艾克爾 (AMKR) 2013 Q3 法說會逐字稿

Hello, and welcome to the Amkor Technology Third Quarter 2013 Earnings Conference Call.

My name is Saul, and I'll be the operator for today's conference call.

At this time, all participants are in a listen-only mode.

Later, we will conduct a question-and-answer session.

This conference is being recorded today, Monday, October 28, and will run for up to one hour.

I will now turn the conference to our host Mr. Greg Johnson, Senior Director of Investor Relations and Corporate Communications.

Mr. Johnson, please go ahead.

Thank you, Saul and good afternoon, everyone.

With me here today are Steve Kelley, our President and Chief Executive Officer, and Joanne Solomon, our Chief Financial Officer.

Before we begin, let me remind everyone that there will be forward-looking statements made during the course of this conference call.

These statements represent the current view of Amkor Management.

The actual results could vary materially from such statements.

Prior to this conference call, Amkor's third quarter 2013 earnings release was filed with the SEC on Form 8-K.

The earnings release together with Amkor's SEC other filings, contain information on risk factors, uncertainties, and exceptions that could cause actual results to differ materially from Amkor's current expectations.

With that, I'll turn the call over to Steve.

Thank you Greg.

Good afternoon everyone.

On today's call, I'll discuss our third quarter performance, the status of several key initiatives, and our fourth quarter expectations.

After Joanne reviews our financial performance in more detail, we will open up the call for questions.

My primary focus is driving revenue growth and improved financial performance for Amkor.

We saw both in the third quarter.

Sales were up 10% year-over-year, and 3% sequentially.

Excluding the power discrete business, which was not included in our guidance, third quarter sales were consistent with the midpoint of our expectations.

Increased test and wirebond revenues more than offset the decrease in advanced packaging revenue caused by the high end smart phones slow down.

We are seeing particular strength in the NAND flash market, where revenues doubled year-over-year, and grew more than 25% sequentially.

Gross margins also improved in the quarter.

We've reduced unit cost in the wire bond business through improved utilization, incremental cost reductions, and lower gold wire cost.

These improvements in the wirebond business drove overall third quarter gross margin to 19.7%, excluding patent litigation charges.

Amkor leads the industry in developing and commercializing cost effective advanced packaging technologies.

These technologies provide increased value to our customers, while typically generating gross margins above the corporate average.

Our advanced packaging technologies include specialized technologies, such as fine pitch copper pillar bump, which is a key enabling technology for [crusilla] [condilla] implementation.

In addition, we perform wafer storage and final test services.

So, the first key to success in the advanced packaging and test area is to generate reasonably quick returns on investments made for driver customers.

The other important factor is to cultivate second wave customers who could utilize assets left behind when driver customers migrate to the next platform.

Many of those second wave customers in the mobile communications space are fabless companies located in Taiwan and China, not a traditional area of focus for Amkor.

But we are working hard to change that situation.

We have a solid foundation for growth, including three factories in Taiwan, and a very large factory in Shanghai.

We lead the industry in wafer level and flip chip technology, and we have installed the most capacity in those areas.

So, we are well-positioned to gain share, as Taiwan and China based fabless established companies move to more advanced package platforms.

To improve our regional focus, we are expanding our Taiwan and China based sales, technical support, and service teams.

We want our customer facing organizations in Taiwan and China to be on par with the organizations we have built in the US, Europe, and Japan.

Our Taiwan and China sales leader, an industry veteran, now reports directly to the Head of Amkor's Worldwide Sales Group.

We are seeing some early signs of success with four recent wins in the smart phone space, which will ramp to volume in the first half of 2014.

We recently hired [Endrim] Tam, as the new leader of our mainstream products business unit.

Endrim is responsible for the wirebond businesses, and reports directly to me.

He and his team, working with our factory managers and sales force, will grow and improve the wirebond business.

They will leverage Amkor's cost structure, strong technology base, solid customer relationships, and long-term commitment to the business.

We offer a compelling value proposition for a number of customers, including those selling into the automotive, high-rel, analogue, and industrial markets.

We will reach out to new customers in addition to growing our business with current customers.

Through our purchase of Toshiba's Malaysia factory, we entered the power discrete business in the third quarter.

The integration is going well, and customer interest is strong.

We have a multi-year supply agreement with Toshiba, and are focused on securing business with other customers.

The acquisition has been immediately accretive to earnings.

J-Devices, our 60% owned joint venture in Japan, is currently generating run rate revenues of just under $1 billion per year.

Although we treat J-Devices as an investment today, we expect to begin consolidating their results in 2016 when we have option to increase equity interest to 80%.

In the meantime, we will work closely with the J-Devices team in a number of areas, including joint purchasing programs, and joint technology development.

J-Devices has a very strong automotive business, and a competitive cost base.

With their recent acquisition of assembly and test factories from Fujitsu and Renesas, J-Devices is executing a consolidate and fill strategy, which will improve their possibility.

Looking ahead to the fourth quarter, our sales are expected to be slightly down sequentially.

Although we will benefit from a full quarter of results from our power discrete business, we anticipate seasonal declines in consumer and computing markets.

We expect that our sales into the high end smart phone and tablet market will be flat to slightly up quarter-on-quarter.

Fourth quarter gross margin is expected to be in the range of 17% to 20%.

We anticipate earnings-per-share to be between $0.08 and $0.19.

We are on track to meet our new 2013 capital additions target of roughly $450 million.

As I discussed last quarter, we established this lower target in response to changes in the high end smart phone market.

We recently completed a comprehensive review of our manufacturing capacity, and concluded that our current manufacturing footprint will be sufficient into 2016.

Therefore, we have decided to begin construction of K5, our inchon-based factory and R&D center in the fourth quarter of 2014.

This is one year later than the plan I discussed in my last conference call.

Construction work will continue into 2015 and early 2016.

This new construction schedule will improve our 2014 free cash flow outlook.

Looking at our performance year-on-year.

Assuming that we achieve the midpoint of our fourth quarter guidance, 2013 revenues would be nearly $3 billion, an increase of more than 7% year-on-year, and an all time record for Amkor.

Gross margin, excluding patent litigation charges in both years, would be 18.4%, an increase of 120 basis points year-on-year.

And earnings-per-share would be $0.50, an 11% increase over 2012, again excluding patent litigation charges.

With that, I will turn the call over to Joanne.

Thank you, Steve, and good afternoon, everyone.

To begin, third quarter sales of $768 million increased 10% over the third quarter of 2012.

Flip chip and wafer level packaging sales grew 3% year-over-year, and test sales were up 42% year-over-year.

Strong growth in mobile communications was partially offset by softness in consumer electronics.

Wirebond sales increased 9% year-over-year, driven by NAND memory and our recently completed power discrete acquisition.

The year-over-year declines in consumer electronics were in part due to decreased sales from gaming.

Our packaging from the CPU and GPU gaming chips largely ended in 2012.

Our 2013 revenues have reset to this new run rate.

Looking sequentially, sales grew 3%.

Wirebond sales and test revenues were each up 14% due to NAND memory and our power discrete acquisition.

Flip chip and wafer level processing sales decreased 10%, due to the slow down in mobile communications.

Gross margin of 18.4%, included 130 basis point impact, due to the litigation accrual.

Excluding this item, gross margin of 19.7% was an improvement of nearly 300 basis points from the third quarter of 2012.

Gross in wafer level processing, NAND memory, and test businesses primarily drove the improvement in growth margin.

Looking sequentially, the improvement in growth margin was driven by increased utilization, efficiency improvement, and lower gold wire costs in our wirebond business.

Our operating expenses of $83 million were up from $80 million in the second quarter.

We expect operating expenses to remain at this level for the fourth quarter.

Our effective tax rate in the third quarter was 32%, driven by a $5 million charge for discrete tax items.

For the full year 2013, we expect an effective tax rate of around 10%, primarily reflecting the benefit of favorable discrete items we'd recognized in the second quarter.

Our capital additions were $97 million in the third quarter, and around $70 million remains to be spent in the fourth quarter.

Also in the fourth quarter, we will make the final payment of around $50 million for the purchase of the land for our new factory and R&D center in Korea.

Moving onto our liquidity and capital structure.

At September 30, we had $591 million in cash and total debt of $1.7 billion.

We have $555 million in available revolving credit line and undrawn secured term loans.

We anticipate drawing less than $25 million under these facilities in the fourth quarter.

In summary, as Steve discussed, 2013 is shaping up as a solid year of improvement and our financial performance, with significant growth in revenues, gross margin, and earnings-per-share.

With that, we'll now open up the call for your questions.

Operator?

Thank you ma'am.

(Operator Instructions)

Vishal Shah for Deutsche Bank.

This is Chad Dillard on the line for Vishal.

Thanks for taking my question.

Can you provide us with an update on your share and your end markets wirebond flip chip test, and how you see those in end markets developing in 2014 verses 2013?

Yes Chad.

Let me provide you a little color there on market share I think is your question, right?

Right.

Yes, so we do track market share on a quarterly basis and on a yearly basis.

And we typically find the best way to look at this is looking at the last trailing 12 months, and see if there are any significant movements.

Right now, we believe overall we're holding market share.

I think when you look at it quarter-to-quarter, we did take a hit because we don't participate very heavily in the mid-range and low end of the cell phone market and the smart phone market.

But I think as we move forward over the next two quarters, you'll see that we'll actually gain share in all likelihood.

So I think moving forward, our objective is to gain share in all areas.

But right now, I think we're holding our share.

Got it.

Thanks.

And I know it may be a little bit too early, but could you talk a little bit more about your CapEx plans for 2014 against the backdrop of postponing the facility until the end of the year?

Just if you could compare what it's going to look like verses this year.

If you take a look at our capital spending for the past four years, we typically spend somewhere in the range of $450 million to $500 million a year new capital additions.

So that's excluding spending on new facilities such as K5.

I think for purposes of this call, we could assume we'll be in the same range for 2014.

Our intent is continue to spend to maintain and improve our leadership in the high technology areas of the business.

That requires us to spend in both assembly as well as in test to keep up with the needs of our mobile communications customers.

But we also have the intent to make better use of our existing platforms.

And so we have a substantial effort underway to increase the usage of our current test platforms in particular.

Great.

Thank you.

Terence Whalen with Citigroup.

Good afternoon.

Thanks for taking my question.

Hi Steve.

Hi Joanne.

So this question has to do with a comment that you made Steve about increasing the strength of the field force in China and Taiwan in order to win additional business in that area.

My question is, can you talk a little bit qualitatively how that's going?

I think you highlighted that in your comments, and I just wanted to understand fundamentally how you're thinking about metrics and measuring that.

And then, the follow-on to that question is if let's say revenue grows 10% next year, what would OpEx grow given that you have to allocate additional resources there?

Thank you.

Okay.

So I'll talk about the plan in Taiwan and China, and then I'll turn the OpEx question over to Joanne.

So how is it going?

So just to recap, I think we've done a lot of the hard work already in Taiwan and China.

We have three plants in Taiwan, a large factory in Shanghai.

We have a lot of boots on the ground in those two areas.

What we haven't done in the past is really invest in our customer facing organization.

But that really encompasses three areas, one is sales, the second is customer service, and the third is technical support.

So we've basically empowered our manager in that region to hire as many people as he needs, and we're already in the process of hiring technical talent.

We're basically looking for package designers who can be embedded with the customer, as well as package designers who will work at Amkor facilities.

And we're looking for basically a higher level of support -- support that's equivalent to what we currently provide our customers in the US, Japan, Europe.

So we have a pretty good standard to measure it by.

And you asked me for a measurement tool.

We're going to be measuring it against our existing support that we currently have in the other regions.

On the operating expense front, obviously, every new year we start off trying to figure out how to do more with less.

So we will look to drive down operating expenses in other areas.

As a reminder, this year, we had some expenses for some acquisitions that obviously hit us mostly in Q2.

We had some litigation expenses that are hitting us this year as well.

So we'll look to hopefully maintain overall operating expenses at this level.

Over time, I do expect R&D expenses to continually go up, and hopefully we can control SG&A to where we are today.

Okay, great.

Thanks Joanne.

My second question, my final question is, you kind of alluded a couple times I think in your prepared comments to processing the challenge of the high end smart phone inventory collection.

I just wanted to understand, obviously, you see different layers of demand, whether it be near-term orders of forecast, you see that out into the different levels of flies.

And I was trying to understand how you're thinking about that going forward.

Whether you've seen stability there?

Whether you've seen any increase in [decorens], or how you think that will develop heading into the first half 2014 based on just in general the smart phone inventory collection that you're experiencing?

Thank you.

Sure.

Let me make a couple comments there.

We saw the correction in June of this year until we made an adjustment there to our Q3 and Q4 plan in June.

And quite frankly, we haven't seen any major adjustments since that time.

So the short answer to your question is, I believe that the situation stabilized in the July time frame.

And so, I would characterize the market as somewhat sluggish, but it's not down.

In fact, we expect to be flat quarter-on-quarter, perhaps slightly up.

We're exposed to multiple areas to the smart phone market.

Not just the actual processor, but we also -- it's actual memory, power management, men's products, and a variety of analogue and micro controllers that go into smart phone and tablet platforms.

So we've got a pretty good penetration in the high end of the smart phone market.

We expect at some point in 2014 things will get better again, and the market will resume the growth rate that we saw in the first half of 2013.

Okay, great.

Thanks, Steve.

Appreciate the insight.

Take care.

Thank you.

John Pitzer with Credit Suisse.

Hi.

This is Farhan asking a question on behalf of John.

My first question is regarding the leading edge capacity.

Amkor has always been a leader in leading edge packaging, and there is a 20-nanometer ramp next year.

And it seems like starting first quarter of next year, there seems to be ramp on 20-nanometer that foundries are indicating.

I was just wondering relative to packaging, have any of the 20-nanometer decisions been made yet?

And how should we think about your share on 20-nanometer packaging next year?

Yes.

So on 20 nanometers, we are working on that and we have been working on that for some months now.

And we're very happy with the progress so far.

We're seeing good yields, engineering lots, and qualification lots that are passing first time through our line.

So I think we're in good shape on 20-nanometer.

And this is really a direct result of the investments we made over the years in this advanced technology, both on the assembly and the test side.

Moving forward, I would expect to have as good a share as we've had in the recent past in 20-nanometers as we have in 28-nanometers, which is pretty healthy.

But we're also trying to do a little better, so our internal objective is to move share higher, but our expectation is to at least maintain what we have.

Got it.

And then in regards to CapEx for your K5 facility, you're pushing out.

I just want to understand, is there some change in demand that you're seeing from your customer?

Was it something -- how are things different now versus when you had planned, which is causing you to push out the CapExs?

One of your customers indicating that they're not going do the business that they were previously planning with you?

I think the biggest issue here is -- and it's not an issue, it's actually a good thing.

And we've been able to make significant improvements over the past year in our own efficiency, in our current bumping locations, and in our current assembly test locations.

The output has improved significantly.

And when able to basically transfer learnings around the Company in a pretty efficient way.

So, our internal capacity is a lot greater than we thought it would be a year ago.

So looking at that, as well as the demand picture in the smart phone area in particular, it just seemed prudent at this point to move the construction schedule out one year.

Now, if it proves that we're too conservative, we do have the ability to start that construction earlier.

As Joanne mentioned, we have purchased the land, and when we say go we can get started.

Got it.

Could you also provide utilization rates on your test and packaging?

I didn't see that on the press release, and I was just wondering what the utilization rates were on this quarter.

As part of Steve's initiatives, we're really working very closely on the shop floor to really challenge what our current capacity is.

As we go through this reassessment of capacity, those utilization numbers we started to have some challenges with respect to consistency between period-to-period.

I would say from a like-for-like basis, the packaging was in the 80s -- low 80s as well as test was.

The wirebond assets were actually fairly well utilized led by the NAND memory, as well as some of the strength in the mainstream that we saw.

And we had some open capacity as you could expect on the flip chip on the advanced side principally going into the smart phones.

We have strong attach to test with NAND.

So, a lot of those test assets were well utilized in support of NAND, less so in support of the smart phones.

Got it.

And then, I'm also trying to understand you had a very strong growth in your wirebond business.

If you could help us break out the discrete portion or give us a sense of how much of the revenue growth was coming from discrete or the Toshiba business that you acquired in Malaysia.

That would help us reconcile what your core growth was for the wirebond versus what you grew inorganically.

Absolutely.

So our wirebond sales in total for Q3 were $351 million, of which memory was $31 million and compared to $308 million a lot in Q2 was wirebond.

Got it.

But in terms of --

$31 million is discrete.

Okay, got it.

Sorry, $31 million is discrete.

Total wirebond including discretes, is $351 million for Q3, $308 million for Q2 2013.

Okay.

And it seems like NAND was also a very big driver of growth for you.

Approximately how much of your revenues, even ballpark, would be great are coming from NAND so that we can think about the business?

Yes.

Our memory exposure is about $100 million.

Got it.

Thank you.

That's all I had.

David Dooley with Steelhead Securities.

Could you help us understand what the impact of the pick up and revenue is from the Toshiba acquisition, or help us understand what the guidance is apples-to-apples?

However you'd like to break it down.

Okay that's very similar to the question we just had, so discrete which was the acquisition from Toshiba is $31 million for Q3.

For Q4, we expect it to be about $39 million for Q4, some level of seasonal correction and starting to see some of the cost savings impacting the price.

So that would be the apples-to-apples Q3 to Q4.

So then the revenue guidance for the quarter, I think it's probably better than normal seasonally because this isn't that big of a change.

Why is the revenue guide in Q4, I guess, better than the seasonal?

Or maybe I'm doing the math wrong.

It's about seasonal, so just to go ahead and do the math Q3 without discretes was $737 million, and at the midpoint without discretes it would be about $715 million.

So down just about 2%.

That tends to -- when we haven't had a real typical Q4, but when we did our calculations we normalized out some of the 28-nanometer RAMs, some of the movement in gaming, which suggested over the last three years Q4 on average is down 2%.

So our decline in revenues, once you back out discretes, feels very seasonal.

Okay.

Final question from me is, on the gross margin front.

Gross margins were up sequentially without the litigation reserve, and your communication business percentage was down.

Could you just talk about the dynamics there?

Yes.

So it was a very positive quarter from a growth margin perspective, especially when you start adding back the discrete charge for the litigation.

What was the drivers was there was a lot of positive news on the mainstream side.

We started to see a lot better utilization on the wirebond front than we've seen in many quarters, which while overall wirebond is lower than the corporate average movement is up from the margins associated with the wirebond it has a favorable impact on our gross margin.

So better utilization.

We started to see positive results from some of our cost initiatives getting larger strips going through our machines, which means that we'll get more units going through each of the machines, which helps again on the initiative of getting more capacity out of our installed base.

And then we did have the benefit of the lower gold prices this quarter compared to last quarter.

Thank you.

I'll re-queue.

Great.

Thank you.

Jairam Nathan with Sidoti.

Hi.

Thanks for taking my question.

The first question was on, you mentioned some new business wins.

And I just wondered, Steve, if you could give some more details and whether it's a new customer or a new program with the same customer?

Okay sir, are you referring to the wins in Taiwan and China?

Yes.

Yes, so those were new customers for Amkor.

And I can't give you their names, but I can tell you when you think of top six fabless Taiwan China companies, they're in that grouping.

And we have more in the pipeline.

So, we have some interesting technology to offer to these companies.

We installed this technology really years ago, and so we're down the learning curve from a technology standpoint, and we're also down the cost curve.

So what we're finding is our quotations are very, very competitive.

So I think if we build up our service organization, we have all of the other pieces in place to be successful here.

Okay, thanks.

And as far as the Toshiba business, how do the gross margins on that business compare to your existing legacy businesses?

And was that one of the reasons -- was that a factor in including incrementals in legends?

Yes, I'll make a comment here and then Joanne may add a few words.

But I think we have the opportunity to make a decent gross margin in Taiwan and China, just as good as we do in other regions of the world.

I think a lot of it is how you view the business, and how you view your assets.

But I'm confident that we can engage heavily in this area, and continue to improve our margins.

And your question, was it specific to the Toshiba acquisition and the discrete business?

Yes, yes, yes.

Okay.

So with respect to the power discrete business, you can think of those gross margins very similar to our mainstream business.

It's very similar to our leadframe package technology.

So it's in the low teens.

On our last call, we talked about net margin contribution from this acquisition was going to be about 5%.

In the initial years as we run through some of the purchase accounting associated with it, it may be closer to our corporate average of 4%.

And just one thing for everybody on the call to understand is, is we broadly service Toshiba.

So Toshiba is one of our NAND memory customers as well.

So with respect to Toshiba, margins overall they're what you would expect across a very diverse portfolio.

Okay.

And my last cost question, and sticking to gross margins here, your guidance for the quarter of 17% to 20%, is that really primarily on volume?

You already did close to 20% with flip chip as well as packaging revenues not doing well, so how should we think about that?

Yes.

No, that's a great question.

With respect to the gross margin guide for Q4, that is based on the expected decline in the legacy business prior to the acquisition.

So we talked about the Q3 we're at $737 million without the acquisition.

We expect to be down at the midpoint to $715 million.

So there is a decline with the decline in revenues that would impact our gross margins.

When you start considering that our guide does include the acquisition, the discrete business because we had talked about it being in the low teens will average down our corporate gross margins as well.

So that it's about a 50 basis point compression on the margins.

So hopefully that bridges a little bit about how at the mid point we'd be at a 18.5% growth margin compared to 18.5% this quarter, but, which obviously includes the litigation accrual.

Okay.

Thanks.

That's all.

Krish Sankar with Bank of America Merrill Lynch.

Thanks for taking my question.

I had a couple of them.

Either Steve or Joanne, off your wirebond or installed base, what percentage is copper wire bonded today?

With respect to copper wire, a lot of our -- all the advanced copper wire bonders are capable of doing both gold and copper.

And with respect to gold verses copper, all by far gold is still the predominant piece of what we do.

And copper is about 20% of our wirebond revenues.

That's the wirebond revenue.

What about is there a way to quantify it in terms of capacity, or is it a close enough proxy?

From significant trends, I wouldn't highlight copper as being a significant trend for us.

Got it, got it.

And then I had a follow-up question your litigation costs.

You guys said that you pockmarked about $60 million to $115 million.

I'm just kind of curious.

I'm guessing this is in regards to the Tesla arbitration.

How much have you paid them so far?

How much has been earmarked in total for this litigation?

So we did increase our reserve to $60 million, which represents the low end of the range.

And as you said, as was indicated in our press release, the range of potential loss was narrowed to $60 million to $115 million.

Of that $60 million, to date we've paid about $20 million in connection with this arbitration.

And as far as earmarking, from a liquidity standpoint, I feel comfortable that even at the highest end of the range we're covered from both cash as well as an available credit line.

Got it.

And then ask a follow up question on this.

To the extent that you can answer this, your press release said that there's a new entry model from the arbitration panel in October.

So is it fair to read that the new interim order in October was more in favor of Tesla versus you guys, or am I reading into it too much?

I don't know how you're reading that.

I think you are reading a little bit into it.

How I would characterize it, is that the range narrows, both the high end as well as the low end.

So both ends of the range narrowed.

Got it.

And what was the range before?

We have disclosure in our 10Q, so I would refer to our 10Q.

We're about to file the updated Q10 at the end of the week.

So, it's fully disclosed there.

Got it.

Thank you very much.

Suji De Silva with Topeka.

Mr. De Silva?

You might want to check your line, and see if you have yourself muted, Suji.

Operator, why don't we take the next caller and maybe Suji can re-pull.

[Aneve] [Tankade] with Hutchin and Hill.

Hi.

I appreciate you guys taking my question, and congrats on a solid quarter.

I think my question sort of more related to like a big picture, free cash flow for 2014.

You obviously moved CapEx around a little bit, and you're sort of seeing the ramp in 20-nanometer.

So in 2014, how do you think about free cash flow?

Are you going to be break even or positive or still kind of the negative side?

So with respect to free cash flow for 2014, as Steve had mentioned, that while we haven't finalized our CapEx plan for 2014, that if you assume it's in a similar range as this year and our prior spending.

And then you'd give consideration to the delay in the K5 timing with some of the construction happening in Q4, I do expect that we'll be free cash flow positive in 2014.

So that is our current expectation is that we'd be positive in 2014.

Which is different from a quarter ago, when we thought we wouldn't be cash flow positive until 2015.

That's great.

And sort of more again big picture.

There has obviously been a little bit talk on consolidation in the space, there's a lot of players in the market.

How are you guys thinking about that?

I'll make a comment here.

I think we'll continue to do or at least evaluate plug in opportunities.

And just as a reminder, J-Devices, our joint venture in Japan, is a great way for Amkor to solidify its position in the Japanese market.

So that's a geographical plug in opportunity for us.

The Toshiba acquisition in Malaysia, it's a great way for us to expand our TAM, moving into a new market that we've never addressed, power discretes.

So we'll continue to look for opportunities like that, but they're more plug-in in nature.

I think as you look at the larger opportunities of combining O stats, it becomes a little more difficult.

Because typically, our largest customers like to allocate their business amongst a number of different OSAT providers.

So 1 plus 1 does not always equal 2 when it comes to merging OSAT suppliers.

Got it.

That's it from me.

Appreciate your time.

Thank you.

David Dooley with Steelhead Securities.

Thank you.

Just curiosity, it sounds like you're flip chip and wafer bump utilization rates are below 80% range of your other businesses.

So, I'm kind of curious why you would need to spend money there in 2014, and therefore it seems like your CapEx number that you're talking about seems high.

Could you just talk about that a little bit?

Yes, I'd be happy to.

So the good news is, we're ready for the up turn.

Because we built this capacity in Q2 in anticipation of up turn in Q3, so you're right.

We've got some empty capacity sitting there.

But there's some more good news.

We noted that and we basically accelerated our new product introduction program.

So the number of NPI lots went up significantly, over 10% quarter-on-quarter.

So Q2 to Q3 number of NPI velocity processed went up by 10%, and the cycle time went down by 12%.

So this is always the silver lining when demand goes down for a quarter or two.

So we're trying to take advantage of it.

I think what you'll see next year is a year that may be back-end loaded.

We're engaged in a number of leading edge programs that will begin to kick in towards the middle of next year.

Some of these programs require us to purchase assembly and test equipment, which we don't currently have on hand.

So that's what's going to drive spending in the advanced area.

Okay.

And did you have any 10% customers during the quarter or thus far this year?

When you refer back to our 10K, we did have one customer over 10%, which was Qualcomm, and they continue to be.

Okay.

And just a final clarification from me, and I'm sorry if you've talked about this previously.

When you look at your flip chip and wafer bump business, you don't manufacture your own substrate, so there's some sort of pass through revenue impact on your margins.

Could you give us a good shot at what that impact might be?

Yes, sure.

You are right, we do not manufacture our own substrate.

I think the best way to look at it is in the press release in the selected operating table, we do break out our cost of sales between material labor and other manufacturing.

And materials run for the quarter was 39%, and that's as a percent of revenue.

Okay.

And so that represents the vast majority of that as the substrate pass through?

That's right.

That's the most significant cost.

Okay.

Thank you.

Terence Whalen with Citigroup.

Hi.

Thanks for taking -- I had actually two follow-ups.

The first is a very simple one.

What's your expectation for memory packaging going forward the next couple quarters?

Obviously, it's (inaudible) I just wanted to understand what was impounded in your guidance going forward into the next quarter.

The second half of your question was breaking up, so I think the answer was over the next couple quarters what's the expectations for memory.

We do expect 4Q to be up modestly.

With respect to Q1, Q1 overall tends to be a trough quarter for us.

So I don't have any separate color with respect to memory for Q1.

Okay.

Great, and thanks.

And then my second question that I had was a very high level question.

And that is as the high end smart phone market comes under price pressure and as potentially architectures go to more integrated-based bands and application processors.

I wanted to understand what the implication is for your business in that sort of a scenario in terms of each of the available market.

Thanks.

Yes.

I'll make a comment there.

I think we're already dealing with customers who have accomplished that integration level, and namely Qualcomm.

That they're a leader in the industry, and we're working very closely with them as a technology partner in assembly and test.

So, I think speaking more broadly, what this means is there will be a greater push towards 2.5D and 3D technology over the next couple of years.

So the important thing to realize is, Amkor has been developing the enabling technologies for many years.

I mentioned earlier about this fine pitch copper pillar.

That's really key to this whole TSV approach, which you can implement at both the 2.5D and the 3D level, so we have all the parts.

The tricky part right now is meeting the cost targets of high volume smart phone and tablet manufacturers.

So we're working very closely there to try to eliminate all the costs we can, and to make these solutions feasible.

Thank you.

(Operator Instructions)

Suji De Silva with Topeka.

Hi guys.

Can you hear me?

Yes, Suji, sorry for bouncing you earlier.

That's okay.

So, the first question is on gross margin, I apologize if you've already covered this.

But should we think about the target longer term, Joanne, being more of a range, or is it something that's going to trend upward through the cycles as you are able to connect the improvements you're trying to do?

Before Joanne comments, let me just make a comment.

Sure, Steve.

My objective is to get us into the 20%s and keep us there from a growth margin standpoint, so that we'd be able to stay there through the peak periods and the trough periods.

So that requires two things.

It requires us to continue to build and improve our wire bond business, and it also forces us to continue to grow and invest in the advanced packaging and test business.

So we'll improve the mix, but we'll also make sure that our legacy business is healthy.

Over the -- the only thing I would add is, over the short run when you look to Q1 and Q2, Q1 tends to be our trough and you start to build back Q2.

So some of the margin improvement as to when will you see more sustained margin improvement, starts to come in the second half of 2014.

Okay, great.

That's great color.

And then also on the wirebond business, can you talk about in the 2014 time frame, do you expect to be a share gainer there?

And can you just talk about competitive environment in the context of that question?

Yes, it's very difficult to predict.

I think the wirebond market will undergo some changes next year.

I think looking particularly into the Taiwan China region, many of the customers who are using wirebond technologies to bring ICs to market for smart phones and tablets will begin to transition to wafer level technologies and to flip chip technologies.

So, I think some of the market may disappear next year.

So I think through a combination of our efforts to gain share and some decrease in some of the smart phone wirebond demand, that we'll actually gain share in wirebond next year.

Great.

Thanks for the color, guys.

Okay.

Thanks very much, Suji.

There are no further questions, so this completes our call.

We would like to thank all of you for joining us this afternoon, and have a good day.

Ladies and gentlemen, this concludes Amkor's recorded conference call.

Thank you for your participation.

You may now disconnect.