艾克爾 (AMKR) 2004 Q3 法說會逐字稿

Good afternoon, ladies and gentlemen, and welcome to the Amkor third quarter 2004 earnings conference call.

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

Following today's presentation, instructions will be given for the question-and-answer session.

If anyone needs assistance at any time during the conference, please press the star followed by the 0 for an operator.

As a reminder this conference is being recorded today, Tuesday, October 26th, 2004.

I would now like to turn the conference over to Mr. James Kim, Chairman and Chief Executive Officer.

Please go ahead, sir.

Thank you.

Thank you for joining us today.

I am James Kim, Chairman and Chief Executive Officer of Amkor.

With me today are John Boruch, President and Chief Operating Officer, and Ken Joyce, Chief Financial Officer.

I will make some brief remarks, Ken will discuss our operating results, and then John will have some closing comments.

At the beginning of this year, there was a general consensus that 2004 was going to be a strong year for the semiconductor industry, and this optimism was certainly reflected in our customers' forecast.

Indeed, the first quarter was by historical standards very strong quarter for Amkor.

In the early part of 2004, we were presented with the opportunity to pursue several long-term growth initiatives, including a 6-year supply agreement with IBM, a new factory to accommodate our growing business in Taiwan, and the Unitive acquisition.

We faced a difficult choice.

We could have decided to pass on these opportunities, and instead view our cash position with a view towards reducing debt.

Given the backdrop of a strong customer forecast, and the generally-held view that the industry was headed for strong growth year, we elected to think long-term and pursue these growth opportunities.

So we moved forward with these initiatives in connection with an operating plan that was based on expectations of robust industry growth.

It has become apparent that over the past several months the semiconductor industry is experiencing a correction.

It has also become apparent that separate from the industry correction, we have lost some market share.

All of the above mentioned factors have put added pressure on our financial performance.

Our visibility into customer demand is cloudy, macroeconomic factors suggest that the correction should begin to reverse by mid-2005.

We fully believe that our supply agreement with IBM, our acquisition of Unitive, and our factory expansion initiatives are right moves for Amkor, and that they will position us for increased market share and exceptional long-term growth.

We have taken steps to ensure that we have adequate financial liquidity due to the industry correction.

We remain focused on providing the right assembly and test solutions for our customers, together with the highest level of service and support.

Ken Joyce will now review our financial results.

Ken?

Thank you, Jim.

Before we discuss our financial results, I would like to remind you that any forward-looking statements made during the course of this conference call represents the current view of Management.

We refer you to today's Press Release which was filed with the SEC Form 8-K prior to this conference call, and our other filings with the SEC for information on risk factors that could cause actual results to differ material from our current expectations.

Third quarter sales came in closer to original guidance than we expected when we issued our readvised outlook on September 29th.

In our view, the strengthening of die support we experienced in the last week of September does not necessarily reflect the general improvement in this business condition.

During the quarter, we experienced a unit reduction in number of package families.

By far, the greatest unit reduction was in our low ASP LeadFrame Business.

We had a much lower unit reduction in our higher ASP laminate packages.

Looking ahead, we will continue to absorb factory overhead and other costs associated with the expansion of production capacity, our new operations in Taiwan and Singapore, and the Unitive acquisitions.

Capital expenditures were 73 million for the third quarter.

We are targeting CapEx of approximately 32 million for the fourth quarter, bringing the full-year spend to around $400 million.

As mentioned in our earnings release, and as previously disclosed, we are structuring a $300 million credit facility to provide additional liquidity during the industry downturn.

The credit facility provides for a nonamortizing term loan with a 6-year bullet payment.

The loan is expected to close tomorrow.

Following the scheduled payment later this year in connection with the IBM transaction, we have no significant debt maturities until May of 2006.

Here is a recap of our fourth quarter guidance contained in our earnings release.

Revenue should be down between 5 and 10% from the third quarter.

Gross margin should be in the range of 15 to 17%.

We expect a fourth quarter net loss in the range of 16 to 23 cents per share.

Now I'll turn the call over to John Boruch for some additional comments.

John?

Thanks, Ken.

During the third quarter we moved forward with several key strategies.

We completed the Unitive acquisitions and now have what we believe is the industry's leading technology for electroplated wafer bumping.

We are integrating the Unitive operations in North Carolina and Taiwan to provide a complete turnkey solution for the rapidly growing market opportunities in FlipChip and Wafer Level Packaging.

We completed the initial integration of our recently acquired IBM test operation in Singapore, and have been expanding our tester-base in response to increased business.

We expect this operation to build critical-mass during 2005, and support not only IBM, but also third-party semiconductor companies.

We expect our [selling] business under the IBM supply agreement to commence in January 2005.

Of course, as with any of our customers, the pace of that business will depend on overall market conditions.

In China, we have completed facilitation of the remaining 75,000 square feet of our existing factory in Shanghai to accommodate anticipated business growth.

We have no immediate plans to [facilitize] the new building we are acquiring in the IBM transaction.

However, that building will accommodate our long-term expansion needs in China.

As both Jim and Kim have indicated, our visibility is cloudy, and this is evident in what our customers have been saying over the past few months.

It is always difficult to predict the timing of an industry correction.

We will keep a sharp-eye on die support levels and customer forecast.

Meanwhile, we have adequate capacity on hand to accommodate a substantial growth -- amount of growth in our core package families.

We will continue to increase capacity in FlipChip, Wafer Level Packaging, and tests, and we will continue to invest in R&D.

Notwithstanding the industry correction, technology continues to advance and we are committed to maintaining our leadership position in developing [die] packaged solutions that can accommodate leading edge fab processes.

Operator, we will now open this call to questions.

Thank you, sir.

Ladies and gentlemen, [OPERATOR INSTRUCTIONS].

Eric Gomberg, Thomas Weisel Partners.

Could you talk a little bit about IBM and that business that will be coming on in January, if there are associated costs that we should be thinking about?

And if that business is also seeing an inventory correction that will result in perhaps a slower revenue ramp than you might have previously thought?

And what type of margin impact that business will have to the Company?

Okay.

This is John.

Yes, Eric, the IBM business will start in January.

We've lined up a number of qualifications et cetera to support that business transfer.

Obviously, it will not come all overnight.

The amount of capital required to support that business is minimal because of the large capacity we have installed in 2004.

The margins in that business are generally in line with our standard margins for the particular products that will transfer to us.

And, yes, as I said in my earnings call, that IBM will be influenced by the general market as all customers, and it's hard to say what kind of impact that will have on the total amount we were looking for in 2005.

But we don't expect anything appreciably different now, but obviously, it will have an impact.

Thanks.

And just a quick follow-up.

Could you maybe give us what the utilization was in the quarter, where you expect it to be in Q4, and what pricing trends were and look like for Q4 as well?

Thanks.

We are a little bit above 70% utilization based on the total potential revenue of our worldwide capacity.

So basically we can produce about $2.4 billion worth of revenue with our existing capacity, or more, if it was a perfect mix.

On the -- what was the second part of that question, Eric?

ASP pressure?

Yes, on the pricing pressure, sure, we have extra capacity, our competitors do [indiscernible] fallen.

So pricing pressure is increasing, and it will show up in different places as we go forward.

So that's all I can say about that right now.

Jeff Harlib, Lehman Brothers.

Can you just talk about a little bit what you're seeing in market, breaking that out in the different end markets demand?

Well, we've given up trying to figure out how much of our business is consumer versus computer versus phones.

It all starts blending into the same stuff.

All I can tell you on there is that we didn't see any strong markets in the third quarter and now.

Probably the strongest of all of them was the computer industry, computer activity, so -- and activity-based around that, and that was spotty, but generally sounded.

And then of course, in the wireless cellphone, that continued to be very weak going through the quarter and as we speak.

Okay.

And just a follow-up.

Are you seeing in the commodity packages some of your customers bring packaging in-house as their utilization declines?

Have you seen any of that?

Or do you expect that?

Yes, there's some of that happens with large IDMs.

We expect that to happen in times of weakness, where they have some of that general capacity in place.

Plus on the standard packages, SOICs and thinks like that, it's a little more inventoriable if you will.

In other words, those are general logic and analog kind of parts that don't suffer severe obsolescence, so our customers tend to buy a little more stock in that.

So we see wilder fluctuations on the upside and downside in those older packages as compared to the leading edge packages.

John Pitzer, Credit Suisse First Boston.

A couple of questions, John.

First you talked a little bit about die support picking up in the end of September.

Can you just go into a little more detail about what you thought happened there, or what caused that, and what the trends look like going into this quarter?

And then secondly, if you look now, you kind of have 2 quarters of below normal seasonal growth for the Company.

Do you think this correction has kind of played itself through so we're looking at a first quarter that's showing normal seasonal?

Do you think this is going to bleed into the first quarter as you're looking at the die bank inventory that you are holding now?

Thanks.

Okay.

On die support, yes we saw die support strengthen towards the end of September.

And we are looking -- as we've talked about over the years, we look for that to be a leading indicator of potential -- at least troughing or maybe not an upturn, but at least a stabilization of the fall.

So 2 swallows doesn't make a summer, 2 weeks or 3 weeks doesn't make it a trend.

But we'll be watching that as we go forward, and all I can tell you is that's what happened the last part of this quarter, and we'll see what it looks like going forward.

John, was it across the board or was it in market specific?

It was pretty general, John.

By the way, John, Jim just reminded me.

It was because of the -- it was a little more strength in the advanced packages, Stacked Chip and MLF and the BGA packages as opposed to the LeadFrames I said, were falling rapidly throughout that quarter.

On the seasonal growth question for Q4, we are in a downturn, plus we have our normal weak December, so you saw our guidance down 5 to 10.

We don't really know and that's how it looks.

It will be front-loaded, which is good for the quarter.

In other words, we expect October to be stronger certainly than December, and October is over, so we can tally up that's what our forecast assumes at this point in time.

On Q1, how we'll play this cycle is it's a mid-term correction.

We don't see huge inventories out there that are ahead of the assembly activity.

We saw numbers here recently that talked about maybe a $1 billion of added inventory in Q3 in the semiconductor supply chain.

And looking at that, we may be wrong, but it appears that a lot of that is in wafer form.

So it doesn't seem to be a whole lot of finished goods in the pipeline.

In some cases for some customers, true, in the some markets, yes, but in general, for correction, it seems that that inventory position seems to be pretty tame to us.

And a lot of -- and based on our die banks going out of sight in Q3, we knew that our customers were building -- they stopped the assembly activity and the test activity because they didn't see any demand.

But they kept building wafers because that has a lot more inertia than assembly, and the die banks have built.

Yes, especially in the advanced area.

So for Q1, our call would be a normal seasonality.

This thing troughing out as we go to 2005, and looking for growth resuming late Q1 and Q2 2005.

Anybody's guess, but that's how we're playing it at Amkor today.

Ted Parmigiani, Lehman Brothers.

Just 2 questions here.

Ken, on the $300 million facility, how much of that will be tapped in the current quarter, and what kind of performance covenants, if anyone, does that loan carry?

And then as a follow-up it would be on the break-even revenue range, what your best guess would be at this stage?

Thanks.

As far as -- we really shouldn't have to use the 300 million in this quarter.

We have the money on our balance sheet to handle our IBM transaction.

With respect to the break-even question, our breakeven right now is around 515 million per quarter, depending on what the mix is, of course, and what was the third question?

Yes, the covenants that the new term-loan will carry?

There are no maintenance covenants in there.

They'll have the same incurrence covenants that are in our current high-yield notes [08].

Maralia Aborhi, JP Morgan.

A couple of questions here.

Given the Delta between your breakeven level and your current revenue run rate, are you guys looking at any short-term cost reduction activities here?

What we're looking at, of course, is all the prudent things you do when you enter a downturn, restrictions on spending money here and there, and we're certainly doing that at Amkor.

We decided that because of our growth initiative action in 2004, we made some big acquisitions, and we need to make sure that we don't continue suffering market share going forward.

So at this point in time, we are not planning on any major reductions in overhead other than controlling it to about where it is now, and then growing forward into 2005.

Obviously, if things get worse there are actions we can take, but right now, that's how we're behaving here.

Okay.

And then if you look at your gross margins in this 15 to 17% guidance range you gave, can you highlight what are the 2 or 3 layers that your looking at to get the gross margin improvements going forward?

The biggest thing to get the gross margins improved as John had just talked about going forward.

It is not so much reducing the costs, although we're focused there, it's to increase the revenues.

And I think if we're able to do that, you'll see the leverage in the model, especially within terms of the -- we brought new fixed costs in with our acquisitions as we fill them up.

Of course, the other area that's very important to us is overall material costs, which runs anywhere from 35 to 50% of the total costs, depending on the product.

So we've been working with vendors to reduce costs and things of that nature.

So they would be the areas that we would be focused on.

Peter Wright, CIBC World Markets.

My first question is a follow-up to one of the earlier ones.

With more inventory sitting in wafer form as opposed to finished goods, it seems to me from a volume standpoint that the back-end [subcon] should be able to outperform some of the front-end foundries from a growth perspective.

Do you think that the pricing trends, pricing pressure is severe enough to offset that, or am I reading into this correctly?

Well, this is John, let me answer that with 2 parts.

One, yes, traditionally whenever inventory is sitting in finished goods, the -- our customers will outperform us on the front-side of the growth curve, because they're shipping out of inventory, and we're sitting there waiting for die starts.

What it's like -- appears to be now, where it's in wafer form, we'll see the demand as it tops up in our customers' radar screens, and we -- the back-end people will be a leading indicator of this -- the resurgence going forward.

On the pricing side, it's very, very difficult to say.

A lot of mix, a lot of products in there, on how that will play -- will offset the potential for doing better, having better growth in the industry.

We'll have to wait and see.

Great.

And 2 quick housekeeping.

Was the utilization of LeadFrame versus laminate anything significantly different?

And usually you provide us with a percentage from tests, and I was wondering if you could update us on that?

It's all in the same ballpark.

Satya Chillara, RBC Capital Markets.

John, I know it's early for 2005 at this point, from a unit growth point, can you take a stab at -- for Amkor and for the industry what kind of unit growth numbers are you looking at?

I don't make the big bucks you guys do, so I'm going to pass on calling that shot.

Okay.

That was an easy one.

In terms of the second caution, the IBM $150 million agreement each year for the next 6 years, how is that going to work out?

You did comment depending on the qualifications and so on.

So how do we start looking at IBM each year, $150 million, or how do you track that.

And just follow-up, if the market share loss, you did say the market share loss, how many percentage points have you lost?

Well, for market share, it's very difficult in this industry because of the poor data out there to put it in percentage points and mislead anybody.

Although, our feeling certainly is the indicators are based on revenue growth, et cetera, that we lost some market share in some areas last year -- this year.

What was the other question?

IBM.

Yes, on IBM, as I said, we'll start loading that business into our factories in Q1, a good indicator of how that's going is at the end up Q1 and Q2 we'll have a much better idea of what that looks like.

We have agreements with IBM on percentages, roughly, and whatever that business is, we're going to get it, so we'll see.

In addition to that we have a relationship with IBM is on the foundry side and a lot of interesting possibilities coming up on the customers there, servicing through their wafer foundry business.

Tom Diffely, Merrill Lynch.

Of the 400 million in capital spending for '04, what percentage of that was from FlipChip and tests?

I don't have that right in front of me, but on the Flipchip and test, I would say it's probably around 35% to 40%.

Okay.

And is that same percentage going to hold next year for the 100 million?

It might be more.

Stanford Nishikawa, Citigroup.

John, now that you've been back in the captain's chair for a few months now, can you give us an update on customer relationships, any new wins, any losses, that type of thing?

Well, we don't talk about relationships too much on -- specifically.

But at Amkor just a renewed focus on the customer aspect of our business and the customer service, and responsiveness and listening, and doing the things that our customers need to build their business.

So it's a whole philosophical approach to how we behave at Amkor.

I can tell you that that's one of the things that we've kind of changed a little bit here.

So we'll see how that works going forward, but obviously it has worked for most companies well in the past, and it should work here.

Jesse Pichel, Piper Jaffray.

One question for John and one for Ken.

John, when does the transition to wafer level CSP occur, and have there been any further developments with your wafer level JV with Casio given your Unitive acquisition?

Okay, on wafer level CSP, it has been occurring, it is occurring now.

The better that technology gets and we think Unitive has a key technology to enable that, the faster that transition will occur.

Four specific die sizes and applications.

So based on what we've seen now, after a few months of owning the Unitive, North Carolina, operation, lots of opportunities there, and hopefully we'll execute on those opportunities and do well with that over the next several years.

So I believe the wafer level opportunities are here and now.

On the JV with Casio, not much as changed since our last announcement.

Okay.

And Ken, correct me if I am wrong, but minus the inventory reserve and the tax catchup, you would have exceeded your original guidance?

So I'm wondering, is that right, and were these built into your original guidance, and how should we think of your tax rate going forward?

Your comment is right, we would have beat the original guidance.

As you're aware, just recently, the President signed some new legislation with respect to tax law, and how it will impact corporations.

That does impact some of our global foreign taxation structure, and so we factored that into our number there.

Sundar Varadarajan, Merrill Lynch.

Just another question on '05 CapEx.

You have a number of about 100 million for next year.

What could cause that to change on the higher side?

In other words, up to what level of revenue for '05 can we say that would cause this CapEx to be around the $100 million level and what ranges would you see any kind of increase to that CapEx guidance for next year?

This is John.

As I said before, we have a significant amount of general capacity on hand entering the year, so it will carry us through most of the year, or all of the year on general kind of capacity.

And we funded MLF and CABG and things like that very, very adequately in 2004.

So even in the leading edge, we're pretty well-covered on general capacity.

So the area that could prove interesting for Amkor is if we're very, very successful in the FlipChip bumping wafer scale level, it will require additional CapEx, and that would mean maybe we would ride above that.

But it won't be until later in the year, and the market, and our investors, and the analysts will be well-aware of the situations as they arise.

And again, just as a follow-up here.

This year we saw the CapEx being front-end loaded.

Is it fair to assume next year that it's more likely to be back-end loaded?

It could be.

I think one might have the double hump here.

If we are preparing for 2005 [ph], if we're successful in some initiatives early on in the year, then we'll have to make a few investments in Q1 here.

And then we have some big opportunities later in the year, and that would mean we would have another spike later in the year, so you might see a double hump.

Shekhar Pramanick, Schwab Soundview.

John, if you could provide a little clarification on the die support comment.

Are you seeing the die support still continuing in October, or has it kind of slowed down?

The forecast has come down.

The die support has not come down as fast, therefore the GAAP is closing between die support and the forecast, which is what you would expect when -- at -- if you want to level out the activity level and reach a trough.

So those are the signs we see.

No sure fire formula here, but that's what we see at Amkor in signifying to us that this thing -- and we don't know quite how long if it's going to trough out here in Q4, and then we'll see where it goes up.

Our call is later in Q1 and the early Q2.

So we will see if we are right.

So but overall, if you look at the monthly run rate numbers and stuff, is October still holding up similar level as September, is it lower, or higher?

I think it's held up on par with Q3, and we think November will be okay, the issue for us is December is traditionally a low month.

Will our customers have a little push at the end of the quarter?

I don't know.

I don't think so.

Folks, our call is down 5 to 10% with a pretty poor December.

Tim Arcuri, Smith Barney.

Two quick ones for you.

Number one, if you look at your CapEx guidance, what you had said last quarter you said that you were going to spend about $80 million in the second half, and it turns out that it's going to be more like 100 to 105 in the second half.

Yet the '05 CapEx plan isn't being changed.

Any particular reason why there's not any follow through from Q4 from that higher level CapEx in Q4 into '05?

No particular reason.

I think it's just the mix changing, the numbers changing, are we looking at what we needed to do in Q4 to support some of these new acquisitions we have, like Unitive in North Carolina, and in Taiwan, and some bumping in FlipChip CapEx we needed to support.

So that was a little stronger than we thought, which means that our opportunities are looking better than we thought a quarter ago, so in that area, so that's all it means.

Okay.

And then I guess quickly as a follow-up.

If you try to kind of zero in on what both Unitive and IBM might mean in Q1, and is it right to assume maybe kind of in the 20 range for IBM, and in the 10 range for Unitive?

So you're going to kind of on a nonorganic basis revenue would be up about $30 million in Q1 versus Q4?

Is that kind of the right way to think about it?

We just can't comment on specific numbers like that for Q1.

That's way to specific.

There will be a positive impact, hopefully certainly on IBM.

We are counting on that.

That will happen.

On the FlipChip, we hope that will happen, and whether that will counter act the seasonality, down seasonality trend for Q1, we'll have to wait and see.

Andrew Stearn, Katona Capital

With respect to market share that you lost, can you comment on the competitive landscape and any strategy that you are planning to recapture some of that share?

Well,, there's no reason why we can't capture some of that -- recapture some of that share.

Some of that market share loss was because we weren't playing strongly in certain areas.

We weren't playing strongly as we've been saying for our last 6 months or 9 months in the FlipChip -- excuse me, in the chip set and graphics market.

We weren't playing at all in the Qualcom CD&A business, et cetera.

We didn't do nearly as well as we thought we should have been and shame on us on camera modules.

We had plenty of opportunities.

So those are just 3 or 4 areas that I can tell you that we know we have missed -- and those are very, very strong growing markets.

And we basically didn't lose market share, we just didn't participate in those areas, and by virtue of math we lost market share to the general market.

So in areas where we play hard I don't believe we lost market share, but these are areas that we should have been in and we weren't in so -- as you know we have been saying we're trying and focused on getting into those areas strongly.

Sysal Alibhi, Chafen Asset Management.

Can you talk about the charges that you incurred?

There's 2 areas I want you to just expand on.

The $6 million in the reserves.

Is that something that stays in the cost of goods sold?

How should we look at that?

And then the $8 million in legal fees, is that something that's going to be ongoing for the new few quarters?

Both good question.

The 6 million in reserves is the -- as you're aware in many cases we buy inventory in support of customer and demand, and in the first 2 quarters of the year, it looks stronger than it turned out to be as we moved into the third quarter.

As a result, we're out there, we're -- we have take or pay contracts, or material responsibilities agreements in many where we collect on those, but that being the case, we did put up a reserve because the absolute dollars of inventory has increased over that 6 to 9-month period.

With respect to the legal reserves, those of you that follow us know that historically, our legal expenses have run less than $1 million.

A million dollars was a high quarter, somewhere around $500,000 a quarter.

As we moved into the third quarter of 2003, we became involved in some mold compound litigation.

It was running about a little over $5 million a quarter.

This quarter we had the mold compound litigation, and we also had -- we went to trial with respect to a patent infringement suit that we have with someone that we believe is infringing our MLS product line.

So those costs are up.

When will they be over?

I'm not sure.

We're currently with one of -- on the mold compound litigation, we're with one of the major plaintiffs in litigation.

It's in trial right now.

We feel we have a very strong defense and we're hopeful to prevail there.

And if we do, we're hopeful that some of the other plaintiffs in that litigation will see the wisdom of maybe folding their tents, and --.

But I guess in the short-term, over the next -- as we look out the next couple of quarters, unfortunately, we are going to continue to have relatively high legal expenses, or fees, probably in the range of 5 million a quarter.

At least for the next 2 quarters out.

And I can't speak beyond that right now.

So we should be modeling something along the lines of a $55 million type SG&A going forward?

That would be a fair number, yes, a little lower, maybe, probably.

Eric Tobin.

Banc of America Securities.

Another question on the $300 million facility.

If you weren't going to draw it in the fourth quarter, what would be the lever that would maybe cause you to tap it?

Is it a certain cash balance?

Thanks.

Basically as we said it's for general corporate purposes and working capital purposes.

As you know, we're in a downturn, and we don't know exactly when we would need to draw upon that.

But as we move into the first quarter, and we start to do some CapEx spending I'm sure there will be some needs for it at that point in time.

William Ingman, Goldman Sachs.

Basically from the first quarter to the second quarter when your gross margin declined, you said it was due to product mix.

Can you give a reason for the decline in this quarter and the further declines going forward in terms of gross margin?

And can you give us an idea of where the bottom would be in terms of -- for us to forecast in terms of where your gross margin would possibly go into 2005?

Well, the bottom is going to be really dependent on where the sales forecast goes, quite frankly, and obviously we're hoping for some improvement there.

As far as where the margins are going, last quarter we were -- and that is in Q2, our margins were around 19%.

This quarter, on relatively flat sales, we're at 18%, but as we indicated in our release, we do have a reserve in there of around $6 million for inventory.

So if that were factored out, you would be flat margins right there on relatively flat sales.

We're projecting, as we go into Q4, that sales would be down 5 to 10%, so obviously margins are going to come down.

The principle driver of our margins on the Amkor model are not any one particular product line, but it's overall utilization.

And as John indicated, there's a good amount of capacity right now, and as we start to use up that capacity, as we bring in the IBM business next year, that will go a long ways to improving margins.

Krishna Rangarajan, CRT Capital.

Two questions here.

One, your gross margin guidance for the fourth quarter, what kind of pricing declines and inventory reserves, if any, is embedded in that expectation?

Well, the pricing declines would be what our normal prices decline is in any one quarter, which is normally somewhere in the range of 2 to 3% per quarter.

That's factored in there.

There are no inventory reserves, per se, factored into that guidance.

All right.

Secondly, looking into 2005, your CapEx guidance is for 100 million.

You have roughly 160 million in annual interest expenses, which brings us to a total of 260.

Do you see any other major cash items beyond this, and how do you -- how are you looking at managing your cash position if you've got any targeted cash balance at the end of '05, please?

That's a loaded question you're asking a lot there.

But you're absolutely right.

I mean, the 160 in interest is a -- it's a large number to deal with, but very manageable.

We've managed it over a number of years, and we can continue to manage that, as we de-lever, we're focused on bringing that number down.

The CapEx of roughly 100 million as we indicated, so you're right, that's 260, and the rest based on where we see our cash flow from operations coming from.

And a lot of that is dependant on what the sales level is.

We're confident that we can mange the cash within the range of the sales that we're looking forward to.

Justin Lou, Zazovy Associates.

I noticed the total debt balance increased by about 45 million quarter-over-quarter.

Was that all due to the Unitive acquisition or what was that composed of?

There is some Unitive debt in there, and there's also, I believe there's like 25 in there, and there's 18 million that was on the -- acquisition of a facility we did in Taiwan, but that's in a prior period, so.

Okay.

And in terms of future payments for acquisitions, how much can -- would you project being paid out in terms of earn outs, or I guess what's a maximum amount that could be paid out in terms of earn outs.

As you know on the Unitive acquisition, there's an opportunity for an earn out of 55 -- it could be up to $55 million.

Our current projections are that it will be substantially less than the $55 million.

Steve O'Rourke, Deutsche Banc.

Two questions.

One, will the capital spending of the first half of this year directly address the areas where you lost share or do you have to do some more investing to kind of neutralize these sort of areas of disadvantage?

Yes, they are directly -- look at some of the areas we lost in market share.

That is in the FlipChip bumping area, which we hope will be in support of recapturing market share in the chip set and graphics markets.

Okay.

And in your prepared remarks, you mentioned that you expect the industry correction to reverse by mid-2005.

What gives you confidence in that expectation?

I guess confidence is the wrong word.

None of us have confidence in this business of forecasting the future.

More the nature of the inventory, really.

I think if we believe this is correction in inventory correction, and general economy stays next year the way we anticipate, then that is a correct assessment.

On the other hand, if demand -- worldwide economy doesn't hold, then the final demand issue is no longer an inventory issue, then I guess we would be wrong.

Therefore, I don't think we can say with certainty what the future is going to hold, but current -- our analysis, including you people, are saying this is generally inventory correction issue.

If that is the case, probably by first quarter next year things should come back to normal.

Especially in our case, as, John has mentioned many times, looks like an inventory issue with our customers, a wafer level issue rather than finished product issue.

Charles Galvinheimer, Satellite Asset Management.

Can you just -- I would like to ask you -- I guess you've answered a number of the questions that have been asked many didn't ways, but let me just come right at you with the convertibles.

So now that you've been able to backstop yourself with the $300 million credit facility, could you discuss what you may be willing to do to take out some of the shorter maturities now, and really I guess shore up the longer path here for the Company, whether it be debt buybacks, whether it be a tender, whether it be some sort of debt for equity exchange?

Can you discuss what your upcoming plans could be for that?

We can't use the $300 million to pay down the subordinate debt, so that's one thing.

We do have plans, albeit, though, to generate free cash flow, combined with, -- the funds could be used at maturity, but not early paydown.

So I believe with the $300 million that we raised, with the free cash flow that we plan to generate, we are going to have more than enough cash flow to pay down both the '06 converts, which is around 233 million, and the '07 converts, which I believe is around 147 million.

So '06 and '07, I feel very comfortable that once again free cash flow from operations combined with our existing cash resources, no problem.

As far as other options, we're always looking at our capital structure to see ways of optimizing on that, as we go along, and we'll continue to do that.

Maralia Aborhi, JP Morgan.

Could you break down what your unit growth expectations are for 4Q for your LeadFrame versus laminate product lines?

We haven't given guidance on unit growth by product category.

Basically, you saw our revenue guidance.

Right.

So maybe at least qualitatively say which way.

Let me just go on.

That the units will probably grow, will do better than the revenue in Q4, okay?

Because we expect price erosion.

It all depends on mix that we actually experienced.

In fact, we see the mix enriching a little bit as we go forth because of the drop-off and the very low ASP activity.

So I think basically the laminates will grow more than, will do better than the LeadFrame business in Q4.

Peter Wright, CIBC World Markets.

Just one quick follow-up.

Since the Unitive deal, have you acquired any new FlipChip customers?

Lots of interest from several new customers, and I think a strengthening of relationships with customers we already had activity with.

The Unitive acquisition was greeted very, very nicely by our customer who were in the FlipChip business.

Is it fair to assume that FlipChip is at least flat as a percentage of revenue on a 3 and 6 month look-back?

No, it's up nicely.

Sual Lunot, Sandal.

I was wondering if you could talk about pension issues, any underfunding, and any payments that are due the rest of this year into '05?

Pensions, thank goodness, are not a big issue for Amkor.

We only have some overseas; for example, in the Philippines, but it's -- it was slightly underfunded last year, but it's a relatively small amount, and, as you're aware of pensions, you can make that up from period-to-period, depending on what the performance is.

But very little exposure in terms of pension liability.

Amkor as the Parent Company does not have a pension plan per se, we do have a 401(k) Defined Contribution Plan, but no defined benefits.

Satya Chillara, RBC Capital Markets

I wanted to ask you on the royalties, on MLF package, I know you have licensed this package to some of your competitors.

So are you collecting any royalties on that?

If so, can you break it out, please?

Yes, we are collecting royalties, and we choose not to break it out.

It's pretty confidential.

Okay.

So on the gross margin front, if you -- without that, what would be the gross margin?

At this point, it's not a major impact.

Tim Arcuri, Smith Barney.

You know, maybe it's in the tax rate, but I'm struggling on the guidance for Q4.

What is the tax rate going to be in Q4, Ken?

We're not using a rate, per se, because we have these huge [anawells] and on a U.S. basis, we're basically putting cash taxes for foreign jurisdictions where we don't have benefits available.

And that's approximately $2 million is where we're at right now.

Charles Galvinheimer, Satellite Asset Management.

Just a quick follow-up on the convertibles, and your $300 million bank deal.

Does the bank deal specifically preclude you from paying these off at maturity in addition to buying them back in the open market?

Not at maturity, no.

Ladies and gentlemen, [OPERATOR INSTRUCTIONS].

Gentlemen, it appears we have no further audio questions.

Please continue with any closing remarks you may have.

Thank you very much for attending this call, and we look forward to the next call.

Thank you.