艾克爾 (AMKR) 2004 Q2 法說會逐字稿

Good afternoon, ladies and gentlemen.

And welcome to the Amkor second-quarter 2004 earnings conference. (OPERATOR INSTRUCTIONS).

As a reminder, this conference today is being recorded today, Tuesday, July 27, 2004.

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

Please go ahead, sir.

Thank you for joining us today.

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

With me today are Bruce Freyman, 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 Bruce will have some closing comments.

Second quarter results were disappointing.

Revenue was in the low range of the guidance.

And unfavorable product mix, together with the cost absorption of recent gross initiatives compromised our profitability.

Over the past several months we have embarked on a series of initiatives to position Amkor for long-term growth.

We extended the production capacity in several strategically important advanced package families.

We acquired a new factory in Taiwan and are extending our operational footprints in China.

We entered into a multiyear supply agreement with IBM, and acquired the IBM Singapore test operations.

Last week we reached an agreement to acquire Unitive and its sister company in Taiwan.

As noted in our press release, we have sufficient production capacity for the foreseeable future.

Now it is time to elaborate our new operations and execute on our business opportunities.

Our organization is forecast and fully committed to achieving success.

We cannot control the macroenvironment, but we're working hard to ensure that we achieve our operational goals and support our customers.

Ken Joyce will now review our financial results.

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 represent the current view of management.

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

The second quarter gross margin was negatively impacted for reasons we discussed in today's earnings release.

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

In addition, we expect the acquisition of Unitive will be dilutive to our earnings in Q3 of this year, quite possibly Q4 as well.

Taken together these factors will constrain gross margin and cash flow this year.

While these growth initiatives will increase our costs over the near term, we believe that these actions will yield the greatest long-term benefits for Amkor and our shareholders.

During the quarter we resolved an outstanding dispute with one of our former software providers regarding certain licensed software that the company was unable to successfully implement.

The $3.8 million settlement amount largely represents the initial license fee paid by the company to the software provider.

Capital expenditures were $124 million for the second quarter, and 295 million for the first 6 months.

We have reduced our second half CapEx budget to $80 million, and should spend most of this in the third quarter.

We're targeting to be free cash flow positive in the fourth quarter.

Payments associated with our IBM transaction and the proposed acquisitions of Unitive and Unitive Semiconductor Taiwan are business acquisitions and will not be treated as capital expenditures.

In April we sold 10.1 million shares of Anam Semiconductor for approximately $50 million.

Our investment in ASI has been reduced to 4.6 million shares or 4 percent for the company's outstanding shares.

Here's a recap of third quarter guidance contained in our earnings release.

Revenue should be flat with the second quarter.

Gross margin should be around 19 percent.

We expect third quarter net revenue -- third quarter net loss in the range of 7 to 9 cents per share.

Now I will turn the call over to Bruce Freyman for some additional comments.

As Jim said earlier, we have made significant investments in our infrastructure.

Now that we have sufficient production capacity we will be very selective in adding capacity and will focus on supporting strategic growth areas such as flip chip.

We're also committed to increasing the profitability of our business.

We're working to enrich the product mix, reduce material costs and increase margins.

We're qualifying low-cost suppliers, and have been successful in negotiating material cost reductions with existing suppliers.

We also are collaborating with our customers to reduce manufacturing costs.

Our legacy LeadFrame package business remains strong in the quarter with units up 17 percent over Q1.

This reflected underlying demand for legacy products and high utilization rates at IBM.

This also reflects additional traction in our strip test business, as our test units were up 19 percent over Q1, and much of this growth was in strip tests.

The outlook for our LeadFrame business is cloudy, however, and we see signs that the inventory correction is now migrating into LeadFrame packages.

In the near-term demand for several of our leading-edge products such as stacked packages, chip scale VGA and MicroLeadFrame will depend on the level of inventory absorption in the industry.

Stacked package unit volume was up modestly in the second quarter.

However, the mix was richer reflecting more shipments of 3 die stacks and memory over logic packages.

System-in-Package shipments were up nearly 60 percent in the quarter reflecting strong demand for power amplifiers, wireless LAN modules and memory cards.

Our flip chip business is gaining traction.

In our Taiwan factory we have commenced volume production of graphics chips.

And we are under qualification with virtually all the major graphics and chipset providers.

Looking across our portfolio advanced package, once we get past the current inventory correction, we believe the long-term demand for these products is very good.

As these products become a larger part of our business later in '04 and in '05, they should help enrich our mix.

Our test business continues to strengthen.

Our new Singapore test Center, recently acquired from IBM, provides Amkor with a complementary test revenue stream and a solid platform for growth in this strategically important test market.

We have been expanding our operational footprints in Taiwan to create a critical mass of technology and production capability to support what we believe are segmentally growth opportunities in flip chip.

In March we acquired a building near the Sinchu Science Park in Taiwan, close to the bumping facility that we are acquiring from Unitive.

With these acquisitions Amkor will be well positioned to provide turn-key wafer probe, bumping, assembly and final test for flip chip applications supporting the graphics, chipset and gaming markets.

The combined benefits of our IBM relationship, the capabilities gained from our Unitive acquisitions and our successful track record working with our customers, Sony, Toshiba and others should place Amkor in a strong position to win the assembly test business associated with the Next Generation gaming platforms in '05.

In China we have constructed a second clean room at our C1 factory, and will add production equipment as demand builds.

We entered the second half of 2004 with the following goals, improve the profitability of our core business by enriching the product mix, reducing material cost and working with our customers to reduce manufacturing costs; focus our new business efforts on areas which yield the greatest return; maximize the benefits of our collaboration with IBM; integrate Unitive and leverage our respective capabilities in flip chip to gain a major share of the gaming, graphics and chipset markets; leverage Unitive's capabilities in wafer level packaging to gain a strong foothold in this rapidly growing market; and finally to generate positive free cash flow in the fourth quarter of 2004.

Operator, we will now open this call to questions.

(OPERATOR INSTRUCTIONS).

John Pitzer with Credit Suisse First Boston.

I wonder if you can comment -- when you look at the landscape today how much of your Q3 guidance do you think is Amkor specific, i.e., could you be losing some share here versus just overall industry conditions?

And I wonder if you can just give us some qualitative comments about which end markets seemed to be the weakest to the strongest?

Thanks.

John, this is Bruce Freyman.

Your question about how much of the Q3 outlook is Amkor specific versus market share, you know all I can say is the forecast for Q3 have dropped across most of our customers -- IBM's as well as fabless.

It is been very broad-based from low pin count devices up to very sophisticated devices for telecom and other applications.

So I think that most of the customers, most of our customers and most of the semiconductor companies in the industry are starting to see some pullback in Q3.

Your question about market share, we do not believe we're losing market share in terms of customers moving business from Amkor to another OSAP company, or in terms of losing a new business opportunity for which we are qualified.

And just one other thing to add on the market share is that the supply agreement that we signed with IBM will result in a sizable shift in business from our competitors starting in the beginning of '05.

Just on qualitative information about end markets?

Thanks, John.

On end markets we saw that the wireless market slowed down for us in Q2.

We saw as result of our PBGA's doing fairly well, we think that the computer market was strong in Q2.

And then going into Q3 it really remains to be seen what happens with the back to school rush and do people start -- do people burn off their inventories and start building more parts.

So just very cloudy right now in almost all the end markets.

Jeff Harvard (ph) with Lehman Brothers.

Can you just talk a little bit about the balance sheet, the increase in short-term debt?

It looks like your net debt was up by about 170 million a quarter.

Yet, I think you had the 50 million of proceeds, and then you talked about a free cash flow deficit of 88 million.

And just to follow-on to that, can you just outline the remaining costs in the second half for the IBM transaction, Unitive, and any other cash outflow item?

This is Ken.

With respect to the increase in the short-term debt about 120 million of that comes from the (technical difficulty) with respect to the IBM acquisition.

And about 19 million of it is tied up in our facility in Japan. 19 to 20 -- let's say about 20 million there.

With respect -- I think that is pretty much what is happening on the short-term debt side.

The second part of your question, Jeff?

And just on that, that was -- that short-term debt is what exactly?

It is a notes payable with respect to the purchase of the IBM supply agreement and facility in China and the operations in Singapore.

If you recall, the purchase price is 145 million, and 20 million during the second quarter here.

The balance is due in the fourth quarter.

That makes sense.

And just in terms of your Q3 guidance, can you just talk a little bit about what you're assuming in terms of pricing versus volume, as well as operating expenses given your EPS guidance is a little lower than Q2?

The pricing we're assuming are moderate price erosion as we would normally see in any one quarter, which could be anywhere from 2 to 4 percent down, which is normal.

With respect to operating expenses, you're right there, there are some increases in operating expenses.

With respect to some of the new operations that we are bringing on that will increase the factory SG&A which comes into our corporate SG&A.

So some of the expenses are up there.

Interest expenses also up slightly.

It will be up during the quarter with respect to the note obligations we took on from IBM and Unitive on payment due them.

So interest will be up approximately $2 million during the quarter.

And how about operating expenses?

Up --?

If you're talking about operating above the line and above the gross margin line, we have said we think margins will be flat at 19 percent, so we are pretty much --.

I'm sorry, I mean SG&A R&D.

How much should that increase?

We're not going to give a specific number there, but it is going to be up slightly.

Up slightly, okay.

And just a last question.

In terms of the CapEx looking out a little further, let's say '05 is a more muted growth year, what do you think you need to spend given the IBM acquisition, Unitive and your based new products development?

We have done a significant amount of expansion during '04.

And we're slowing down our spending going into the second half of the year.

So we will probably -- we will spend on the order of about $80 million in the second half of the year.

Next year we foresee we will spend somewhere in the area of about $100 million for total CapEx.

And does that factor in normal growth in the 10 percent range for semiconductors?

We haven't talked about what our growth will be next year, but currently we're running about 73 percent capacity.

We put capacity in on those products that we're fairly confident will grow through next year.

The only type of investment that we need to make for the most part are for flip chip and the bumping business that we've acquired from Unitive.

So that is where most of our CapEx will go.

Tom Diffely with Merrill Lynch.

Can I get a little more information about the software settlements, any details there?

I don't think we are going to talk -- yes, we purchased some manufacturing software.

And as we said earlier the initial license fee was around $3.8 million.

We paid some consulting fees, had it installed, it never worked properly.

And then just one more question then quickly on the CapEx.

Where is flip chip right now as a percentage of business, and where do you think it goes at this point by the end of '05?

We don't break it out package by package or family by family, but it has enjoyed nice growth so we think it will most likely double next year.

It is really one of the biggest growth opportunities that we have have.

Amkor traditionally has not played in the graphics and chipset business.

We're getting -- we're already in volume production with Anvitia (ph) on flip chip for graphics.

We're getting qualified with ATI, another major producer of graphics chips.

And we're getting qualified with virtually all the Taiwanese chipset producers.

So our acquisition of Unitive in Taiwan, our purchase of the FICTA building earlier in the year, and our deal with IBM we feel are really putting us in the driver's seat for competing for the chipset graphics and then the next generation gaming consoles.

Pierre Maccagno with Needham & Company.

Could you comment on the inventory that increased quite a bit?

And if you could give some color as to the breakdown on die banks versus finished goods, what could you said on that?

Fair question.

Inventory was up 30 percent during the quarter.

As you're aware, we don't have any raw materials or finished goods, we just have -- we don't have finished goods or work in process, we only have raw materials.

We buy them to customers' forecast and we have customer guarantees that if they are not -- if the inventory is not used than they will pay for it.

What has happened during this quarter was our customers maintained their forecast.

We bought the inventory to support their forecast.

They just didn't draw down on those sides for production during the quarter.

So they tell us they are still going to produce these products.

They have asked us to hold this inventory.

At such time after several months under terms of agreement they will have to pay if they don't use it.

Having said that, we have taken steps to reduce inventories as we move forward.

So during the last month that we have already begun reducing or working to reduce our inventory levels.

How would you do that?

Just by ordering less materials -- ordering less materials, keeping less on hand and then having less in the pipeline.

And finally are you going to give some breakdown as to, for example, advanced versus legacy product or also the end markets communications and computers, the breakdown of revenues?

We have elected not to break down the revenues that way particularly on the end markets.

On the end markets it is very, very difficult for us to understand whether these parts are computing, consumer or communications.

There has been such a convergence in those industries in particular.

So it has been -- we've spent a lot of time and we found we're not getting much value for us or for our shareholders by doing that.

So this time around, what I've said is we think that the wireless business is taking a pause during the third quarter while people burn off some inventory.

That is what we see in general in the wireless business.

And then it is very difficult -- obviously the storage business went down quite a bit in Q2 from -- you can see that from the announcements of our customers that support that industry, particularly the hard disk drive part of the business.

And so we will have to see if that comes back during Q3.

Eric Gomberg with Thomas Weisel Partners.

I was hoping you could talk about any changes you might be doing in terms of how you do forecasting?

You go back to when you reported the fourth quarter and you gave a big CapEx number for '04 and said that you had very good visibility into advanced package demand for the second half of the year.

And you reported Q1 and gave Q2 guidance, which also ultimately you had to take lower.

It seems that there has been a disconnect between what your expectations have been from your customers and what reality has been.

So I am just wondering if you're rethinking forecasting tools?

It is important to note that for the last several years our forecasting system of working with our customers to understand what their demand is by line item has been very successful for us.

Obviously in the first half of this year it has been very unsuccessful for us and caused us to over judge or be too optimistic on what the first half demand would actually be.

You know we have very, very tight relationships with virtually all of our customers.

We constantly validate, revalidate these numbers.

And what has happened over the last 6 months is that the forecast from our customers have become unreliable.

So the die support versus what the customers are forecasting to load has come in obviously below 100 percent.

What we have done is we have gotten even closer to the customers.

We have incentivized our sales team in terms of getting the die support number right, the forecast number right.

But to make a long story short, is we are working closer with our customers to understand what true end demand is.

And can I also just ask you, you said you don't think you're losing share.

And I am just wondering you grew 1 percent sequentially in March and 6 percent in June, and you're talking about your customers working down some excess inventory.

I'm just wondering how did they get this excess inventory because 1 percent and 6 percent growth isn't something that would drive a balloon in inventory unless they are building a lot more from other suppliers?

I don't know what to say to you there.

You know we have talked to the customers in terms of the standard products and advanced products.

We're winning most of the opportunities that come our way.

But I'm not sure exactly how to answer your question.

Satya Chillara with RBC Capital Markets.

Bruce, I was hoping what you could comment on unit growth in Q3 and Q4, particularly Q3?

And in terms of pricing, what was the ASP per pin for Q2?

So if you can answer those questions?

Let me talk about the ASP per pin.

So first of all pricing has been in the normal environment.

It has been a normal pricing environment.

In fact during Q2 we probably raised more prices than we lowered.

But increasingly the discussion of ASP per pin or ASP per unit is becoming meaningless in our business.

The reason for that is an increasing part of our revenues is coming from things like way wafer bumping, flip chip, camera modules, DLP, System-in-Package and test, all items or all product families that are very difficult or impossible to really talk about in the context of ASP per pin.

So having said that, like I said we raised more prices than we lowered in Q2.

It was what I would characterize as a stable pricing environment.

And I expect in Q3 we will see a similar situation.

What about unit growth in Q3?

We're not going to -- we will have to wait and see how that comes in.

Certainly the growth in legacy product in Q2 was big.

We put the number out that we grew quite in the mid teens in terms of unit growth.

And we will just have to see how Q3 comes out.

It is really the whole miss for us, or a huge part of the miss in Q2, was mix related.

Somewhat sales related, but almost entirely mix related.

So we are -- our high margin products, the product that -- such as ChipArray and Stacked CSP and MLF and others we had forecasted them to come in at the same healthy level they had come in or a little bit higher than they had come in Q1.

And these products just didn't sell.

So we will have to see what happens in Q3 and beyond.

Ali Irani with CIBC World Markets.

A couple of questions for you.

First one, if you give us some color on pricing.

Obviously at 73 percent utilization at Amkor one would suspect that pricing would be a significant factor in your sequential revenue guidance next quarter.

Secondly, and again not to be up too much on this, but coming back to Eric's question, I am wondering if over the course of the year all the revised guidance you have given at some point aren't market forces but again are Amkor specific in terms of market share.

And with the Company no longer giving us end market breakdown, I am hoping you can give us some color for confidence that this isn't again an Amkor specific market share loss over the last couple of weeks?

I think there's been enough – there have been enough announcements from our customers now to talk about that -- to show that there's some type of inventory correction going on in the market.

So I don't need to go through our customers one by one, but I'm sure you guys are aware of them as well as I am.

So there is definitely an inventory correction being seen out there.

There is an inventory correction in the wireless market, in the standard linear and logic business, in the analog business.

So we're pretty confident that that is what we see.

And then as far as market share, I stand by my original comment that for business that we have qualified on, for business that -- with our existing customers we have not lost market share and we have not lost the opportunities.

That's all I can say on that.

One would suspect that when we get to the fourth quarter in the next year, excluding the IBM business, your core business will growing at least at the rate of market demand, is that what you're saying?

Historically our business has outpaced the growth of the semiconductor industry.

And we would certainly expect that -- I can't specifically say what is going to happen next year, but we would expect that over some medium period of time that we will outpace the growth of the semiconductor industry as we have for most of our 35 years in existence.

Timothy Arcuri with Smith Barney.

I actually had two things.

First of all, Bruce, can you comment a little bit on this low-end mix?

How long is this is going to continue to impact the P&L?

Or was this kind of onetime opportunistic activities that you kind went into the market on this quarter, or is this something that can have some more longer-term impacts associated with it?

Good question, Tim.

Our mix, or the resultant P&L that we had and the lower than expected sales, although we met the range of 5 to 8 on our sales, really wasn't driven by the fact that we've got a lot of low-end products.

When we say low-end we just mean low lead count, LeadFrame product.

Actually a lot of these are very, very good business for us.

So what I said before, and I think what we have outlined in our release is, is that it was really the high-end, the lack of the high-end that really hurt us.

So if you look at our numbers -- we look at our numbers internally from Q3, Q4 and going into Q1, all this stuff like MLS, Stacked CSP, ChipArray and some other products were selling very, very well.

We had very robust forecast for these packages from our customers.

And all I can say is that maybe because of the inventory glut out there these packages didn't sell for us in Q2 like we had anticipated.

And almost that alone was really the major, major reason for the P&L mess in Q2.

Can you maybe looking into '05, can you give us some idea, if not quantitatively at least qualitatively, can you give us some idea of how much revenue the IBM and the Unitive deals might add?

We haven't said by year.

But I believe we have said that the value of the IBM deal could be up to a $1.5 billion by 2010.

Shekhar Pramanick of Schwab SoundView.

Two-part question, Bruce.

Maybe you can talk about when did you see the weakness show up in terms of your customers coming in with actual package wafers?

And how you are seeing the first three or four weeks of July what are the unit numbers, how are they tracking.

And then I have a question, one more question.

Yes, when we saw the weakness was probably starting about the second week of June.

And forecasts started coming down -- our six-month forecasts from our customers.

So by now it is pretty widespread, as I mentioned, pretty widespread and pretty broad based.

And then July we can't comment on what is happening in the third quarter at this point other than to say that it is very cloudy out there.

People are in the midst of reevaluating what their inventories are and how much supply is in the channels.

And so right now we're sticking with our guidance of flat to --.

Do you feel it has bottomed or -- how much has it come down?

Has it come down 10 percent on a unit basis or less or more?

We're not reporting that on this call.

It is -- like I say it has been pretty broad based.

And we have talk to the Presidents and COOs of many of our customers, and most of our customers are seeing it across a pretty broad range of products and markets.

Stanford Nishikawa (ph) with Citigroup.

A couple of clarifications.

In regards to the cash flows in and out for IBM and Unitive, you've got the 125 in the fourth quarter, but when does the Unitive payment hit?

And when did or does the IBM payment hit?

The Unitive payments will hit in -- 33 million will hit here in Q3.

And then there is another payment that will be payable out one year from now, and that is approximately another $18 million.

Then there are earn outs associated with those transactions, and we will have see how they turn out.

And then I am sorry on the $20 million for the first IBM payment you said that hit in the second quarter or the third quarter?

For IBM?

Yes, for IBM?

That was in the second quarter.

The other thing, I think in response to Jeff's question did you mentioned that your CapEx could potentially grow (indiscernible) to the $100 million, is that what you said?

Yes, I think that's what Bruce mentioned.

I think that is a fair number.

We are revising our numbers down somewhat.

We believe that with the capacity that we're running at right now we have plenty of adequate capacity.

And in terms of what we're seeing here on the revisions downward by some of our customers that we believe we have plenty of capacity and we won't have to spend a lot of money other than funding some of our operations for Unitive and in the flip chip areas with IBM.

One of the real benefits of the IBM deal is how much overlap there is between IBM's outsourcing requirements and the capacity that we already have in-house.

It won't require a lot of capital to satisfy IBM's requirements for next year.

Sundar Varadarajan with Merrill Lynch.

Just another question.

You (indiscernible) you have about $183 million of payments for IBM and the Unitive deal, plus about another $80 million of CapEx, bringing about total cash outflows just for CapEx and acquisition to about $263 million for the second half.

Given where you are from a liquidity standpoint, from my calculations it seems like you would be running to close to about $100 million or so of cash balance by the end of the year.

Are you guys comfortable with that kind of number, or do you think you may have to do some form of financing to provide a little bit more of a liquidity cushion as enter '05?

We feel very good with where we are at from a liquidity position.

As we indicated, we have revised our CapEx outlook for the balance of the year to about $80 million there.

I think even with the payments that you talked about on both IBM and Unitive and that CapEx, we should still have cash in our mind of around 200 million.

That is our target.

That will keep that balance and we're comfortable, and we are very confident that we can meet that target.

Eric Ruble (ph) with Miller, Tabak, Roberts.

Ken, could you give us an update on how depreciation you expect that to ramp over the next two quarters?

Eric, that looks like depreciation is probably in the range of around -- we are at 58 million in Q2.

I think it will be around 61 in Q3.

And as we look out at Q4, we should be probably not much different than that -- hold on.

Right around the same range in Q4, because we are cutting back on the CapEx.

So it will be right in the straight same range in Q4.

A question for John.

Can you comment on the global footprint?

You've got operations in Japan, Korea, building operation in China and focusing more on operations in Taiwan.

Can you rank for me what the order, how they are going to drive the business here going forward?

And particularly if you can comment on how Japan is doing?

How they are doing -- I can just talk out -- we don't break out the P&L's for the different countries.

But China has been a tremendous growth engine for us over the last couple of years, albeit from a green field site.

This year it will grow about 3 or 4 times what the revenues that we did there last year.

So we are facilitizing our second building there.

And the first building, our first 75,000 square feet is full, so a lot of customers are interested in going to China with us.

Taiwan is growing.

It is the whole reason that we bought the FICTA building, the FICTA building.

And Sinchu Science Park was to accommodate the growth that we see there in our flip chip business and also laminate business that is starting to pick up there, that we just trying to develop.

Our DLP business with Texas Instruments is also in Taiwan.

That has done very well for us over time.

And then finally, Japan.

Japan, we have excellent relationships with Toshiba as result of our joint venture with them that we bought out at the end of last year.

Our business with them is growing over time and it has enabled us to get in the door of a number of Japanese semiconductor companies that we never had as customers before.

So I would say Japan is doing well.

Both -- it is doing well this year and we think it will provide growth for us with the Japanese semiconductor manufacturers over time.

Jesse Pichel with Piper Jaffray.

A couple of questions on inventory.

First of all, your inventory is mostly substrate and components used in SAP, right?

Substrates and LeadFrames are probably the two biggest components of inventory.

But we're not talking about chips, right?

No, not chips.

Not ships and not finished goods.

Can you comment -- in your customers die banks what does the die bank situation look like at your leading-edge plant in Korea?

And then for some of the LeadFrame business in the Philippines, are you seeing an increase there?

Yes.

So in the Philippines it has been strong for at least 3 quarters because it is kind of mirroring what has been happening in our LeadFrame business.

So that continues to be strong in terms of die bank.

And then in Korea, our dye banks had been going down in Q4 and Q1.

And now die banks have risen in Q2, which kind of suggests that people have built more wafers than they need to assemble, and so they're stopping the whole supply chain process in our factories.

So the they've got the die there to release, but they're not releasing them because they don't have the end demand, is our thinking.

So having said that, it is important to remember that when we talk about die banks it is only about 25 percent of our customers die bank with us.

Jay Dana with J.P. Morgan.

This is actually Maralia Wolf (ph) for Jay Dana.

A couple of questions.

On the last conference call you had said that about 65 percent of your revenues came from IBM.

How did that change in the June quarter, and how do you see that going into the September quarter?

I think it's about the same.

I think it is about same, and if anything maybe during the quarter the IBM share of the revenue rose.

But I would say it is about the same as last quarter.

A couple of quick housekeeping questions.

You used to break down the assembly revenues between advanced package and traditional packages, did you guys give that out, or did I miss that?

No, we have stopped doing that just because of the difficulty in understanding which ones were advanced and which ones were not advanced.

And final question, cost for materials as a percentage of sales, do you have a number for that?

It is been running, depending on the factory and the product line, it is running in the 38 to 40 percent range.

Michael Gordon with Orix.

My questions have been answered.

Thanks.

Krishna Rangarajan (ph) with CRT Capital.

My first question is on your CapEx guidance for '05.

I believe around SEMICON West you were publicly stating a $250 million number?

That's correct.

And you brought it down substantially.

What has changed between then and now.

Once again as a result of the decrease in the level of business, we did put a lot of capacity in place in Q4 of last year, Q1 into this year.

So we have plenty of capacity in place.

So you had to reevaluate our plans as conditions changed.

And we believe as result of that we have more than adequate capacity.

As we indicated a little earlier on the call, we would have to spend some money to support our new acquisitions, the IBM business, and the Unitive flip chip business, but other than that, we really don't see a lot of need for CapEx next year so we revised that downward.

So with 100 million in '05 CapEx what kind of growth could you accommodate in '05?

So right now we're running at 73 percent, and so you can just figure the math on what the revenues could be.

Be in the a 30 percent range then, right?

Right.

And then also we anticipate that we will have the -- we know you'll have the business from IBM going into next year and then additional business from investments that we will make in flip chip in Taiwan.

Todd Jaric (ph) Sanfield Capital (ph).

Just curious on a going forward basis, let's say hypothetically this is your ongoing run rate, around 360 million of cash flow, and CapEx next year of 100 million, interest of 140.

What do you see using that excess cash flow generation?

Would you potentially revisit your debt reduction plans from earlier?

You have bought back bonds in the past, just curious?

Absolutely.

Deleveraging is certainly something we're going to look at, and we would differently be accumulating some cash for that purpose.

Is it safe to say that you are pretty happy right now at your current footprint with three major acquisitions you've made and the CapEx.

Is there any other areas were you might want to expand into?

No, we're done with acquisitions for the foreseeable future.

And as our Chairman, Jim Kim mentioned, we need to execute on integrating those activities and growing into them.

Jon Rogers with Morgan Stanley.

Just in the context of potential M&A, and you said that you don't have any interest in doing that.

But just in the context of your competitive positioning for the upturn, how are your strategies in China and Taiwan played out?

Have you sought to be closer to your customers?

Have you been aggressive in qualification or do you think you have better technology and processes than your competitors?

And on that front are there any other technologies that you may be lacking that may ultimately lead further M&A at some point down the road, maybe 2, 3 quarters?

No.

Absolutely in a 2, 3 quarter timeframe we're not thinking about any more M&A.

But in terms of how we have done in Taiwan and China, in China our business has really done well.

We've gone in there with the multinationals, the large IBMs, the Intels, Phillips, TIs of the world and so we are very happy with the wave we have grown our business in there.

One of the attractions for doing a deal with IBM was that we foresee the need to have a large operational footprint in China to take care of our customers and grow our business in the long-term.

In terms of competitive advantages in Taiwan, our acquisition of Unitive gives us the absolute best electroplated bump technology in the world.

It plugs the one hole that we had in our flip chip supply chain that customers were concerned about, which was the electroplated bumping ability on 12 inch wafers.

So now that we've got that, we really are positioned -- we are in a very, very strong position to win in the chipset graphics.

And then maybe even more exciting is the upcoming gaming business like PlayStation 3 and Xbox 2.

Have you guys started qualifying for PS 3 and the next Xbox?

I can't comment.

Avy Dennis (ph) with J.P. Morgan.

Given that your liquidity is going to end this year somewhere between 100 and $200 million, you have no bank debt -- no real bank debt in your capital structure, have you thought about going back to the banks, giving them back that CapEx covenant now that guidance is about 100 million for '05, kind of shoring up liquidity so we don't have to worry about the upcoming maturities, '06, '07?

We certainly have been ability.

You're right.

We just went through the process of paying down our term loan, and we do have a revolver that we renegotiated just recently with our banks.

For $30 million?

For $30 million.

We really don't see the need to put any bank financing in place, but it is an alternative, but it is not something we're looking at right now.

Philip Joel (ph) with Goldman Sachs.

I was wondering you guys (indiscernible) for capacity utilization allowance 73 percent.

We've heard from a number of foundries that their capacity utilization is the mid-90s.

I was wondering how you guys bought about that and how you intend to reconcile it?

You're talking about wafer fabs foundries?

Yes.

So the foundries are close to being full, although there is another sign that people are decreasing their orders particularly at the second-tier wafer fab.

I think there's been quite a bit written about that the orders for some of the major fabs are starting to soften.

So yes, those guys have been running 100 percent most of the year -- and kind of comparing apples and oranges between the assembly and test business and then the found business.

Eric Tolden (ph) with Banc of America Securities.

Thanks, they have been answered.

Mark Barkman (ph) with Pacific Crest Securities.

On your 73 percent capacity utilization does that include Unitive or not?

No, it doesn't.

What will it be at the end of Q3 then with Unitive -- let's say that your status quo right now, what will it be at the end of Q3 with Unitive included in there?

About the same.

I don't know.

I don't know.

I have not calculated that.

Bruce, I saw your presentation at SEMICON West just a couple of weeks ago, and you seemed pretty bullish at the time.

And then you made some comments that you actually started seeing this slowdown in mid-June.

But I am just not getting the same message today.

What has changed in the last two weeks to --?

So that the entire presentation, right, was on what is happening in the OSAP (ph) sector?

So why are people outsourcing more?

So it is not necessarily what did they do in June for outsourcing more, but what we talked about is that packages are getting more complex.

It is more and more difficult for IBMs to make the investment in a whole different host of packages.

Things like DLP, memory cards, flip chips are all examples of that.

But it is not an indication of what is happening in the here and now it is more of what kind of trends are happening like in the EMS industry or in the OSAP sector.

William Matthews with Canyon Capital.

Just kind of a general philosophical question.

If in the past we have heard the Company priority is delevering, that there's no acquisitions on the forefront, and then for probably a second time in a few years the Company has kind of missed an opportunity to issue equity over 20 percent, potentially over $20 a share potentially to delever.

How are we supposed to have faith going forward that there are no further acquisitions and that the Company's intention is in fact to delever?

We're very serious about delevering and we continue to be serious.

But we also have to think for the long-term best interests of the Company and its shareholders.

And we saw these opportunities -- the IBM is an excellent example -- where it is going to provide somewhere between 1.5 and $2 billion over the next 6 years.

That was an opportunity we couldn't pass up.

The Unitive acquisition is almost a perfect complement for the business that we have.

And as Bruce indicated, it fills a hole that we have in our organization.

So once again if an opportunity -- short-term, believe me, we have had a lot of heated discussions in-house.

Short-term there's a negative consequences -- margin compression, cash flow.

But certainly doable, and certainly we believe in our judgment after we put our team together it is in the best interests of the Company.

I can really tell you that we are going to have to ask you to give us the benefit of the doubt on this one in terms of we're acting in the best interest of the Company and its shareholders.

Between, say, December of '03 and when these acquisitions took place, when did they -- when did the opportunities actually present themselves?

Was this stuff you had been looking at for years and then the opportunity came up or was it --?

No, actually, they developed quite quickly.

The opportunities presented themselves, as I indicated, and we had to make some quick choices and some hard choices.

They weren't easy decisions to make.

And we knew the commitments.

We know what we have been saying to the markets.

And not just the markets but to ourselves.

It was clear to our path that we had to make a change in strategic direction to do what we felt was in the best interest of the Company long-term, realizing some short-term pain.

So we believe we made the right decision and we believe that the results of the Company over the next couple of years will bear that out.

And we will go back to trying to build our cash and delevering the balance sheet.

And comfortable right now.

I have over 205 -- 295 million in cash and we believe adequate capacity.

We cut back on the CapEx.

We cut some of our other expenses.

We will maintain that cash.

Our target is to keep cash above 200 million.

And we believe that is a doable objective.

And once we get pass that, we start generating some free cash flow next year that will put us in position to take down some of that debt.

For example, there is 233 million in converts coming in '06, as you're aware.

We will be in a position to pay that.

John Pitzer with Credit Suisse First Boston.

Bruce, just a quick follow-up.

I guess when you look at a subset of some of your customers, on an absolute dollar basis inventories are actually somewhere between 10 to 15 percent above where they were at the peak of the last cycle.

And when the inventory build in the last cycle kind of dissipated it led to 4 consecutive quarters of down revenue for you guys.

Given that, I guess, help me understand your confidence level that this is just a one quarter inventory issue?

You know, John, it is hard to say what is going to happen in fourth quarter and beyond.

All I can tell you right now is that there have been an inventory correction.

And some people, including some of the analysts on this call, have written that it is summer seasonality.

Some of our largest customers have talked in those terms.

But it is too close to call.

So we will have to -- we don't have the crystal ball on it.

All we can tell you in our guidance for this quarter at flat, which certainly doesn't please us, was based on a broad-based softening.

So whether we are at the bottom or not we will just have to wait and see.

And then, Bruce, this might be splitting hairs.

I am just kind of curious when you talk to your customers was it really end market demand falling below forecast, or did we just get to a point where supply growth at the fabs began to accelerate and so supply caught up to end demand which ended the inventory build?

We don't have a perfect vision of that, but it is our thinking that the wafer fabs got ahead -- people build more die then were required.

Every thing that we read says that end demand is still fairly robust.

Ladies and gentlemen, that concludes today's question-and-answer session.

I would like to turn the conference back to management for any concluding comments.

Please go ahead.

We would like to thank you for attending the call.

Ladies and gentlemen, that concludes the Amkor second-quarter 2004 earnings conference.

If you'd like to listen to a replay of today's conference, you may dial 303-590-3000 with pass code 11000487 pound.

Thank you again for your participation in today's conference.

And you may now disconnect.