意法半導體 (STM) 2003 Q3 法說會逐字稿

Good morning.

My name is Linda, and I will be your conference facilitator.

At this time I would like to welcome everyone to the STMicroelectronics Third Quarter earnings conference call.

All lines have been placed on mute to prevent any background noise.

After the speaker's remarks, there will be a question and answer period. (Operator Instructions) Thank you.

I will turn the call over to Ms. Lynn Morgan, ST Group Vice-President of Corporate Development.

You may begin your conference.

Thank you, operator, and thank you all for joining us today.

I would like to introduce the people sitting around this table.

First of all, as you know, Pasquale Pistorio, our President and Chief Executive Officer, will host today's call.

He is joined around the table by Aldo Romano, Corporate Vice-President, TPA, which is Telecom, Peripherals and Automotive Group, Carlo Ferro, Corporate Vice-President - Chief Financial Officer.

We are sorry to say that today Alain Dutheil, who is normally with us, our Corporate VP of Strategic Planning, and Carlo Bozotti, our Corporate Vice-President in charge of Memory Products Group, are not available because they are attending the funeral of one of our former colleagues, the gentleman who formerly ran our supply chain division.

Also, Philippe Geyres is traveling on business.

So, those gentlemen were nice enough to download as much as they possibly could to Stan March, our Vice-President of Investor Relations, and Benoit Dulisse, our Director of Investor Relations, and those two gentlemen will, hopefully, be able to your questions on MPG and CMG.

At the same time, I would also like to take this opportunity to say two things.

Number one, what I am legally bound to say is that on this call we may make various forward-looking statements regarding the future business and financial performance of STMicroelectronics and the semiconductor industry.

These statements are predictions that involve risks and uncertainties, and actual results may vary materially.

We refer you to certain pages of our 20-F, and to previous announcements, and to ST's press release issued yesterday regarding some important risk factors you should consider in evaluating this information.

That I was forced to say, what I am pleased to say is that our 2004 field trip is to take place, I believe, in Paris.

And the dates would be our dinner on May fifth, and the event on May sixth 2004.

And, We are hoping that we will have a great turnout, both from the European financial community, but also from the US and Asian financial communities.

At this point I would like to turn the call over to Pasquale Pistorio.

Thank you, Lynn, and good morning and good afternoon, ladies and gentlemen, and thank you for joining us today.

I will review Q3 financial and operating results, discuss our outlook for Q4, and share our thinking on ST positioning for 2004.

Then, my colleagues and I will be glad to take your questions.

As you have seen, the Q3 revenues came in about to $20 million above the high end of the guidance range we provided three months ago, positioning a 6% sequential increase over Q2, and a 9.6% increase over last year's third quarter.

Several applications within our target market segments were strong performers in the quarter.

In Digital Consumer, set-top box, and NDTV showed a strong sequential revenue growth.

Wireless, comprised of the differentiated Memory Products and others, reached about $430 million, a double-digit increase over the prior period, and accounted for about 24% of our Q3 net revenues.

Data Storage led the computer peripherals segment, as market applications benefiting from recent acquisitions, were an important factor in the Industrial segment.

Whilst traditional Automotive performed well, overall Automotive performance dipped slightly from Q2 levels.

The balance, which characterizes our end-market performance also served us well in our product families.

In addition to a solid 5% sequential growth rate in differentiated products, Micro & Memories was up 21.4%, and Discretes increased 4%.

Despite continued press pressure, Flash memory product sales increased over 15% sequentially, to $191 million, and represented 54% of MPG's revenues.

This relates only to our low Flash business.

In October we started sampling customers with 12 Megabit and 1 Gigabit end products, which could be in volume production by Q2 next year.

As anticipated, overall pricing pressure penalized our Q3 gross margin performance, which, at 35.1%, was 60 basis points below Q2's 35.7%.

Product mix was also a contributor, given the sequential increase in sales of Flash memories.

Utilization rates declined in the first half of the quarter, bringing down the average for the period to about 78%, from approximately 83% in Q2.

This was a function of reduced loading of certain 6-inch wafer fabs, and an increase in our overall capacity.

Inventory levels remained constant with those of the prior quarter.

As the reduction of software-related inventory was offset by a build-up to support increase in the markets of demand for shipments inQ4.

Inventory tools, however, increased to 4.25, just at the point of our targeted range for Q3.

We maintained our emphasis on discretionary expenses control in Q3, helped by these seasonal factors.

R&D and SG&A expenses together represented 27.4% of our Q3 revenues, an improvement of 130 basis points over Q2 levels.

With respect to the breakdown of the operating income by product group, you will have seen that TPA's operating margin declined sequentially.

This is the result of the inventorial reduction that I mentioned before.

Conversely, CMG's operating margin in Q3 benefited from a better mix, and increase in inventory to accommodate strong bookings for Q4.

Our news release details the restructuring plan, which we announced at the time of our Q2 earnings, and defined during Q3.

We are projecting that the full plan will result in a pre-tax charge of about $350 million, or $240 million after tax, of which about 50% involves non-cash items.

Approximately 55% of the charge was included in Q3, and the remainder will be taken over the duration of the plan, which should be substantially completed over the next 18 months.

The plan's objectives are to increase cost competitiveness, and to enhance the capacity, which is part of our overall strategy to raise profitability levels by expanding gross margin.

Although impermanent restructuring charges resulted in operating net losses for Q3, we believe that the restructuring plan is an important step toward enhancing ST's performance potential in the period ahead.

Our balance sheet at the end of Q3 remained very strong.

Cash and marketable securities exceeded 2.7 billion, versus a total debt of a little over 3.1 billion, giving us a total debt-to-equity ratio of 0.05.

Thus, we have the flexibility to maintain this position, or to leverage it, depending on market conditions and business opportunities.

And, our Q3 debt restructuring initiatives have reduced our average interest rates on top of debt to 1.2%.

Importantly, we continue to generate enough cash from operating activities.

At the end of the first nine months of 2003, free cash flow, before acquisitions of $135 million, was $290 million, after sustaining at the pre-investment levels to support the transition to 0.13 micro and 90 nanometers.

I would like to point out that this is our eighth consecutive quarter of positive free cash flow before acquisitions.

Leveraging our silicon and system annual, we are taking full advantage of internal growth opportunities.

In Q3, revenues from our 12 strategic customer alliances of $760 million, were about 3% above Q2 levels, and represented 42.1% of total revenues.

Thus, much of our sequential revenue increase in Q3 was tuned by product sales to a broader number of key customers.

Our accelerated marketing programs involved a broadening of our product offerings to an expanded customer base through three key initiatives.

A - by creating regional competence centers to focus on specific applications;

B - by increasing design activities in the demand of the products with multiple applications; and, C - financing a new generation of e-tools for customer support.

Looking by application, we remain focused on our target market segments.

I mentioned our virus number earlier, currently tracking at an annualized rate of over $1.7 billion.

And a word for Flash, it is getting better.

Not yet profitable as it needed to be, but after a long period of price erosion we are seeing some stabilization and our orders are picking up significantly.

Now on to our short to long-term perspectives.

As we noted in our new release, we expect Q4 to be another quarter of solid growth for ST.

Our guidance of 6-12% sequential revenue growth is based upon our Q4 backlog, and indications we are getting from key customers.

Our revenue growth drivers are likely to be Wireless, printers, and mostly digital consumer applications.

And - Flash memory should achieve a significant double-digit sequential revenue increase.

Gross marginal costs on the currency basis should be between 36-37% range, which we anticipated in July.

This range takes into account the effects of Italy's blackout, which will cost us about 50 basis points in gross margin.

The positive contributors, in addition to increased revenues, should be a higher utilization rate, and related manufacturing efficiencies, while pricing will continue to be challenging.

As you read in our new release, our 2003 capital expenditures should be closer to 1.2 billion than the $1 billion we had anticipated previously.

This reflects our increasing confidence in the industry's near-term growth prospects.

And, our preliminary capital budget for 2004, of 1.6 billion, is up about 33% on a Year-over-Year basis, with the majority of these funds marked for the leading edge 0.15 micro and 90 nanometers technologies at our fabs.

Moreover, ST's long-term strategy of building enough demand, manufacturing infrastructure, will make us capable of expanding modularly.

This leading-edge capacity has demand and falls, thus we believe that ST will be very well positioned to profitably take advantage of the favorable industry growth trends that are expected for 2004 and 2005.

And I will end with our focus on our top-line growth and market-share gains and our emphasis on increased profitability levels.

Therefore, we are taking measures that, we expect, will enable us to accelerate and expand our gross margin.

These measures include manufacturing restructuring and efficiencies, and expansion of the feature region customized products we are selling to a larger customer base, which should benefit our overall pricing and product mix.

At this point, ladies and gentlemen, we will be pleased to take your questions.

(Operator Instructions) We will pause for a moment to compile the Q-and-A roster.

Your first question comes from Nav Shira of Citigroup.

Hi, good afternoon.

Nav Shira calling from Citigroup.

Just two quick questions please, Pasquale.

Firstly, you mentioned your rollout onto 0.13 micron.

That's obviously for some of the more commoditized parts like Flash memory, but how much of your other product range do you expect to have on 0.13 microns by the end of 2004?

And I have a second question on gross margins, you indicated in the press conference earlier today that your gross margin target of 40% is likely to be achieved during 2004, what is your belief in seeing that being raised above 40%?

Because competitors to you, especially in the US, have seen gross margins of 43% or even advertised 49% even for recent results that they have reported.

Thank you.

Thank you.

Well, first of all, the 0.13 microns and, even more important, the 90 nanometer, which in 2004 will evolve, is not only for Flash.

It is also for the multiplicity of the system chip that we are building for Wireless, for Consumer Application, for Disc Drive Application, and so on and so forth.

So, we expect that a larger variety of our system chip, and of course Flash, will be in those technologies.

Today, I would like to recall that we have several fabs that have been built to these capabilities.

Of course, Crolles, too, is by nature a 90-nanometer R&D and pilot line.

Then, our plant in Rousset is 0.13 microns and 90-nanometer capable, as well as our fab in Singapore.

While we could upgrade, and we will, also our facilities for example at Crolles or Catania.

So, we have plenty of facilities that are built for 0.13 microns and 90 nanometers, or are upgradeable.

And products involve Flash as well as system chips.

As far as the gross margin, one thing I would like to note, we have chosen, since a long time, a strategy, of a mixed portfolio, focusing on the fringing of the products, but also memory products on this preset standard.

We do internal benchmarking with companies compatible.

So we take companies that are, essentially, differentiated products, and we benchmark our differentiated product with those companies.

We take companies that are essentially memory and we benchmark our Memory products with our competitors that are essentially in the way.

We take the discreet and standard product companies.

When we do that benchmarking, our gross margin is absolutely comparable.

In other words, if you take a company that is essentially discreet and standard, and benchmark with our TSG group, they are not only comparable, but we have nothing to envy.

Similarly, if you take, I don't know, EPM and MG with companies that are essentially the fringe of the products.

This is for the establishing of the fact that our gross margin capabilities, right now, are not inferior.

Moving to next year, there is an overall progress towards - because, remember, for example, Memory products have a very volatile gross margin.

During the high price competitiveness and excess capacity, they undershoot the company average.

In the other opposite, they overshoot, so they become even, a higher gross margin than the company average.

So, an improvement in the environment of the market, continued introduction of new higher value products in all the product groups, and an increase of manufacturing efficiency that will come from the rationalization of the mature 6", but also in a debt saturation level of the leading-edge hubs to the capital investment we are doing.

All those factors are contributors to gross margin improvement, and without saying when, because we do not give this guidance, but we expect next year, at a certain point in time - the sooner, the better - to go above the threshold of the 40% gross margin.

Thank you very much.

Your next question is from Mark Lepassis of Prudential.

Thank you, a couple questions, if I may.

On the revenue guidance for Q4, does that assume more or less returns than you received in Q3?

And regarding a model, how should we think about modeling depreciation and operating expenses over the next several quarters?

Thank you.

Ok, about the inventory return business, we expect less inventory returns because we have a more solid backlog, but in any case if they come we welcome them, which means sales will be higher.

But, if I can say so, he point of entry of Q4 is even more solid than was our point of entry of Q3.

And as far as the rest, I think Carlo Ferro can comment.

Yeah, good afternoon, good morning everybody.

Depreciation and amortization this quarter has been between 4 and 5 million.

We expect in Q4 a slight increase to range between 430 - 450 million.

And so, for the year, the total depreciation falls in the high end of the range that we have always anticipated, of between 1.5 and 1.6.

Indeed, this is going to be 1.6 for the year.

Next year we expect depreciation and amortization to increase up to 1.75 billion over the year, the major reason of that being the effect of the exchange rates that the company is incurring during the year 2003.

I guess your second question is on the operating expenses for the next quarter, and in this respect the remarks by Mr. Pistorio refer to seasonal factors for Q3.

We should also take into account that by a technical reason, calendar days in Q4 are 3% higher than Q3, and all these factors combined may lead us to expect expenses next quarter to be higher than Q3, in arrange between a plus 5, plus 7%.

In any case, on the expenses front, I want to say that the company is always very, very cautious in cost control.

But one thing we do not want to control is R&D, in the sense that we want to continue to accelerate R&D expenditure, and, secondly, with the new drive we are keeping, as I said before, to leverage on our product portfolio on a broader customer base, and also approximate our marketing campaign.

Thank you.

Your next question is from Andrew Griffiths of Merrill Lynch.

Hi there, I just have some questions on the division profitability figures you gave.

Starting with Discrete and Standard, which had a big increase in operating profit - higher, even, than the sales increase - could you explain that and talk about why, whether that will be sustainable, please?

On what?

On Discrete and Standard?

Yeah.

Maybe I calculated this wrongly.

I got a 14% margin from that division.

What is wrong with that?

It is just a very large increase compared to Q2, which was a few percentage points below that, and revenue does not increase by very much.

This is Pablo.

Indeed, what happened in DSG is a very beneficial selection on the product mix, which, somehow, have depressed the ability to grow this quarter, with only 1% sequential growth, but has boosted the margin, since Salvatore Castorina, the General Manager of the business, has been very selective in order to improve the overall profitability of his product portfolio.

And we expect it can somehow continue in the next quarters.

So that is sustainable.

Ok, then, if I 'X' out your restructuring charge in the other income lines remaining in the divisional breakdown you gave, it looks as though that number increased, about doubled to about $60 million.

I just wondered what kind of number we should be looking at modeling, going forward into Q4 and into 2004, for that other income loss item.

Yes, I'll say that usually this number in the previous quarters used to be around 35 million, and this, as you know, includes the famous strategic initiatives under the sponsorship of the corporation, and in order to boost this initiative, and this is the reason why it increased also in Q3, this number could increase in a range between 40-45 million, per quarter in the next quarter.

And, this number also includes the stock top of Q2 that we are in a range of 17 million this quarter.

We expected to continue basically along the same path in Q4, but they should discontinue, starting from Q1.

OK, and in the final, to wrap up these divisional questions, in TPA, I would have expected sequential growth in revenue, given that there is some wilder stuff in there, and also from the computer peripherals.

What was the problem here?

Was the Automotive part of TPA seeing negative sequential growth?

Not at all.

This is Aldo Romano speaking.

I am in charge of TPA so I am the right guy to answer your question.

So what happened in this quarter is that basically the Computer Peripherals were flat overall.

So when you see the numbers, sequential growth was negligible, but, in effect, what happened was excellent growth in Data Storage, and instead some, decline, not totally unexpected, in Printers.

This is only due to temporary inventory reductions that our customers decided to do during this quarter.

In effect, in the next quarter, we are planning to grow further in Printers, also.

So, you mentioned Automotive and, in effect, we are giving a number concerning all automotive products, because it means other products that are going to Automotive customers.

That is why we do our statistics.

There are the true, dedicated products specific to automotive.

There are also commodities that are going in the same segment.

And what happened in this quarter was that there has been a growth of about 7% on true automotive products, and some decline on Commodities, which is probably due to the decline in the production, which is seasonal, in the third quarter of the year, particularly in Europe.

This is my interpretation of this decline.

So, there is nothing to mention, and I restate the fact that, in Automotive, ST overall, particularly in this dedicated product, is performing very well.

We will grow in next quarter, also, and we will grow a lot next year.

So that 7% number, would that be the TPA portion or automotive dedicated parts?

The dedicated products are obviously more than one half of the total.

I do not have the precise answer to your question, but it is in the region of two-thirds Dedicated products and one-third the other, more or less.

This includes transistors, Memories, whatever.

Ok, that means that the Telecom part of TPA, then, was weaker, was it, in Q3?

Could you elaborate on that?

Well, partially correct, in the sense that the performance in the quarter, in the Access and Networking has been below our expectations.

Instead, the quarter was very good in cellular phone devices.

And so the growth has been exceeding the company growth, so it has been a good quarter for cellular phone devices.

And again, also here we expect, clearly Q4 is always the best quarter of the year, so there will be continuous growth also for cellular phone applications.

Everything included.

And I would like to mention here what has been recent in the- has been mentioned, what Pasquale has mentioned that, overall, the growth that we are having in products for cellular phone application in the quarter, is, sequentially, has been 12%.

So, it has been a good quarter.

And also, if you compare this to one year ago, this is about 18-20%.

So this is the rate, what is happening in Wireless products?

And that's for all your Wireless products across the board?

Both TPA and other?

Yes.

Great.

Thank you very much for the comprehensive answers.

Your next question is from Adam Parker of Sanford Bernstein.

I have a couple of questions.

One, I just could not hear you.

I'm just not sure I heard you right.

Did you say you expected operating expenses to be up 5 to 7% sequentially in the fourth quarter, and if you said that is that on a constant currency assumption?

yeah, well - Carlo Ferro can comment again.

This is on a constant currency assumption, correct.

Usually we base these expectations on the exchange rate when entering the quarter, and of course these forecasts include already the deterioration in exchange rates incurred during the third quarter.

Ok.

Great, I've just got a couple other questions.

One is just sort of, on the gross margins going forward, on the incremental gross margins, I'm trying to think about how I should think about the leveraging your company going forward over the next couple of years.

I know you articulate your plan here for expanding your customer base, and restructuring and improving efficiency, but I'm trying to think about how I should think about incremental margins will there be over the next couple of years?

Could it be as high as the mid 60s or higher?

And, why is it so much lower than, say, the 70%-plus incremental margins that the other large semiconductor companies get?

Let me first say again that it is not that much lower than other comparable companies.

As I said, if you take a company that only makes differentiated products you can choose, and compare it with TPA and CMG, gross margins are very comparable.

Pasquale - Sorry, maybe I was not understandable.

I meant on the incremental gross margins.

In other words, if I look at, say, Texas Instruments, and they show me 80% -

Sorry.

That is a very good point.

Yes.

Compared to last year the incremental was fantastic and our incremental was negative.

But this is product mix.

Adam Parker.

So it is all mix?

How should I think about it going forward over the next couple of years, is sort of my question?

Let us take one big family that of course is under the attention of everybody - Flash.

Flash has been the one growing the most, in terms of percentage, and, therefore, increasing their weight sensibly in the last several quarters in those several months.

And this has been the family with the worst gross margin deterioration, due to the extreme price pressure.

And this has an impact on the total company.

Now, once the Flash price stabilizes with the recovery, the gross margin tends to be above the company average, like it was in the year 2000 and 2001.

At that point what you have is not a detractor, but a predicator.

So, I think that the company with a big product mix has a kind of cost and trend in some product families, and an oscillation due to the variability of other product families.

This variability will be very positive when the market is positive, and very negative when the market is negative.

So, I expect that in the coming years 2004 - 2005, with the market in a much better condition, the families that are being more vulnerable to price pressure will become into major contributors to the gross margin.

Ok, I guess I just meant maybe on a normalized basis.

Let's just say the next few years, so we're just saying what is the normalized ability of this company, as revenue increases, what should your gross margins be on that additional revenue you get on a normalized basis?

On a normalized basis, I think that over a long period, the company would be a step over the 40, and I would say that my desire is to see the company going above the 42 -4 4 range, once we move into a solid market.

I think the core portfolio of the company in the full cycle the service to be above 40, and that is what we are going to build for the future.

Ok, one last question.

You mentioned here a couple of times that you think the industry conditions are expected to improve in 2004 and 2005, and I just want to, given your experience of all the cycles you have managed through here, what gives you any conviction that the cycles are going to last into 2005?

Well, what drives our industry has been always in this field, the same.

On one side, demand driven by the economy, and, on the other side, the balance or unbalance between demand and supply.

The economy looks like it is picking up, and we make the assumption that there will be the expectation that the economy is to project for the next couple of years.

On the side of demand and supply, this is very evident.

If you look at the capital spending of the industry in the last many quarters, you will see that this has been below the median of the industry.

So there has been the build-up of potential under-capacity, and we believe that this is exactly what has happened if you go through all the other cycles, the conditions were a couple of years of sustained growth with price stabilization.

Now, to be more precise there, I believe, frankly, that in the more mature technologies, the situation of overcapacity will be standing longer than in past cycles, because there is a lot of the capacity that disappears from the European or American markets and reappears in China.

So, it is not capacity disappearing, but just transferring to more competitive areas.

But, as far as the leading-edge technologies, there is not too much capacity available.

Therefore, and there we will see the net debt better opportunity to reasonable price establishment, and even price increases.

In the leading edge technology companies that have the capital capability, the financial capability, and the long-term positioning to invest, will be in a better position, and that is what we expect.

So we are trying to accelerate our migration to final geometries, both in products and fabs.

Great.

Thanks for your time, Pasquale.

Thank you.

The next question is from Jean d'Anjou of CSFB.

Hi.

Good afternoon, gentlemen.

I have two questions.

The first one is on Flash.

Could you tell us more what was the price decline on quarter-to-quarter in Q3, and what you are expecting for Q4?

It seems that still discount prices are dropping and that's pretty surprising because volumes are picking up.

Why do you think it takes so long to see pricing Flash or of non-Flash.

Jean, this is Stan March, just to reply.

The pricing pressure that we experienced in the Memory products group in the third quarter was approximately 6.5%.

Without being specific about Flash, it was higher than that, in the neighborhood of 8%, which was slightly higher than we had anticipated, going into the quarter.

So, with respect to fourth quarter activities, while there is broad based demand, there is still some pressure on price at the high densities, and a little bit of stabilization at the lower densities of Flash, so we expect some pricing pressure to continue into Flash in the fourth quarter.

There has been demand at both ends of the density spectrum, to your point, and certainly it is important in this regard to consider that there are dynamics, because they are different end markets that they are drawing, whether it is in the Consumer and Wireless in the higher densities, or whether it is more products along the lines of Computers or Peripherals in the low density.

So there are slightly different pricing dynamics.

So, it is difficult to answer, based on the industry or the end market dynamics, or to give a holistic view for us at this point.

It is safe to say we expect some continued pricing pressure, hopefully not on the same magnitude as what we experienced in the third quarter.

Ok, right now the follow up, what makes you so confident on the future to decide to go for such a big increase in CAPEX in 2004, whereas as your margin score is still below average for recovery?

I would like to recall what has been the strategy of the company in this area.

We build a demands infrastructure.

So we have a big plant ready in Singapore, big plant ready in Rousset, big plant under construction in Catania, and a big plant ready in Crolles, et cetera.

But we bring in equipment capacity in a very modular very.

So should the demand unfold lower than what we expect, we would cut substantially the 1.6 we are saying.

Should the demand unfold, as we want - we want to put this capacity in, because it will translate into higher gross margin products, because of leading edge technologies.

So, the modularity is the secret.

We build infrastructure rate of a schedule, and then we move in the equipment.

As you know, by infrastructure I mean the building and the limited extent of the facilities.

This represents a small portion of the total capital.

The real big capital is in the equipment, and the equipment we bring in modularly.

So, if the growth of the industry will not be realized, in which today most experts believe will be in the range of 18%, and then, with our confidence that our portfolio to service a much better growth than the industry, then we need the capacity and then we are ready to go.

On this expectation, we are saying that capital spending we anticipate to be 1.6.

With even more the growth, we can spend more; the less the growth, we can spend less.

The modularity is the secret.

Not to be sure that we have the crystal ball.

I don't know if that was clear.

Basically, my question was referring to the concern that some of your competitors are spending less and are having a much better margin leverage at this point.

Some of our competitors are spending less, and some, like SumSum or Intel, are spending more.

But they have high margins.

They have higher margins, I do not know the assumption, their margin is not reported.

But if you compare apples and apples - no.

In other words, if you could compare the Flash margin of our competitors with our Flash margin, I do not think that there would be a difference.

And if you compare our leading-edge system-chip margin with our competitors' leading-edge system-chip margin, the answer is that we are absolutely comparable.

And in this capacity, we will do- vary the demands of products like the new generation of wireless platform products, the new generation of consumer platform products, the new generation of automotive platform products, as well as Flash.

So, there are all kinds of things.

Ok.

Thank you.

Your next question is from Cody Yakri of Legg Mason.

Thanks.

I hate to harp on the gross margin question, but with the restructuring coming through, obviously that is part of the improvements to the incremental margins.

Can you talk about next year, as you are working through the program, how quickly we can start to see some differential to those incremental margins?

And you talked about kind of normalize margins on today's base, but how much of an additional lift can we expect to see to the bottom line leverage through this restructuring?

Well, the restructuring is, essentially, for manufacturing, and not for the bottom line.

We are referring mainly to 6" plants.

So this will improve the manufacturing costs of products that run thought those lines, because at the end what we will have is increased of the daily output using that head count.

So, it is making much better efficiency if we reduce head count, reduce power and gas, or reduce materials, etc, etc.

The cost benefit we mentioned is about $120 million per year on a standing base, once it is done.

I am sorry - $140 million on a normalized base once the restructuring is completed, and it will be all in manufacturing costs, because what we do, we are just migrating 6" products to 8" final geometries when possible, or the same 6" into the6" fab in Singapore, which is cheaper than the US or Europe.

So, those things will give us substantial manufacturing advantage for those products, but not for the bottom line.

And the bottom line will be the leverage of the volume.

The leverage of the volume will be able to increase the margins, because the R&D on the platform is about the same, and bigger than the expenditure by the group that will leverage by the volume.

Ok.

Maybe you could talk a little about seasonality and the commodity price declines that we are still seeing?

You said you were seeing maybe a little larger price declines in Q3 than you expected, especially in Memories.

Assuming that we have some normal seasonality in Q4, I'm sorry in Q1, can you talk about what you would expect normally as far as seasonality into Q1?

And then if we are not getting the stability of the prices during the seasonally stronger Q3/Q4, what would you expect for prices in the first half?

Of next year?

Of next year.

If the situation between demand and supply improves, in other words if the imbalance in the overcapacity will be reduced as we expect, the overall pricing picture will be firmer.

If not, and we do not know how to anticipate better than what we have said until now, if not, there will be price pressure.

But the company will not be depending on that.

We will keep doing the things we are doing, which is, again, migration to more advanced products and to final geometries.

When you move Flash to 0.15 microns, which is the bulk of our production today, to 0.13 microns, which is the bulk of our production next year, to 90 nanometers, which will be the bulk of our production in the year 2005, every time we do this we will dramatically increase the margin, because you get many more dies from the same wafer, and, therefore, you have the ability of leverage in your technology.

This is depending from us.

If we can do this better than our competition, which we hope to be able to do, we will have pricing capability in the industry.

So we will do the things we need to do internally, and the market will add whatever component will be determined by the capacity relationship, which we expect to be variable in the coming couple of years, compared to what it has been in the last two years.

Pasquale, looking back over the years.

Can you give maybe your view on what your normal seasonal pattern is into the first quarter, and do you think that this year you now have a different product mix and a end market mix, that would change that normal pattern?

Historically, I would say that Q1 is weaker than Q4.

In the company's history, due to our market mix, Q3 tends to be the lowest quarter, followed by Q1, with Q2 and Q4 being the strongest quarters.

This is the historical pattern.

Now, even in our history, in years of strong market recovery, Q1 is even better than Q4.

It is rare, but we have had cases of Q1 being higher than Q4 when you are in a stronger recovery.

This could happen, but I don't know.

It is too early to speak about Q1.

My point is that seasonally Q1 is usually weaker than Q4, and seasonally Q3 is usually weaker than Q2.

This has been the seasonal pattern.

This year, Q3 being higher than Q2 in terms of revenues, is an indication that the market is in recovery mode.

Great, thanks guys.

Thank you.

The next question is Didier Simama.

Good afternoon, gentlemen.

I had a few questions, maybe starting with CAPEX, you are increasing your CAPEX on easement on next.

I would like to know, if you could possible tell us if your CAPEX is more geared towards Memories or the Logic parts.

And I have a second question, related to Hard-Drives.

I think you said your Hard-Drive business went up sequentially.

It seems, though, that this growth was relatively muted, especially compared with the volume growth that we saw in TPA, I think 33% volume in gross sequential in ST.

So, I was wondering if I was missing something in this perspective.

And the last question would be for Pasquale, related to Motorola, they have announced a spin-off of their semiconductor division, and I would just like to know what you think of that.

Do you think that is missing an opportunity for consolidation in this industry?

Thank you.

Well, about Motorola, it is easy to answer.

Motorola is a very, very major partner of ST Motorola semiconductor sector.

We have a fantastic guarantee association in Crolle, which is working very well to reciprocate with the section together with the field section of the department, and we see this as a continued success story.

As far as Motorola Corporation, it is a good customer, we love them, because they are good purchasers of our products, and a part of that is their strategy, whatever is their strategy is there position.

I don't want to make any comment about what my customer, my partners do.

we are able to do everything in one sheet, which brings a huge cost reduction, and also a price reduction, and not a margin decline, however.

I am trying to say that part of the growth that is happening in the data storage work is also coming, associated with reduced overall semiconductor content because of the technology evolution.

I hope I was clear.

Ok.

You announced in the press release some design wins in notebooks for read/write channel I believe in SOC.

When do you think this will have a meaningful impact on your top line?

This is already taking place this quarter.

I am very pleased with this question, because it is the first time we have a product in the Notebook.

This is not only the Power section, but also the SOC.

We have a complete chip center for this application, and some of our traditional customers decided with us a year ago to try and enter this segment, and their plan is to gain, as they mentioned themselves, so I can repeat, to gain at least 12-15% market share in this segment.

Therefore, we will see some advantage in the next quarters.

Ok, and just a last question and I'll let you go.

Looking at the installation rate, I'm not talking utilization but installation rate, given your CAPEX, where do you think Rousset and Singapore, clearly these two, Rousett and Singapore will be wrapped up at the end of 2004?

Given the saturation rate?

No, at the moment I think Singapore is only 40% installed or 30% installed.

Do you think at the end of 2004 you will have it 100% installed?

Maybe not 100%, but above 70% for sure in Singapore, and we should be as close as possible to 100% in Rousset.

Ok great.

Thanks very much.

Ok.

Your next question is from William Convoy of Sanders Morris.

Good afternoon gentlemen.

A follow up, Pasquale to something that was discussed earlier.

You mentioned you anticipate the gross margin increasing in '04.

Can you speak a little bit to how much or what the pricing assumption is in that?

Not exactly obviously, but sort of just more conceptually.

Are you banking on a pretty good improvement in the pricing environment, or is it really driven more by utilization?

Can you just discuss that a little but more?

Of course we welcome a price increase if it enters the market, but we do not plan on that.

All our assumptions are, essentially, based on the fact that we will improve our mix through new products and technology migration and our manufacturing efficiency for better saturation of leading-edge appliances and better efficiency of the 6" plants.

Our back end is very, very efficient, but we keep going ahead with our leading edge manufacturing in back.

So, we expect to see significant efficiency improvements in our manufacturing machine, and significant improvement in our product mix.

Then, if you add to this just a better pricing environment, I'm not saying price increases, with a stop in decline at levels they've been in the past, this will be a major contributor.

We really- I do not count on - and I don't think it will happen - a situation with prices increasing significantly, as they did in the 1999-2000 period.

There is nothing to lead to a situation of that kind.

We do not count on that.

We count on an environment where pricing will back declining more or less according to the traditional learning curve.

That is helpful.

If I could follow up and change gears a little bit.

Earlier I thought you mentioned that in the current year the bulk of your wafer reduction was at the 0.15-micron node -

Flash.

That was all for Flash?

Yes, I was referring to Flash.

That clarifies it.

Thank you very much.

The next question is from Jean Ardenne-Mignon of DRKW.

Just a couple of questions from me.

If I look at your third quarter versus second quarter growth of 6%, but I take away the MPG group from that altogether, just looking at the company excluding MPG, the growth of the company was only about 2.6% quarter-on-quarter.

So the growth was largely led by the Memory products group.

I was just wondering, in your Q4 guidance of 6-12%, and given that you have said that Flash will continue to be a strong growth driver into the fourth quarter as well, how much of that growth do you think will be from the rest of the company, and how much will be pulled along by Flash?

Maybe you cannot give a number to that answer, but just some kind of qualitative indication.

We expect that Flash will continue to be a major contributor, but I do not think it will be representing such a bulk differential.

Flash is going to grow faster.

Say, probably, I look at the number more precisely.

Let me give you precisely what I can.

We expect that Discrete and Standard products will keep growing in the range of 3-4% sequentially, which was what they did roughly in third quarter.

Then we think that the differentiated products overall should be growing again another 4-5%, and the rest will be the Memory products.

OK, thank you.

Just a question on your CAPEX, two questions actually.

One is the additional $200 million that you were going to spend this year, where exactly would you be spending that?

I mean, one of the additional-

Excuse me, one specification I would try to say.

I tried to indicate what the mix was that led to the Q2 Growth, right?

Q3 growth over Q2.

I think, since we expect, in Q4, an overall growth better than the Q3 growth, what I am trying to reflect is that the same proportion we expect for Q4.

In other words, we see Q4 over Q3 being lower for Discrete and Standard, a little bit higher for Differentiated, and higher for Flash, in this proportion.

What exactly are you doing with the additional $200 million this year?

Where is that being spent?

Does the $1.6 billion targeted for next year include any front-end equipment moving into the Catania fab, and if so, roughly how much?

And my last question is, what are your expectations around utilization in Q4?

Ok, capacity utilization in Q4 will be better than the capacity utilization in Q3, which was about 78% overall.

In Q4 it should be well above 80%, which is something we are much more comfortable with.

As far as CAPEX next year is concerned, it does not include production equipment for the Catania 300 millimeters, but facilities equipment for the (300 millimeters.

The bulk of the capital is going to be in Crolles, Rousset, Artoute and Singapore.

And the $200 million this year?

Where will that be going to largely?

To those fabs in Crolles, Artoute, Rousset and Singapore.

Thank you very much.

Your next question comes from Tory Sandberg of US Bancorp Piper Jaffray.

This is actually Jeremy Klein calling for Tory.

Pasquale, you mentioned earlier that you are entering backlog position for Q4 was stronger than Q3.

Could you talk a little about what your lead times are right now, and where you think they will be going in the next quarter?

In Flash?

In some areas, we start having time concerns, such as Flash, but we are responding adequately still.

Overall, our ability to respond is pretty good, and in the Differentiated products there is no problem, because our medium to long-term reclining capability we have with our strategic partners allows us to have frame contracts with flexibility built in, in order to be ready when the customer is ready.

This represents the bulk of our products, as you know.

Something like two-thirds of our products are of this nature.

With Standard products, of course, there is much more general purpose, so we see lead times stretching, particularly with Flash.

In Discrete and Standard there are some Discrete products, but overall we have responded pretty well.

We do not see, let's say limitations for us to respond in Q4.

Could you talk about your returns requirement for Q4?

Could you give us some guidance?

The way the order intake shows up is a little different than in the past.

There is more of the same kind of backlog.

A lot is what is called a freight, which is an agreement with customers to be ready to supply, and then the customer calls you with their needs from the freight.

This gives the customer a lot of flexibility, and a relative uncertainty to the supplier.

What I can say in relative terms is that the portion of our Q4 guidance, which is solid in the books or is in the same is higher in the books than what was in proportion.

We need less percentage of returned business in Q4 than we need in Q3.

Ok.

In your gross margin guidance of 36-37%, you mentioned that this based on a constant currency assumption.

Can you quantify what type of impact a 1% change in currency exchange rates would have on your gross margins?

This needs the accurate calculation of our CFO.

Yes,Lynn is also recalling the fact that the reference point mentioned before is that it is the exchange rate at the entry to the quarter which was somehow above 1.16.

Today there is a margin of level estimation in respect to this forecast. 40% of our manufacturing cost, typically, is denominated either in dollars or in other currencies, such as the Euro, which are volatile in respect of the US dollar, and as a result of that the sensitivity on gross margin for one point of exchange rate volatility is quite marginally in the range between zero and half a point for each pin point.

In gross margin we are at the level of volatility between a few percentage points of change, and we are not significantly affected.

Of course, we have been very positive this year after 20% of volatility of the exchange rate due to devaluation of the dollar.

Is expense is opposed, as you know, and this is also one of the reasons why the guidance for expenses provided before is suggesting an increase in expense.

Let me make one comment.

Currency fluctuation is becoming more and more a way of life.

Rather than anticipating what they are going to be, we are working on being indifferent to currency change as far as possible.

Therefore, what we have been doing continuously, and are now accelerating, is to balance our worldwide sourcing in terms of materials, supply and manufacturing capability; wrapping up our Singapore 8" plant faster, and restructuring plant of the 6", will give us more balance in the origin of our manufacturing.

Increasing our indirect account, in terms of marketing guarantee people, as we are doing in Asia, for example, with several design centers, application and other technical marketing activities in China, in India, and in Singapore, will, again, contribute to those balances.

The company has to react, and the more we grow, the less we will depend on the dollar-Euro exchange rate.

Thank you very much for your time.

Your next question is from Stuart Adrian of Morgan Stanley.

Could you give a little bit more granularity on the CMG growth in the quarter, and could you talk more about SEMOS census?

In the previous quarter you talked a little about DVD pricing being pretty tough.

Could you give us an update on that as well?

To comment on the CMG performance, first of all I would like you to note that this is the eighth consecutive quarterly increase in revenue for CMG.

This quarter the performance was mainly driven by the set-top box sequential increase.

The backlog that we have in place today, and able to think that we can support another record quarter for this market.

Looking at the DVD side, it is lower than it used to be.

We believe that we are now ready to bring solutions to the market place next year, and to come back and grow through a new solution on the DVD encoding recorder.

On the SEMOS census, you ask, we continue to perform well, as we have new models coming into the market, and we believe that our performance in 2003 H2 will be significantly above that of H1.

I think you have talked historically of being at a run rate of around $100 million by year-end in that business.

Is that still on track?

Yes, we can confirm this $100 million for the year.

In terms of the rationalization process that is going to go on in your 6" facilities, how much of ST's total capacity is that?

Is that going to impact in any way the way in which you use foundries going forward?

Is that all going to get absorbed by your 8" facilities, or do you intend to outsource some of this production?

No, the 6" restructure will have zero impact on the foundries.

Until now we have used foundries only for 8" technologies in the 0.25, 0.35 and 0.18 microns, and, therefore, there is no relation to the 6" - or perhaps a marginal relation, and it will continue to be marginal relation on the 6".

The 6" restructuring will imply that some of this capacity will be integrated and moved into 8" ST plants.

For example, if you do a 0.5 micron-ensue you can move toward 0.25 microns, and then make it in an 8" plant.

This gives you the possibility to have many more dies out of the 8" plant, and a productivity increase.

If you have a mature 1.2-micron technology, oran older bill, you just move the production to Singapore.

Capacity on those products will increase.

We will generate more good dies and not fewer good dies.

In other words, the restructuring is a pure productivity and cost mechanism, not capacity.

We are not shutting capacity off, but migrating 6" wafer either to final 8" geometries, or to our 6" capability in Singapore, which we are expanding.

At the end of the process, we will have more good dies out of those products, not less, but done in a more cost-effective way, and that will increase the manufacturing margins of those families.

Finally, on MPG, have you said that you expect MPG to be profitable in Q4?

MPG will be profitable in Q4.

Thank you very much.

The next question is from Giuseppe Paglisi of Intermonte Securities.

Good afternoon.

Firstly, in your release you say that the backlog for Q4 is solid.

What about Q1 2004?

Is it already strong?

Secondly, what about the utilization rate in 2004?

When are you going to reach 90%?

Never.

We hope never, because 90% is dangerous for our service to customers.

It depends how people count.

For us the definition of capacity utilization is based on what we call standard line capacity.

The standard line capacity is the ideal capacity you have if all your equipment performs to its specifications, which is practically impossible with the variety of product mix that we have.

We are very happy with a capacity utilization of 85%, which is our ideal.

Our preferred range is 80-90%.

If pushed towards 90% we feel it is risky for customer service, and if it goes below 80%, it is risky for the bottom line.

Ideally, we aim at 85% utilization, and I believe that, with our expectations in H2 next year, we should be in that range.

What about the backlog for January?

We do not give this guidance, because the mood of the industry is not yet to long term commitment, so we are grateful to say that entering Q4 our position is more solid than it was when entering Q3, but Q1 we will see later on.

Thank you.

Your next question is from Marisa Baldo of SG Equity.

Firstly, on restructuring, correct me if I am wrong, but I thought that in Catania you had 6" lines as well.

Can I have the reason why these lines have not been touched?

Is it the same reason as for the Tourres facility?

The Discrete product family has not been involved at the moment, because those plants are capable of giving good results at the moment.

So the SG is not dead in performance, both in terms of profitability and, even more important, in terms of Roll-out They are a very good Roll-out and very good profit margins.

We have no need to address these facilities.

They are cost-effective, the revenue is good, and the profit is good, so they are not involved in this restructuring.

Thank you.

Now a housekeeping question.

On the other items in the P&L, it has been level in Q3.

Can we have a composition of this negative item, and can we have guidance for Q4, and, ideally, for 2004?

In addition to that we report in this line values items of internet expenses, NT funding, exchange rate, cost for patent litigation, so on and so forth.

Basically out of the 2003 -

Marisa, can you pick up your handset?

Yes, that is done.

Is that better?

Much better, thank you.

I was saying that the 2003 Q3 9.4 net expense includes 17 million.

The other is the miscellaneous items, and the reason why they are higher than the previous quarter is expense is increasing, and also having less income from the other side.

For the next quarter we expect a loss in the range of -5 or -9, -10 for this quarter.

As I said before, at the beginning of Q1 2004 the start-up costs of Crolles should discontinue, or discontinue during the quarter.

Based on that, for the year 2004 we do not expect a negative net result of other income expenses.

It could be either break-even, or slightly positive.

Thank you very much.

Could I have the same type of comments for tax rates?

What was the normalized tax rate for Q3?

What I have is a 20% expectation for next year, but usually you tend to come out with a lower tax rate, so can you give us an update on that?

Sure.

I understand the question, given the fact that the tax rate for this quarter is not easy to understand.

I appreciate the question, given the fact that we feel the tax rate is a competitive advantage versus competitors, since on an operating basis we enjoy a tax rate between 17-18%.

This quarter's results show the tax benefits, which is a combination of operating income at a tax rate on a normalized yearly basis performed at 17%.

The reason for this improvement is that the Euro-dollar impact is affecting profit in those European jurisdictions where the tax rate is higher.

For next quarter we expect a tax rate, again, of between 16-17% when talking about operating.

The restructuring charge happens to occur in jurisdictions with very high tax rates, and provide a tax benefit, which is in the range of 30-35%.

I guess this justifies the benefit on our P&L of the restructuring charge as far as our tax is concerned.

For 2004 is it possible to have an idea, because you were talking about 16-17% for Q4, which is lower than the preceding quarter?

Is that something that can be kept on next year?

There are always these figures when entering the year.

I will confirm as and when entering this year that a tax rate of 16-20% is reasonable for us.

Thank you very much.

May I interrupt for a moment, please, and suggest that maybe we have time for one or two more questions, really, at the most?

Your next question is from Ujay Orgay of JP Morgan.

Firstly, on inventory, can you give us an idea where inventory will be by the end of Q4, given that your guidance to reduce inventory in Q3 did not quite happen, because you have seen a very strong order surge.

What do we expect to see for absolute inventory by the end of Q4?

What we consider to be more important is inventory to orders, because you supply inventory to the level of expectable business level.

Our target turns for Q4 are 4.5 to 5.

We expect our inventory turns at the end of the quarter to be in this range.

They were 4.25 at the end of Q3, and in general the company wants to be a minimum of 4 and a maximum of 5 because, with our variety of products and product mix, more than 5 is not convenience in respect of service to our customers.

We have targeted to be anything between 4.5 and 5.

Secondly, of the $120 million cash charge in restructuring, can you help us categorize that?

How much will be spent on compensation and how much on the other major categories?

A large portion of that is for termination benefits, but there is also, of course, a very important amount related to transferring processes and equipment, and other closure-related costs that we will recognize as incurred quarter-by-quarter as incurred.

Of the remaining restructuring charges, can you tell us how much of that will be saved over the next 18 months?

Basically, it is talking about the charge before tax, so the $350 million and $190 million have already been incurred this quarter.

For the next quarter we expect no more than $10 million, and the remaining balance of $150 million could be roughly estimated to be two-thirds in fiscal 2004 and one-third in the first part of 2005.

So it will be equally spread over fiscal year2004?

Yes and no.

Each quarter may fall in a range of -20 to -30.

On TPA, if you exclude the inventory reduction in the quarter, what would the margin have been?

Thanks a lot for that question, because I was expecting a question concerning the 1.5 decline on total profits in the quarter compared to Q2.

I effect if you compare the manufacturing margin, which is similar to the gross profit, there is no raise or decline.

If you look at the R&D cost, there is neither an increase nor a decrease.

If you look at allocated costs, there is no variation in marketing or general expenses.

This small decline in profitability is, therefore, coming only from inventory reduction performed during this quarter.

That is the only difference.

This is one shot.

In other words, to be very precise, the profit margin would have been the same as it was in Q2 without the inventory charge.

Also, while we are on this subject, if you compare it with last year, you will see that the profitability last year was around 18-19%, and this year it is one or two points less.

In spite of the huge price pressure that we had also in TPA, we have been able to compensate this with reductions in manufacturing costs.

Therefore the manufacturing margin from TPA is still the same.

Compared to last year there is only a difference in R&D efforts, so TPA is now engaged in bigger R&D efforts, that will, hopefully, pay next year in terms of growth.

What we are trying to do now is put more resources towards sustaining future growth.

Operator, I would like to intervene here and say that, unfortunately, we do not have time to take any more questions, but we would like to ask Mr. Pistorio to make his closing remarks.

Well, I would like to start off by saying that the signs we see indicate that recovery is accelerating in pace.

ST has been able to go through the very challenging past two and a half years, performing very well compared to the market.

We always maintained profitability in those years.

We have improved our balance sheet, and we have accelerated investment in R&D through direct R&D investment alliances, moving to platform generations, strengthening our portfolio, and preparing the manufacturing machine.

In entering what we consider to be a more significant recovery pace, the company is really positioned in terms of portfolio, customer base, and manufacturing machine, to perform better than industry average, as we have done in the past, and we are committed to do so with gross margin and operating profit being our priority.

Thank you for being with us this evening.

Good evening and good morning, everybody.

Thank you, operator.

This concludes today's conference call.

You may now disconnect.