台積電 ADR (TSM) 2005 Q2 法說會逐字稿

Welcome to the TSMC's quarter two 2005 webcast conference call.

Today's event is chaired by Ms. Lora Ho, Chief Financial Officer and Vice President, and Dr. Rick Tsai, Chief Executive Officer and President.

This conference call is being webcast live via the TSMC website at www.TSMC.com and only an audio mode.

Your dial-in lines are also in listen-only mode.

At the conclusion of the management presentation we will be opening the floor for questions.

At that time further instructions will be provided as to the procedure to follow if you would like to ask a question.

Please be advised for those participants who do not yet have a copy of the press release, you may download it from TSMC's website at www.TSMC.com.

Please also download the summary slides in relation to today's quarterly review presentation.

Once again, the URL is www.TSMC.com.

I would now like to turn the conference over to Dr. Elizabeth Sun, TSMC's head of Investor Relations for the cautionary statement before the main presentation by Ms. Ho and Dr. Tsai.

Good morning and good evening to all participants.

This is Elizabeth Sun, head of Investor Relations for TSMC.

Before we begin I would like to state that management's comments about TSMC's current expectations made during this conference call are forward-looking statements subject to significant risks and uncertainties and that actual results may differ materially from those contained in the forward-looking statements.

The information as to those factors that could cause actual results to differ materially from TSMC's forward-looking statements may be found in TSMC's annual report on Form 20f filed with the United States Securities and Exchange Commission on May 16, 2005 and such other documents that TSMC may file with or submit to the SEC from time to time.

Except as required by law, we undertake no obligation to update any forward-looking statements whether as a result of new information, future events or otherwise.

Now I would like to turn the conference call over to Ms. Lora Ho, our Chief Financial Officer and Vice President.

Good morning and good evening; welcome to TSMC's second-quarter conference call.

Due primarily to the net recovery from our customers, our second-quarter business has improved better than we guided previously.

We expect the recovery to continue for the remainder of this year.

I will now take you through our overall results of the quarter.

We have prepared some slides for you to look at as we look into the call.

These slides are available on TSMC's website.

During the call I will refer to the slides by their individual numbers.

To start with let's look at some highlights of our second-quarter 2005 operating results.

These figures are shown on slide number 4.

Our Q2 revenue was NT$ (indiscernible) billion, 5% higher than that of quarter one.

Net income after tax totaled 18 billion.

Meanwhile earnings per share reached NT$0.74.

TSMC ended the second quarter with 128 billion in cash and short-term investments; shareholder's equity amounted to 384 billion in key dollars.

During the second quarter we shipped a total of 1.27 million (indiscernible) equivalent wafers which is 16% higher than the 1.11 million pieces achieved in first quarter '05.

Our overall utilization rate, which includes TSMC fabs, wafer (indiscernible) SSNC and TSMC Shanghai, went up to 85% for the second quarter. (indiscernible) equity for the quarter was 18.4%, about 2% higher than the 16.5 achieved in quarter one.

Let's now go to slide number 5.

Compared to the first quarter revenue increased by about 5%.

Due to demand recovery and industry (ph) depletion, the overall utilization rate improved to 85% this quarter from 78% previously.

The positive effect from higher utilization was partially offset by a lower ASP.

Overall our gross margin increased slightly to reach 39.7% as opposed to 38.9% achieved previously.

Operating expenses totaled NT$0.5 million, slightly lower than that of last quarter representing 9.3% of total revenue.

Nonoperating income for the quarter was NT$0.2 billion higher than that of Q1 mainly due to the reversal of (indiscernible) losses of short investments and marketable securities and foreign exchange gain from hedging activities partially offset by a small scheduled payment of the lawsuit settlement (indiscernible) from SMIC.

With higher losses from TSMC Shanghai and lower profitability level from (indiscernible) partially offset by the improved WaferTech performance our long-term investment losses came to about NT$0.6, NT$0.4 billion higher than that of Q1.

Net income before taxes reached NT$70.7 billion; income after-tax came to 18.4 billion while EPS improved 9% to $0.70 per share.

You can see from slide number 6 that on a year-over-year basis our revenue decreased by 10% due to lower ASP and unfavorable exchange rate.

Gross margin was also 4% down year-on-year as the utilization rate dropped from 106% to 85% with unfavorable exchange rate adding to the negative impact.

Despite higher spending over 90 nm development, there was significant reduction in operating costs (indiscernible) production.

As a result our operating expenses went down about NT$0.6 billion and amounted to 9.3% of net sales, the same as quarter two last year.

On the nonoperating expenses item went up around NT$ (indiscernible) mostly due to foreign exchange gain from hedging activities, the reversal of (indiscernible) losses of short-term investments and the receipt of SMIC settlement. (indiscernible) of NT$1.3 billion in Q2 '04, the total investment items were a loss of NT$0.6 billion this quarter mainly due to a lower level of WaferTech profitability and higher losses in TSMC Shanghai.

Pretax income was 17.7 billion which is 23% lower than the 23.1 billion achieved in Q2 last year.

However, with a higher effective tax rate overall tax expenses for the two periods remain at a similar level due to increased capital expenditure tax credit available was higher than for this quarter.

Our profit after-tax was down 22% than that of Q2 '04, EPS also reduced by 21% from $0.95 in Q2 '04 to $0.70 per share in this quarter.

Slide number 7 will give you a view over a few important items from TSMC's balance sheet and some key financial ratios.

First of all, our cash and short-term investments for the quarter was NT$128 billion, 16 billion higher than the previous quarter position of 112 billion.

Accounts receivable balance also when up from NT$22 billion to NT$30 billion reflecting higher sales in June.

Accounts receivable turnover was 43 days for the quarter.

Current liability at the end of quarter was NT$87 billion, NT$49 billion higher than the 38 billion at the end of Q1.

This was mainly due to the accrual made for the 2004 cash dividend and employee bonuses.

Inventory wise, our Q2 inventory came to NT$14 billion, 5% higher than that of Q1.

Partly in preparation of the higher demand in the first quarter and partly due to higher value of (indiscernible) work in process tells us our advanced technology production continued to increase.

Current ratio for the quarter was 2.1 times;

TSMC continues to maintain a very solid balance sheet.

Slide number 8 summarizes TSMC's cash flow for the quarter.

During the second quarter TSMC generated NT$28 billion cash inflow mainly from net profit of 18 billion and depreciation amortization of 18 billion.

Net cash used in investment activities totaled NT$10 billion, (indiscernible) capital expenditure of 13 billion, partially offset by a reduction in short-term investments of 2 billion. (indiscernible) financing activities during this quarter came mainly from a high level of guaranteed depositing from customers.

Overall speaking, our cash balance went up by NT$18 billion for the quarter, cash and short-term investments together was also up by 16 billion.

You can see from slide number 9 that our capital expenditure for the quarter was a US$458 million, mainly from $450 million for TSMC's own fab and US$42 million for TSMC Shanghai.

About 80% of the total capital expenditure this quarter was spending to build up capacity for 12 inch fabs.

For the year 2005 total capital expenditure will be close to the high end of our guided range which was US$2.5 to US$2.7 million.

Now let us look at some business analysis.

Slide number 10 provides a quick view overall revenue breakdown by technology with marking gradually recovery in the second quarter, revenue for the most -- technology went up more so in mature technologies.

You can see from the chart that a percentage of advanced technology took up this quarter was 43%, 2% lower than the previous quarter.

However, if we compare that with the previous quarter the growth was still far superior.

Slide number 11 provides a view by application.

We saw significant Q over Q growth in the computer and communications sectors.

This was in turn reflected in the overall weighted shift.

With demand recovery and seasonal expectations for the second half of '05, revenue for computer sector, including PC graphics, chipsets and LCD monitors, went up across the board and contributed to 38% of total revenue.

Rather than the 41% achieved in Q1, communications revenue went down and accounted for 37% of total revenue.

On the wireless there was growth and most segments except for mobile phones due to seasonality and inventory digestion (ph).

Revenues from all wireline segments also improved significantly with heightened demand.

The same outward trend was visible for the consumer sector, pushing the overall (indiscernible) to 18% of total revenue.

Growth in set-top boxes and HDTV were the most prominent.

Slide number 12 shows the geographical breakdown by our revenue.

Revenue for North America, Asia-Pacific, Europe and Japan now account for 77%, 9%, 8% and 6% of our total revenue for the second quarter respectively.

Compared to the last quarter we saw greater improvement in Asia and Europe.

Side number 13 provides a revenue breakdown by customer ties.

In the second quarter revenue generated from fabless and system (ph) customers accounted for 74% of total sales.

On the other hand, IDM went down to 26% due to less outsourcing this quarter.

Now let's move to slide number 14.

This slide shows our fab utilization and ASP trends. (indiscernible) TSMC's utilization for the second quarter went up to 85% due to market recovery.

ASP, however, dropped 5.4% as a result of higher percentage of a mature product shipped and the continued price pressure.

Let's turn to slide number 15; this slide summarizes our installed capacity by fab.

Our total capacity during the second quarter was 1.45 million 8" equivalent wafers.

We also estimate our fab 12 and fab 14 capacity continued to increase which would bring our overall capacity for the quarter to 1.55 million 8" equivalent wafers.

Our capacity for the year 2005 is expected to be around 5.96 million pieces of 8" equivalent wafers which is 24% higher than the last year.

Lastly, slide number 16 and 17 provide a recap of the major events during the second quarter -- I will leave it for your own reference.

Based on the current business outlook the management expectations for the third quarter 2005 performance are as follows -- wafer shipments to increase by a mid teen single digit percentage point sequentially; overall utilization rates to exceed 90%; gross profit margin to be in a range of 41% to 43%;

ASP to decline by a low to mid single digit percentage point sequentially.

This ends my presentation today.

Thank you very much.

Operator, please open the floor for questions.

(OPERATOR INSTRUCTIONS).

Michael McConnell, Pacific Crest Securities.

A couple questions.

Can you talk about the pricing trends, what you saw in Q2 that led to the declines and why you are expecting a decline again in Q3 despite the fact that utilizations seem to be moving upward?

There are two factors that have to do with Q3 ASP.

We are seeing improving technology (indiscernible) 90 nm.

However, because the transition between 0.13 to 90 nm our overall revenue contribution from advanced technology will be in the same level of the present quarter.

Plus we are seeing more shipments in the third-quarter (ph) on mature technology and plus some pricing pressure.

So our third-quarter ASP will drop to low to mid single digit percentage point.

Is there a particular product category that seems to be weaker in Q3 than maybe leading to some of the less robust growth at 0.11 micron 90 nm?

No, we do not see a particular product, no.

Okay.

And then looking at Q4 -- although your commentary probably has to be -- it's a little bit too early to tell, but are you expecting to be fully utilized in Q4 just looking at your forecast right now or is it too early to tell?

We're not ready to guide to Q4 utilization.

However, we are seeing a recovery for the second half of this year.

Okay.

And just one last point and I'll go away.

Just with respect to capacity, what was the number you're expecting for the year?

Did you comment on that?

I just said total year capacity would be 5.96 million pieces for the full year.

Thank you very much.

You're welcome.

Timothy Arcuri, Smith Barney.

Actually I have two things.

Number one, I asked you last call what your 2006 thinking was for CapEx.

And I know that you don't want to pinpoint that right now, but at the time you said that your philosophy is that you don't want to lift CapEx around; you don't want to spend up too much or really down too much.

So is your general philosophy that you want to keep CapEx about -- in a range of up 10% say to down 10%?

That you'll want to spend more linearly?

We have not finalized -- actually it's an ongoing issue, especially when we move to the second half of this year we will continue discussing and trying to figure out a 2006 (indiscernible).

At this moment we have not finalized our number so we cannot share with you all of our 2006 CapEx.

What you said is probably true that because the CapEx has to do with capacity for the following years, so we don't want to over invest on our (indiscernible) for the same reason we don't want to under invest in a downturn.

When we look at CapEx we are looking at a longer range of material, we're not just looking at the next year.

But it's too early to tell about 2006 CapEx at this moment.

We will let you know probably January next year when we do our fourth-quarter release announcement, we will let you know by then.

Of course, okay.

And then I guess kind of a general question, in terms of when you think about adding new capacity historically if you kind of look at when you've gone out and really bought new tools to add capacity it's been right around the 90% utilization range and that's about where you are today.

So you've not been very aggressive out there in terms of placing orders.

And is there any reason why that historic relationship of 90%'ish range for you to go out and add capacity wouldn't hold true any more?

Well, I'm not sure that -- well, basically the way we do it now, we do it -- right now for instance we have a pretty good idea as to what our third-quarter and fourth-quarter capacity needs are and we know that we have enough capacity to support the demand.

And we are also looking at next year's demand picture from both a macro picture and the market segment application point of view plus of course our customers' input.

We will come up with a demand profile for next year and then we will make up our investment plan accordingly.

Sure.

But I guess what I'm wondering is that most of the capacity you're adding right now is simply you're taking tools that you've already ordered and I'm wondering -- historically you've gone out and placed new orders right around that 90% utilization level and I'm wondering if there's any reason why that philosophy would change?

I don't think we have that philosophy.

Basically what we're doing right now is we are planning on our investment reasonably well back in late last year and early this year.

So right now our capacity is about right for our demand.

And this is quite important -- very important actually for our profitability, our gross margin.

Indeed.

Okay, thanks.

Matt Gable, Calypso Capital.

I had another capacity related question.

With what seems to be migration to more of the volumes going to your mature technology, how does that possibly affect your CapEx decisions in 2006, just regarding the types of equipment you buy?

Would you ever have to be beef up more equipment and not buy as much leading-edge equipment; is that a possibility?

If so, what does that do to your margin?

Okay.

We are not adding more capacity to (indiscernible) technology.

We will add capacity for advanced technology.

So for the (indiscernible) what we're trying to do is trying to utilize as much as possible the current equipment.

However, for the advanced technology we do watch the (indiscernible) seriously, and based on that we view our capacity.

And (indiscernible) we will actually use as much as possible to achieve the highest level of utilization.

That's the philosophy around this (indiscernible).

Okay, thank you.

Robert Maire of Needham & Company.

The ASP decline you saw in the quarter and the ASP decline you're looking at going forward, could you give us some sense as to how much of that was competitive and how much of that was mix related?

You have indicated that mature product demand was strong.

Is that one of the reasons for increased competition, or could you give us some more detail on that, please?

I think generally speaking, we do see ASP growth quarter over quarter, even in spite of optimal (ph) downturn.

In the first quarter, we guided low to mid single-digit decline.

Actually, it was a combination of both.

However, we have seen the pricing pressure has been relieved a little bit, so the rate of erosion would come down in third quarter.

Okay, so what you're saying by that is the competitive pricing is reducing a little bit, and now it is more of a mix issue that is responsible for the ASP?

If you would look at our track record, TSMC regularly maintains stabler ASP trends compared to our competitors.

I mean, as the partnership to our customer, we don't easily (indiscernible) the price and we don't easily reduce the price.

So that would (indiscernible) page 14 of our presentation, the ASP actually maintained quite stable.

Second question somewhat related to that; your customers who are taping out product to you, are you seeing substantially new designs or are customers being more conservative in what they are taping out to you?

Is that responsible for perhaps more mature designs, or is there any concern about their design capability more leading-edge that would cost those to be more conservative or produce more mature designs?

I think almost all customers are designing new products (indiscernible) the advanced technology, 90 nm in this case.

You are also right in saying that the (indiscernible) in 90 nm technology is getting more difficult, more complex, and the customers are teaching more the (indiscernible) before they do that.

However, what we have observed, I think the ramp up of the 90 nm product and technology have been quite smooth.

The e-learning has been very good, so (indiscernible) success.

Now, with the customer base we now have even more customers, what we call second (indiscernible), coming in with their new products.

Related to that, the move to 65, it may be a little early to talk about that, but do you see a similar transition to 65 when that occurs, or do you expect that to be any different than the transition to 90?

I would expect about the same for 65 compared to 90.

I think both customers and we are learning through this period of time.

We have actually incorporated some of the experiences we learned in our factories into the design for manufacturing guidelines, so that we can work with our customers better and help them with their design.

Okay, thank you.

Fayad Abbasi of Prudential Equity Group.

I've got a couple of questions.

One, looking at your gross margin, you have a nice gross margin improvement heading into Q3.

I was wondering despite ASP decline if you could kind of -- let's say we were to get back to peak level, full utilization levels of 100% plus utilization.

Do you think that you could see your gross margins improve over your prior peak levels?

Yes, we believe that is going to happen, depending on which (indiscernible) you (indiscernible).

Okay, so as you get to like 105% utilization, you think you can exceed the prior peak level gross margin?

If you're talking about in the year 2000 level, which I think for the whole year is around 46%, and I think that is very likely going to happen.

On the margin improvement side, other than we have been doing very good job in execution to bring the costs down, both in advanced technology and in the (indiscernible) technology.

And you need improvement, ongoing efforts.

And also the (indiscernible) depreciation that is coming down continually.

So those factors are, I think, we have very efficient and competitive cost structures.

The ASP will maintain -- cry whale, whale.

Of course, we can generate pretty good margins.

I think last quarter you had given a sense of what the 90 nm revenue was as a percent of your overall.

Could you give us how that looks for Q2?

For Q2, our 90 nm revenue actually (indiscernible) so only representing 2%.

But, however, for this quarter, quarter three, we expect the 90 nm will contribute 10% of our total revenue for the quarter.

We will be well above 10% in fourth quarter.

Can you talk maybe about some of the end-market producing improvement, and that's again to Q3?

Thank you.

We are seeing the end market for this year acting pretty good, especially in the handset and P&C area, notebooks in particular.

I think we believe the business improvement is caused mostly by the demand.

Operator?

Mark Fitzgerald of Bancof America Securities.

I was wondering if you could comment in terms of the change in business model that seems to be going on in the foundry business where many of the customers are starting to outsource design and how that impacts your own business going forward?

Mark, I'm not sure -- you're talking about our business model.

Well, in terms of how do you deal with these companies that just design houses like e-silicon and silicon systems; does that change the way you do business in terms of dealing with some of the established companies you are outsourcing at this point?

No, no.

Not really, no, we have not.

We have been working with companies with what you would call the (indiscernible) companies.

We have been working with them for quite some time.

There is really -- we have not changed our model at all.

Can you give me an idea of what percentage of your business now comes from those type of companies?

We don't have that data with us.

We don't have -- we haven't done any statistics on that.

Is it a growing trend that you do, or is it always going to be a small part of the business?

It has been growing; it has been growing.

How far and how soon it will continue to grow, I don't think we can really comment on that.

Okay.

Just one last question.

Can you give us an idea how big the revenue base is for your revenues from the mask shop, mask operation?

We only -- we (indiscernible) talk about that only a very small part of our business.

The majority of our business is coming from wafers.

Okay, thank you.

Dan Heyler of Merrill Lynch.

My apologies if this has been asked.

I joined the call late, but I wanted to ask about the 200 millimeter efforts to use those (indiscernible), whether or not -- what the major product will be there?

And the second part I guess is on CMOS (indiscernible).

This is an area you guys have strategically focused under and talked a lot about in the past, and just some senses in general, if that could actually be one of the major (indiscernible)?

What is your first question?

Progress in promoting 200 millimeters.

Progress -- 300 millimeter, was that your question?

200 millimeter.

I am sorry, Dan, we still don't understand the question.

I'm calling from Taiwan, so maybe the lines here aren't as good since you're next door.

So 200 millimeter product (indiscernible) is about 20 to 25% (indiscernible) globally, right, and you guys have talked about building out your capabilities in high-voltage, CD drivers, and other areas that take advantage of your legacy technologies.

So I am wondering -- and a lot of companies seem to be talking about the similar products.

So I'm wondering how you are doing relative to your expectations and whether or not image sensors or other areas can fill that (indiscernible)?

Okay, all right.

I think we have been a pretty successful in areas like the CMOS image sensor.

I think we are doing reasonably well with the high-voltage products, the LCD drivers, and now we are moving in to the power management type of products.

We have leveraged our embedded Flash technologies quite well; microcontrollers for automotive, now also for more 16-bit, 32-bit (indiscernible) for other applications.

In addition, I think (indiscernible) very competitive in the market in terms of our performance and our cost structure.

So this comes back, those technologies are being utilized very well.

Thank you, that is great.

Is your internal target -- do you think potentially that you could build the 200 millimeter eventually when the cycle gets a little bit stronger, or would you be happy (indiscernible) would still be very profitable with say 90% utilization?

Is there an opportunity to actually fill it?

We are very happy with our mature technology business.

Our goal, of course, is to (indiscernible) it up, and I feel we have proven in the past and now that we -- always cycles, you know that, Dan.

But overall, our utilization for our mature technology has been quite a satisfactory -- it has been a very good business.

We do not have any plan to get out of this business at all.

What we are doing (indiscernible) in that business will find all kinds of applications.

Right, different applications.

Thank you.

Shaley Jadely (ph) of Fillmore (ph) Securities.

This decline in 90 nm from 10% to 2% (indiscernible) application did you see the decline coming from?

Actually, the percentage of revenue for 90 nm, 2% in second quarter, about 10% in third quarter.

You just said the reverse.

And one quarter it was much higher, right, than the 2%?

First quarter was 4%, you are right.

First was higher than second quarter.

Let me repeat question two.

Which application was responsible for this decline?

The major applications including the (indiscernible) graphics, VOD and DVD recorders.

Okay.

In your (indiscernible) you mentioned three of the segments doing very well; that is graphics, LCD driver and chip sets.

If you could help rank the growth as to where did you see the highest growth this quarter?

Mobile phone.

I think the driver business pretty good, the chipset business pretty good; graphics is kind of flattish.

Okay. (indiscernible) IDM's part of the business if you look at fourth quarter, this accounted for 33% decline to 31; and this quarter it is 26%.

Is it primarily one IDM responsible for this pullback of orders, or you are seeing multiple items responsible for this decline?

No.

I mean this is really -- again, if you go back to several years or even -- you'll notice that our IDM and fabless business has always been roughly at 30%/70%, give and take.

We really do not -- I am not losing my sleep on this change.

Somehow you will come back to (indiscernible) 30%.

What I was trying to get at, is it more secular in nature, given these IDMs are building their own 90 nm capabilities and capacities.

So is it likely to come back in the next two quarters, or are you going to see the decline continue?

Well, we are not really worried about that.

We work with our IDM customer also very closely.

We understand their plans, and the IDM also to advancing our (indiscernible) is continuing.

They will, of course, build their own capacity, but I think it's part of their (indiscernible) policy that they will not build full capacity.

So I think we are a strategic proprietor to all of our IDM customers.

Based on our conversations with our customers, we are seeing our IDM customer is giving us higher demand forecasts with advanced technology in coming years.

Okay, thank you.

Shekhar Pramanick of Moors & Cabot.

Good evening; two questions.

Rick, if you could give us a little color, what do you see in terms a wafer start trends in the Q3?

Is it equally strong as (indiscernible) around May/June time frame?

The second, isn't it (indiscernible) to see the pricing improvement actually comes after overall industry utilization goes up; it somewhat lags?

Lastly, my question is I know you have a -- you'll be getting a major outsourcing from a major Flash player.

Do we need to worry about margins for you in that space?

Thanks.

I think one of the comments commenting on the wafer start, let me say it this way that our booking has been quite good for several months, and this month I think we are still seeing good booking.

We are expecting a good second half of the year.

I think (indiscernible) point of view, there is some kind of a lagging effect.

We believe the pricing environment is improving.

We do not expect pricing will get worse after third quarter.

On the Flash question, our main business has been logic foundry, using logic technology.

We definitely will continue focusing on that.

However, we have been also looking at different markets, different applications, where we have low penetration or low marketshare to see whether we can gain, we can expand global business (indiscernible).

Of course, all those areas we look at, including (indiscernible).

We do not rule out memory for our future business.

An important thing for us is the (indiscernible) business level we take on needs to be a profitable business be they memory or logic.

That is basically how we view those opportunities.

Thank you.

Donald Lu, Goldman Sachs.

I just have a question on your third-quarter inventory.

Do you have an estimate on want will be the inventory level end of Q3?

Are you asking our customers' inventory level or our inventory?

No, your internal -- your own inventory level.

Our inventory actually (indiscernible) into orders.

So our inventory majority was in work in process.

And in the second quarter we had a 14 billion inventory level which was slightly higher than second and first quarter.

But we are also monitoring our number of days very carefully.

We see 12" capacity continuing to grow.

Inventory value -- I will say that number is probably going to grow slightly.

However, in terms of number of days, we are trying to have very tight control of that.

Second quarter we are 41 days and, if you look at our track record in the past few years, it's around 40 days.

So we'll try to maintain that.

So end of the Q3 on an inventory days basis it will be around still 40 days?

Around 40 days.

It has been our historical record for many years -- 40 plus/minus (ph) few days.

Okay, great.

Thanks.

Nicholas Gaudois, Deutsche Bank.

First question, to clarify the Q3 (indiscernible) trends, you're saying you said SCD (ph) drivers, PC chipsets, wireless up, graphics plus.

Would you care to make any comments on consumer networking in terms of the trends you're seeing in the first quarter?

Networking I think -- or actually second quarter we have seen pretty good growth compared to first quarter already and I think that will continue.

Consumer products -- in the third quarter is seasonally a strong quarter and we certainly expect that to be the case this year, too.

Okay, great.

In terms of your 90 nm design you're adding incrementally in the next couple of quarters according to what you just said two questions ago, in which application markets would you see (indiscernible) adding incrementally?

Right now the main drivers are in the wireless base band, the graphics for P&C for game consoles, we have POD and the DVD recorders.

These are also ramping these products and the -- more products are being ramped up in higher volume.

I think going forward we are seeing actually other applications coming in.

I think networking, for sure some of the network processors -- some of the set-top boxes.

Actually (indiscernible) I think (indiscernible) yes, digital TV.

Since people are designing in -- especially since the adoption ratios really are quite high now.

Okay, that's great.

Last question on the capacity (indiscernible).

If we look at your estimate for fab 12 A&B and fab 14, fab 12B your estimate will be about 12,000 to 13,000 wafers per month capacity by Q3, fab 14 about 15.

Would you care to give us an indication about where you would be by the fourth quarter?

Would you be higher than this or are we going to see a flattening of ramp there?

I think our fabs 12 and 14 continue ramping at 12 and 14.

By the end of this year they'll be -- about 12 -- in the end of Q4 we will have 46,000 roughly.

46,000 in total.

Okay, thank you very much.

Jacky Tu (ph), Jefferies & Co.

This is Jacky calling for John Mal.

I have a couple of housekeeping questions first.

How should we model the investment credit tax and the share count for Q3 and Q4?

The first investment -- you're asking investment tax credits?

Yes.

You can assume the same as Q1 and Q2.

How we -- we amortize that investment credit over years.

Okay.

And what about the share count?

Outstanding shares -- after the stock dividend of 24.7 billion shares.

Okay.

And then finally, I was wondering what your company's perspective is that's in your discussion with customers on the revaluation and further revaluation of the yen in subsequent quarters and years?

And what's the impact to your business and specifically what's your business in China in terms of the impact there?

You're asking (indiscernible) appreciation?

Correct.

Firstly our operation and China is still very small.

Given the appreciation is only 2%; the impact will be insignificant to us.

Any thoughts on what's going to happen in 2006 let's say if your capacity increases in China and your valuation actually continued to increase?

Depending on how much appreciation would happen in the coming future, but if you're taking into account that our China fab roughly speaking versus (indiscernible) capacity is too small.

So I don't think that impact is going to be huge.

Okay, thank you.

Pranab Sarmah, Daiwa Institute of Research.

Thank you.

I have one question regarding CapEx and revenue guidance.

AT the beginning of the year you had guided like you'd like to spend about say low 30 percentage point of revenue in CapEx in 2005.

And now you said that you are probably going to spend probably high end of your CapEx.

Do you still hold that particular guidance?

We normally (indiscernible) link the percentages of revenue as our guidance, but it just happened.

So this year it's very close to 30%.

And this year we are going to spend roughly 2.7 billion. (indiscernible) for next year we have not decided yet.

We're looking to the demand and the capacity for 12" and then we can make a decision on our end.

But as I said earlier, we are looking at longer year trend, not just for one year.

I think the CapEx for 2006 will be heavily related to the outlook for 2007 and it's pretty much premature at this moment to talk about 2006.

What about 2005?

Still you hold that low 30 percentage point of revenue (indiscernible) CapEx.

In that case, your revenue will reach probably 260 to NT$270 billion.

CapEx, which was (indiscernible), will be 2.7 billion.

I don't know how to comment on percentage to revenue, but you have your model.

You have revenue estimation.

You just 2.7 billion divided by the revenue you are modeling, you get percentage.

Okay.

Next one is have you seen like your orders, probably demand is coming more from the consumer and that demand rather than corporate (indiscernible) who has a more seasonal buyers in this up-cycle compared to last up-cycle?

I think it is difficult for us to tell in good (indiscernible) what's between consumer and the corporate.

Because it is difficult for us to say, for instance on handsets, you call that consumer or nonconsumer.

But what we can say is the (indiscernible) the wired application which probably related to corporate more than the consumer, it is going quite well.

I think what we are seeing is a fairly broad-based growth among all applications.

The end-market demand seems to be doing fairly well.

Thank you very much.

Operator?

Can we have the next question?

Bhavin Shah of J.P. Morgan.

I just had a question on .18, .15.

Can you comment on which maybe is top five applications that (indiscernible) .18, .15?

That's a pretty difficult one, Bhavin.

Very broad based application.

I think -- I cannot guarantee you that it's very accurate.

A chipset -- (indiscernible) chip from mobile phone applications due to wireless LAN, DVD players -- we've got probably a pretty major -- you asked for 5, right?

I gave you six, so --.

And I guess earlier you commented on gross margin being higher than the previous cycle at the given level of utilization.

I'm not sure if you want to quantify that or not but let me try.

Let's say at the current level of ASP your utilization was about 100%, what will your gross margin be?

Well, the gross margin would be quite good.

You also need to take into consideration the exchange rate because we are expressing our books in NT dollars.

In the 2000 year the exchange rate was 33+ and now it's 31.

So if you're just looking at the 2004 versus 2005, the NT dollars actually appreciate around 7%.

So that has to do -- a huge impact to our margin as well.

So if you assume the same level of the exchange rate the margin is going to me in the same level or the (indiscernible) peak or even higher.

Okay.

And I have a question on -- there's been some debate about alternative (ph) minimum tax in Taiwan.

Could you offer your perspective on that if at all it effects TSMC or not?

It will have some impact, but actually we are going to pay some tax starting from next year.

So depending on what is the tax rate will be considered for a minimum tax.

If the tax rate is 10%, for example, we will have slight impact for 2006 and 2007 and a little bit more impact on 2008 afterward.

Okay.

Thank you.

Vishal Shah, Lehman Brothers.

Can you just clarify your guidance for Q4 capacity?

You said that would be 46,000.

Was that for the full quarter and that's for both fab 12 and 14?

Thank you.

We did not give guidance for fourth-quarter capacity.

But however, we did say the total year capacity is 5.96 million pieces.

And we have said the first-quarter capacity was 1.32 and second quarter 1.45 and third quarter 1.55 and fourth quarter -- naturally if you take 5.96 minus those three numbers you get --.

Right.

No, I was talking about your ramp plan for fab 12 and fab 14.

I believe you said that your fab 12 -- the wafer shipments in that fab will be about 46,000 by fourth quarter.

I was talking about capacity, not shipments.

Okay.

That's it for fab 12, right?

Fab 12.

Okay, great.

Thank you.

Ivan Goh, Dresdner Kleinwort Wasserstein.

I have two questions.

The first one is you have listed a couple of devices that will drive your 90 nm ramp into the second half.

Can you just pinpoint the most important two, the fastest-growing application in 90 nm going into the third and the fourth quarter?

That would be the base band chips and graphics.

Those are the two major drivers.

Okay.

And my second question is today you said that the CapEx for the full year will be closer to 2.7 billion, which is at the upper end of your range, and you also made a comment saying that next year's CapEx will very much depend on your outlook for 2006.

There's a two-part question -- for 2007 -- a two-part question.

The first part is -- is there any change internally between your previous guidance of 2.5 to 2.7 and now saying that it's closer to 2.7?

And secondly, how does this change back -- how is it caused by a change in your outlook in 2006?

Actually the range was (indiscernible); it was only 2.5 to 2.7 is number one.

And over years we are looking at the demand for 2006 especially on advanced technology.

So when we see a stronger demand we adjust a little bit, that's how we operate the CapEx decision.

I think the plus/minus .2 billion is not considered a significant change.

Okay.

I asked this because also in the afternoon you mentioned that you didn't change your forecast for 2006 for the industry.

So I was just wondering how it all comes together?

We believe really -- earlier in the year we believe our demand for advanced technology will be strong in the second half of this year.

We also believe that that momentum will carry itself into first quarter next year.

So we decided to invest accordingly. (indiscernible) both of the numbers which, of course, (indiscernible).

It's about $100 million.

Plus or minus 100 million.

Thank you very much.

That's very helpful.

Operator, in the interest of time we will only accept one last caller's question.

Mehdi Hosseini, FBR.

Mr. Hosseini is not on line.

Can we just move to the next one?

Shekhar Pramanick, Moors & Cabot.

Just a follow-up question for Rick.

What's your sense in '06?

Do you think '06 over the year when TSMC again outperforms the semiconductor industry because you might be picking up some significant outsourcing particularly maybe in the memory side?

Our view on '06 semiconductor is that we'll be -- the semiconductor industry will grow high single digits for '06.

We also believe that the foundry industry will grow better compared to the semiconductor industry as a whole.

As the leader of the foundry industry we are -- certainly expect that our goal was to outperform the foundry industry.

Overall the growth will be from our overall business, not necessarily as you stated from memory mainly.

Thank you very much.

Thank you.

Operator, I think that would be the end of today's Q&A, right?

Yes, ma'am.

Thank you very much for attending TSMC's second-quarter conference.

We are looking forward to seeing you next quarter.

Bye-bye.

Before we conclude the TSMC second-quarter 2005 results webcast conference call today, please be advised that the replay of the conference call only be accessible through TSMC's website at www.TSMC.com.

Thank you all.