility in terms of that capacity in that tools or would you need to spend additional CapEx to ramp at another facility?

Well, we certainly have the sufficient flexibility in looking at the overall situation.

In terms of EMS, we do have multiple sites in different locations.

And we are actually seeing maybe a handful of customers looking to move some of the orders outside of China.

And we certainly have the capacity or the flexibility to entertain that.

In fact we are beefing up the Mexican -- Mexico factory, also our Taiwan factory.

We are setting up a new footprint in Eastern Europe.

We are still on track with the JV that we were supposed to setup with -- in Brazil.

So things moving along.

And we are capable of entertaining whatever requests from our customers in the scenario.

Do you have initial view on CapEx for 2019?

Well, I think it's a little bit early to give you a number.

I think the overall probably remains to be below the D&A that where you have over the years.

Yes, this is Rick Hsu from Daiwa Securities.

Could you -- Ken, could you put back your Q4 guidance because I -- my handwriting is slow, I missed the last part, EMS part.

Can I move it again?

Thank you so much.

Right.

So my first question, again this is a housekeeping question, could you run through your Q3 utilization rates among your 3 magic category wirebond, testing and bumping?

And what's your outlook for Q4 in terms of capacity add and utilizing rates?

Utilization rate?

I think Q3 for both wirebonding as well as advanced packaging, the utilization is overall over 80%.

And in fourth quarter we are expecting around 80% level.

In terms of testing, it's above 70%.

And that should remain relatively the same in fourth quarter for test.

All right.

Yes, thank you so much.

That's clear.

The next question is, could you talk about your customers' inventory level?

I think the, we are not in real good position to comment on inventory level, but from our forecast, I think the -- the [debt] situation, I think the operation is still going at a normal level.

Okay.

One last question.

This value EMS business in Q4, it looks the scale -- the scale keeps expanding, but your OPM is going to be similar to Q3 and your GM will be, I don't know, maybe a little bit lower.

Could you elaborate a little bit more about this remix?

Well, as I said, the -- there is going to similar, well little bit lower growth rate in the fourth quarter in terms of our EMS business.

But the bulk of the increase is really coming from higher material content business.

And therefore, the overall gross profit margin could come down a bit.

But with the operating leverage that we have I think at the operating level the margin, we can keep the same margin at the operating level.

Bill Lu from UBS.

Follow up to Rick's question.

If you look at EMS and specifically the consumer products, ramping up quite nicely, I think in the past, you've talked about margin getting better with volume.

Can you give us an update on that?

Can you -- can you repeat the question again?

Yes, the consumer EMS product that is ramping up quite nicely, in the past you've talked about margin getting better when volume starts ramping.

Can you give us an update on where margin is right now in the outlook?

You mean the product's specific margin or the overall EMS margin?

I guess I'm talking about -- you've got this one big product in SiP, right, consumer-related SiP.

Yes, I think the SiP itself, the margin is improving.

But as a whole, if you count the EMS business as a whole, the SiP business does represent a higher material content business.

Therefore the more we do in SiP, the -- that will put some pressure on the overall EMS margin.

But like I said, the operating leverage itself will carry the operating margin at the same level.

Yes, thank you.

I guess I'm wondering, if volume continues to ramp where does margin go for that SiP business?

Sorry?

If volume still ramps even higher in the next couple of quarters, where should we expect margin to go?

For EMS?

Just for the particular product.

Well, we are not disclosing particular product margin.

Secondly, Ken talked about test gaining market share.

Can you talk a little bit more about where those shares will be coming from in terms of applications?

It's part of the strategic initiatives that we're focused on.

We believe that there is lower hanging fruit out at test for us to go and get.

Lastly on the trade tension, the geopolitical risk that Ken talked about.

It seems like lot of the concerns are on auto and industrials.

Can you talk about what you're seeing from those customers?

I always think that we are not serving any particular customer in any particular region.

I think we're serving the whole world.

I think the -- it doesn't really matter where the product is being built, we still provide the same services.

I mean short-term wise, there is going to be fluctuations because of the lot of the uncertainties ahead of the industry itself.

But I think in the longer term, I think the business remains intact.

So we just have to make the necessary adjustments along the way when the customer request changes.

And we need to -- we certainly have the flexibility to respond to that.

And our exposure to automotive tends to be fairly small.

So I don't want to project that on to the rest of the industry.

This is Roland Shu from Citigroup.

So Joseph, what you just mentioned for your [CapEx] level in 3Q is still at normal level, did you mention that?

Yes.

So how does that compare with 3 months ago?

I don't think there is any real differences between the -- during these 3 months.

I think the situation looks, the -- pretty much as expected.

And we're not seeing anything abnormal.

Okay.

And for TSMC, I think now is -- just projecting very strong growth for 7 nanometer.

And the market actually had very positive comment on that.

So how many or what percentage of this 7 nanometer wafer packaging using conventional flip-chip BGA or flip-chip CSP packaging.

And how many products are you engaging with the [Tepa]?

Well, I wouldn't know.

I think the -- whatever the customer give us -- asks us to do, we'll do it.

So for this 7 nanometer revenue contribution, will it be a significant part for your revenue growth next year?

So Roland, we don't necessarily focus on the nanometers in terms of what comes to us.

Our focus is more upon the dye pads and the dye pad density on the dye, right, and it could be 14, it could be 28.

There are associations and trends in terms of what you use or what type of packaging you use, but it doesn't necessarily mean that like a 7 nanometer packet.

I mean you could feasibly do wirebond on a 7 nanometer package -- on a 7 nanometer dye.

Yes, understood.

So this 7 nanometer growth path, I think what's the business you are seeing for next year.

Well, I think TSM looks at things probably slightly differently than how we do it, but I don't believe that we really look at 7 nanometer.

I think 7 nanometers to them is not as unit driven, right, we're much more unit driven.

Okay.

So next year for your IC ATM, well these are -- [these] leading-edge technology to be big growth driver for your IC ATM next year.

I think -- are you referring to fan-out, to be specific?

Fan-out is my next question, yes, but just specifically for this, maybe your 10 nanometer, 7 nanometer or 16 nanometer.

Well, it depends on how you define advanced packaging.

I think for us bumping, flip-chip, wafer level CSP, those are the so-called events.

And that represents about 30 some percent of overall ATM revenue.

In terms of fan-out, I think it's still -- right now it's still small.

I think for the 300 millimeter we are getting into mass production in -- starting from now actually in fourth quarter we are moving into mass production.

But for this year I think the percentage of our total revenue is still going to be quite low in the year.

I think the real meaningful number, should start -- we should start to see some meaningful numbers starting from first half next year.

That will be coming from 300 millimeter?

Fan-out.

Yes, fan-out.

Is 300 millimeter wafer type or panel type?

Wafer.

How about the panel type?

I think that will be late-2019 or even early 2020.

That also start from the 300 millimeter pipe, 300 millimeter, right?

I mean, for your panel.

No, I think the 300 millimeter is really wafers.

Then we go to 600 and above for panel.

Charlie Chan from Morgan Stanley.

So I actually want to follow up on Lu's question on our SiP margin improvement, right.

So first of all, why this time around your margin can improve for that SiP projects?

Is that because of a better yield or higher volume?

So my question essentially, I want to get a sense, whether this part of margin can keep improve in the coming quarters or years.

Well, I think there is -- it's not a linear type of pattern in terms of volume growth with -- margin growth with volume.

It really depends on the characteristics and the structure of the product itself.

I think the margin improvement that we mentioned before is a result of overall streamlining the overall operation.

It's not just because one particular product volume goes up and the margin starts to improve.

It really depends on how you streamline the overall operation, how do you lay the production line, how do you bring the yields up.

And also then how do you structure the deal, the business terms and so on and so forth.

So there is a lot of different factors involved.

And it took us roughly over a year for us to streamline the overall situation.

And then it becomes a profitable business for us.

So I can tell you, when volume goes up and then margin will improve how much, it doesn't really work that way.

Generation to generation, products characteristics may be very, very different.

With the previous generation, the material content could be lower and be higher in next generation.

So it really depends on how you structure the overall business just to decide what kind of margin you will be getting.

And so for the current margin value, are you kind of happy with the current margin?

I mean, how would you describe the entry period of this business?

And is that possible you can potentially hike the price for this project going forward?

Well, how do I characterize the margin, is that?

Yes, so I mean to -- from the surface right, this business is -- looks like a good business to have, right, but you continue to dilute your corporate margin, right.

So from CFO's perspective are you happy with the current margin?

And I think that is associated to your competing edge or the entry bidder of this project.

Do you have any competition for these projects?

From the CFO's perspective, I'm never happy with any margin that we're getting.

But when you look at different businesses you really look at -- need to look at the potential of it and also the return that it's getting.

I think as far as the return, the SiP business is as good as any other business that we're currently doing now.

And also on this EMS business overall, right, so long term, do you expect this business to outgrow the ATM business?

Outgrow by what?

What do you mean by outgrowing?

Let say, ATM to grow 5% to 10% and EMS maybe grow more than 10% CAGR in coming years?

Well, that's certainly possible.

But it really depends on what kind of business you're getting and whether you get the business or not.

Yes, but is there any kind of more, kind of secular transfer of tempo, customers wants more modularization?

Or you think you can really differentiate your EMS business with other industry peers?

Well, I guess what I can say is we are seeing the so-called SiP business regardless it's in the EMS arena or is it more ATM-related kind of SiP projects.

I hope I pronounced that right.

Is SiP project not [SiC] projects.

But I think the overall, we're seeing the movement being built, being built up.

And like I said it's really actually starting from this year that we are starting to have meaningful multiple customers and multiple projects.

And also in multi -- in different applications and functionality, different types of SiP projects that are going through the pipeline.

And we're seeing the momentum continuing into next year.

We are actually engaging in more different projects now.

So you know, we still firmly believe that SiP is really going to be the next big driver, a growth driver for us.

And we believe the trend is moving to our favor.

I think in terms of technology we need to continue to put more investment into our technology offering to grab that business opportunity.

Sebastian from CL Securities.

My first question is to -- asking about the fourth quarter IC ATM guidance, can just clarify that it's going to be down like 4% Q-on-Q, but if we compare to the past few years average seasonality in 4Q, it's about flat, sometimes up a little bit, sometimes down a little bit, but I think that down 4% Q-on-Q has been the very low end of the 4Q seasonality in the past 5 to 8 years.

So look, can you -- I mean illustrate more or elaborate more details about what is driving this weaker-than-seasonal 4Q?

I think this year the peak came a bit earlier than previous years, yes.

I think the -- so the overall fourth quarter forecast looks a little bit soft comparing to previous years.

I think that's the result of the overall uncertainties ahead of us.

I think there's a lot of things going on at the same time in this world economy and therefore I think the overall pace is kind of slowing down a little bit and that's what we're seeing from our customer at this point.

But I don't think this is, this is something that's structurally different.

I think the market or the industry goes up and down with different pace.

So this is something that we are seeing.

And now we're -- we just have to respond to it.

So with the fourth quarter guidance then we roughly get the full year ATM business compared to last year if we use a pro forma basis, probably about up like 1%, 2% year-over-year.

Correct me if I calculated that wrong because of the currency difference.

But I think that certainly is less than 3% year-over-year growth.

But so I think that -- if I remember correctly, that will be another year for ASE, the IC ATM business to grow less than the overall semiconductor industry growth.

So would this be the norm going forward?

Or is there any chance that for ASE the pattern grow, business grow to accelerate, to reach the average semiconductor grow or even above the semiconductor grow again?

I think our overall position is getting stronger, although there is some hurdles for us to cross.

I think the integration that we need to go through, we're still having some hurdles in front of us, including the restrictive conditions put on us by MOFCOM.

But going forward, I think the overall momentum, that will start to come back again.

But having said that, I think one thing I want to point out is it looks, from the surface it looks like if we are growing slower than the overall industry growth then, it may look like we are losing market share which I -- in my mind I don't think that is really the case, is really we have to strike a balance between return and our margin or -- and business growth or revenue growth.

There are businesses that either we can afford to get into or businesses simply that are money-losing proposition, those are the ones that we would not be interested in moving in.

So it's really matter of choice, how do we want to grow our business and where areas -- which areas we want to move in more aggressively or not.

As an example, I think starting from this year we are really putting more investment or efforts into our test business.

We see better potential because of the higher turnkey ratio that are going through our factories.

So it really depends on the how the market moves and how do we respond to it.

But I think the overall situation, we are the leader and we will remain to be the leader in the industry.

And there will be some earlier investment that we need to make to maintain our leadership position that could create some capacity timing differences, gap between revenue and investments.

So there are lot of factors that are in play and we just have to hopefully make the right choices at the right time.

My third question is on the trade war between U.S.-China.

Just wondering, given the tariff, are you seeing like some customers that used to use the Chinese outside players and now or increasingly coming back to you because of the concern about that?

That's really unclear.

I think that could be the theory.

But the customers really don't come and tell us that they are purposely moving their business from one place to another.

There could be some business moving to or the theory says that, that should be the logical move.

But different customers have different agendas or different perspectives in terms of running their own businesses.

And we'll just try to be as flexible as possible.

If there is more business coming in, we will have to be able to take it and with different locations and different sites.

Last question from me is, I'd like to know more about your, ASE Holding's strategy in memory packaging.

I think this is somewhat related to your strategy in China as well because I see your collaboration with [Fujen Tsinghua] and also earlier collaboration with the Tsinghua Unigroup.

So it seems to me that these strategic move -- action is more related to the memory development in China.

So I just wonder, just want to ensure that I get it right that the ASE Holding as a company will become more aggressive in the Chinese memory packaging opportunity, so that means that the discrete memory packaging will become more serious, aggressive in the future, is that right?

As I -- as we mentioned earlier on, I think the -- whatever we do in our different locations or the investment that we make, we're trying to capture whatever business opportunity there are for us to materialize.

No, China memory is certainly one area that worth our attention to it.

But that doesn't mean that we will be just putting all the resources in China solely for memory.

I think whatever investment that we make, we have to make a balanced decision and, you know, looking at different scenarios so that we have the flexibility to make the required changes.

Tsinghua is one example.

I mean, from day 1 the investment was never meant to be a captive operation to serve Tsinghua only.

It's really aimed at serving the customers that are coming out of Southern China.

That could be logic, that could be memory.

It really depends on how the market moves and how the -- what kind of progress China semiconductor industry is moving -- or is making.

So we will maintain -- we will remain to be flexible.

And whatever is required we will have the capability or the capacity to service that -- that demand, that will be the strategy that we have.

So I don't think we -- we're not saying that we had our mind set on China memory.

It's whatever China opportunity provides we will have the -- we will have the suitable capacity of investment made in that region.

Thank you.

Just one follow-up on that.

I think I remember earlier Joseph, you talk about that you still have the flexibility in Fujen Tsinghua, this collaboration that you can also -- you have the flexibility to do something else if Fujen Tsinghua, their progress like is not on schedule or slower than expected.

So in terms of the -- sorry, this is dumb question, but I just wonder, in terms of the equipment and that capacity that you're planning for memory and planning for other [evolution] by analog or logic.

Would it be the same?

Or can you leverage the same equipment for memory and do logic?

Well, the test -- testers will be different.

For packaging more or less they are fungible, but for Tsinghua, in particular, I think we haven't really started the installation of equipment.

And we certainly have the latitude to make any changes that we need to have in terms of both the equipment that we will eventually install or the timing of the installation.

Any more questions?

Any questions online?

We don't have any questions online.

So last call?

All right.

Thank you for attending.

See you next quarter.

Thank you.