英特爾 (INTC) 2003 Q3 法說會逐字稿

Good day, everyone, and welcome to the Intel Corporation's third quarter earnings conference call.

Today's call is being recorded, and at this time I'd like to turn the call over to the Assistant Treasurer and Director of Investor Relations, Doug Lusk.

Please go ahead.

Thank you and welcome to the Intel Corporation third quarter earnings conference call.

Attending from Intel are our CFO, Andy Bryant, and President and COO, Paul Otellini.

I'd like to remind everyone that the earnings release and this call are available on our IR web site at intc.com.

For those of you who did not see the earnings release, revenue in the third quarter was $7.8 billion and earnings per share were 25 cents.

The third quarter earnings report discusses Intel's business outlook and contains forward-looking statements.

These particular forward-looking statements and all other statements that may be made on this call that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially.

Please refer to our press release for more information on the risk factors that could cause actual results to differ.

The specific forward-looking statements cover expectations for product mix and demand, revenue, gross margin, expenses, tax rate, interest and other income, capital spending, depreciation and amortization of acquisition-related intangibles and costs.

These statements do not reflect the potential impact of any mergers, acquisitions, divestitures or other business combinations that may be completed after October 13, 2003.

Lastly, if during this call we use any non-GAAP financial measure as defined by the SEC in Reg G, you will find on our web site, intc.com, the required reconciliation to the most directly comparable GAAP financial measure.

And with that, I will now turn it over to Andy Bryant.

Andy?

Thanks, Doug.

The third quarter turned out to be among the better quarters we have seen.

Sales of microprocessors and chipsets in emerging markets of both Asia and Europe and continued strength in Japan led the company to double-digit growth in revenue and triple digit growth in operating profits compared to the third quarter of 2002.

The results for 2003 are expected to show substantial progress over 2002.

First, I will comment on the quarter just ended and then look at the rest of the year.

Revenue of $7.8 billion was slightly above our September outlook and a billion dollars above the second quarter.

This 15% increase is the best sequential growth for a third quarter we've seen in over 25 years.

The growth from the third quarter of 2002 was 20%, the best third quarter year-to-year growth we have seen since 1996.

Unit shipments of Intel architecture microprocessors, chipsets and motherboards each set a record.

Flash memory unit shipments were slightly higher and Ethernet connectivity products shipments were higher.

Growth in the Intel architecture business was exceptional.

Revenue of $6.8 billion increased 17% sequentially and 26% year-over-year.

The networking business also grew.

Its revenue of $544 million was up 7% from the second quarter and 13% from the third quarter of 2002.

Revenue for the wireless communications business, comprised mostly of flash memory, was down slightly from the second quarter and down 23% from the third quarter a year ago as we continue to see weakness through the first part of the quarter.

Gross margin improved substantially.

The growth in gross margin dollars exceeded the growth of revenue, with a sequential increase of $1.1 billion and a year-to-year increase of $1.4 billion.

Gross margin percentage of 58.2% is up 7.3 points from the second quarter and 9.4 points from the third quarter of 2002.

The key drivers of the growth were anticipated in July: higher revenue, lower processor unit costs and lower start-up costs.

The strong improvement in gross margin coupled with discipline on spending is reflected in operating income of $2.3 billion which translates into a sequential increase of 81% and a year-over-year increase of 139%.

Operating income was 29% of sales, about 10 points higher than the second quarter and 15 points higher than a year ago.

In the Intel architecture business, operating profit was $2.9 billion, representing sequential growth of 58% and year-to-year growth of 106%.

The networking business reduces operating loss to $94 million, which is 2/3 the size of its operating loss in the second quarter and a little more than half the loss a year ago.

In the wireless business, the loss of $124 million was flat with the previous quarter and over four times the size of the loss a year ago.

Spending on R&D and SG&A was $2.2 billion and consistent with our outlook.

As a percentage of revenues, spending was down 3 points from the second quarter and 4.5 points from the third quarter of last year.

The size of the workforce at the end of the quarter was 71.-- 79.1 thousand, approximately flat with the second quarter and down 3% from the comparable number a year ago.

The total for interest income, other income and gains and losses on equity investments was a loss of $29 million.

Within this category of the income statement, interest and other income was $34 million.

The net loss on equity investments was $63 million, primarily as a result of impairment charges of $92 million.

The tax rate for the quarter was 27%, down from 29.5% in the second quarter.

There were several reasons for the change.

You may recall that in July we anticipated a tax benefit related to a divestiture.

We recognized the benefit of $125 million after the divestiture of the wireless LAN business due to the sale with the stock of Zircom.

The positive impact was a little larger than anticipated.

It was partially offset by higher taxes, reflecting our expectations of more profits for the year with a larger percentage of these from higher tax jurisdictions.

Forward diluted earnings per share, which includes potential dilution attributable to employee stock options, was 25 cents.

Basic earnings per share, which does not include potential dilution, was also 25 cents.

For the quarter just ended, average shares for calculating diluted earnings per share was 6.6 billion.

We did not repurchase shares during the quarter.

On the balance sheet, inventories overall were slightly higher.

Finished goods closed the quarter up 19%.

We have put some additional supply in place in both our own inventories and in the distribution channel to support anticipated increases in fourth quarter revenue.

Total receivables increased by $440 million, consistent with higher revenue.

Days outstanding were 37, up from 36 in the second quarter.

Cash, short-term investments and fixed income trading assets ended the quarter at a total of $15.2 billion after capital spending of $1.1 billion and an investment in Micron of $450 million.

In summary, then, for the third third quarter, revenue of $7.8 billion, up 13% from the second quarter and 20% from a year ago.

Gross margin of 58.2%, over 7 points better than the second quarter and over 9 points better than a year ago.

Operating income of $2.3 billion or 29% of revenue.

Spending on target and SG&A and number of employees down from a year ago.

Forward diluted and basic earnings per share of 25 cents, incorporating a lower than expected tax rate.

As we turn now to the outlook for the fourth quarter, please keep in mind that the forecast data do not include the effect of any new acquisitions or divestitures that may be completed after October 13th.

I will use the midpoint of forecast ranges when making comparisons to prior periods.

We expect revenue in the fourth quarter to be between $8.1 and $8.7 billion. $8.4 billion would be an increase from the third quarter of 7%, consistent with seasonal patterns off an exceptionally strong base in the third quarter.

In the previous five years, revenue growth in the fourth quarter from the third has ranged from flat to up 13% and averaged growth of a little over 8%. $8.4 billion would also provide year-on-year fourth quarter growth of 17%.

This is the highest we have seen since 1998 and is a little more than the 10-year average for the fourth quarter.

Our forecast for gross margin percentage in the fourth quarter is 60% plus or minus a couple of points.

This forecast is higher than the third quarter due to expectations for higher revenue and factory utilization.

Spending for R&D and SG&A should be between $2.2 and $2.3 billion.

Fourth quarter spending is usually a little higher as is revenue, producing seasonally higher spending on marketing programs and other revenue dependent expenses.

Gains and losses from equity investments and interest and other income in the fourth quarter are expected to be zero.

The tax rate for the fourth quarter is expected to be approximately 31.5%, higher than our prior expectation of 30.5%.

This reflects expectations of more profits for the year with a larger percentage of these profits coming from higher tax jurisdictions.

Depreciation for the fourth quarter is expected to be approximately $1.2 billion, and amortization of acquisition-related intangibles and costs should be $65 million.

For the full year, there's some slight adjustments to our outlook for R&D and capital spending.

We expect R&D spending to be approximately $4.3 billion for 2003, compared to our previous outlook of $4.2 billion.

This is the result of a combination of small increases, including some profit-related expenses.

For capital spending, we have revised our forecast to between $3.6 and $3.7 billion.

This compares to the previous target of $3.7 billion plus or minus $200 million.

Our goal is to continue to drive efficiency in capital spending.

The results for all of 2003 are expected to show substantial improvement in revenue and profits over 2002.

The nine-month results, combined with our fourth quarter outlook, point to a year with revenue of approximately $30 billion, gross margin percentage around 55 to 56%, and gross margin dollars between roughly $16 and $17 billion.

We expect to achieve double-digit revenue growth for the year of 10 to 12%.

Gross margin percentage should be higher by 5 to 6 points.

As business improves, Intel will be leaving the downturn in better financial shape than when it entered.

Spending and head count are down.

They have been held flat as revenues have picked up.

Inventories are disciplined and credit controls are outstanding.

Profits are growing, cash has increased and shares outstanding have shrunk.

We have redeployed resources to higher areas of productivity and strategic importance.

Aggressive investments are delivering better products to our customers, while improving productivity and creating value for our stockholders.

Over the last several years, we have spoken many times of our efforts to attack non-essential spending, protect profits, preserve cash and proceed with investments that make us more competitive.

This quarter's results are a solid indicator of our progress during the downturn.

With that, let me turn it over to Paul for additional comments on the business.

Thanks, Andy.

Intel had an excellent growth quarter driven by strength in the Intel architecture business across all geographies, channels and product lines.

Emerging markets continued to be strong worldwide, but we also saw good demand in existing markets as well.

We set a number of new records in the quarter, most notably, microprocessor unit shipments, which beat the previous high set in the fourth quarter of last year.

On a geographic basis, our Asia Pacific region set another revenue record and surpassed $3 billion for the first time.

Revenue was up 18% sequentially and 31% year-over-year driven by continued growth in emerging market consumption along with worldwide growth in motherboard demand.

China, Taiwan, India and Australia all established new revenue records.

Japan continued its recent recovery and is up 73% year-over-year, driven by notebook growth both for internal consumption and exports along with higher demand for camera phones.

Our Americas region grew 11% sequentially, primarily due to strength in the consumer segment as a result of a solid back-to-school selling season.

Local revenue for Latin America was also a bright spot with strong sequential growth.

EMEA sequential growth was 19% in Q3, making it our fastest-growing geographic region.

While the emerging markets of Russia, the Middle East, Turkey and South Africa continue to grow at a very fast rate, we also saw revenue growth in Western Europe, driven by an improvement in corporate PC purchases and strong growth in notebook sales across both business and consumer segments.

Our channel business set records in both CPU revenue and unit volume.

Our distributor channel continues to represent approximately 1/3 of our total CPU unit shipments and has delivered significant growth beyond the desktop as a result of our mobile and server channel programs.

The Intel architecture business revenue was up 17% sequentially and 26% year-over-year.

We had record processor unit shipments in desktop, mobile and server as well as in chipsets and motherboards, all of which grew by double digits sequentially and year-over-year.

We believe we gained a point of market segment share in microprocessors and several points in chipsets.

ASPs grew slightly, driven primarily by the growth in mobile and server units.

On the desktop, we saw a good mix within the Pentium 4 processor family, led by continued adoption of Intel's hyper-threading technology.

We remain on track to our goal of exiting the year with over 50% of our performance desktop shipments being enabled with hyper-threading.

At IDF, we announced the Pentium 4 Processor Extreme Edition, a high performance desktop processor, targeted at gaming enthusiasts and other power users which features hyper-threading technology, 3.2 gigahertz clock speeds and 2 megabytes of level 3 cache.

We also increased the performance of our Celeron line to 2.7 gigahertz and likely gained share in the value segment.

Our mobile business had record units with growth in nearly -- of nearly 30% versus the third quarter of 2002.

For the first nine months of this year, mobile units have grown at nearly twice the rate of the desktop thanks in large part to the continued success of Centrino Mobile Technology.

Pentium M shipments nearly doubled sequentially, and we expect it will represent over 1/2 of our mobile shipments in the fourth quarter.

We recently launched a new advertising campaign for Centrino Mobile Technology, promoting the wireless computing lifestyle among consumers as well as business users.

Our 80211 unit shipments also grew sharply versus the previous quarter.

The attached rate continued to exceed our expectations and is well above 50%.

We plan on shipping both dual-band 80211 AB and BG silicon for revenue in the fourth quarter.

Our server business set unit and revenue records as the benefits of Intel architecture continue to make inroads against the installed propriety risk base.

The Xeon family grew by more than 20% sequentially, driven by strong demand for dual processor and multiprocessor products.

We also extended the speed of the dual processor one meg cache version of the Xeon processor to 3.2 gigahertz.

The Itanium processor family's performance and price performance leadership continues to produce results for us.

The Itanium 2 processor had the best volume quarter ever, driven by industry-leading performance along with the maturity of operating systems and business applications that have been optimized for the Itanium family.

We also expanded the Itanium 2 family with our Deerfield processors for lower cost, lower power DP systems.

During the quarter, we announced an agreement with China's Ministry of Education to develop an Itanium 2 based computing grid, designed to span 100 of China's leading universities and deliver performance of more than 15 [INAUDIBLE].

Our chipset business established unit and revenue records, and we continue to gain market segment share.

We also improved product mix as we continued to see excellent adoption of our latest desktop and mobile chipsets.

Even with strong chipset sales in Q3, we see no signs of inventory buildup in the channel.

Looking forward, our plans to ramp 90 nanometer and 300 millimeter technologies are proceeding, and we expect to ship for revenue desktop and mobile processors built on this process this quarter.

Our existing chipset families provide customers with high volume launch vehicles for these new processors, which can be slotted into existing designs with minimal changes.

As a result, we will be in position to ramp the capacity and capabilities that our 90 nanometer technology will deliver across most of our premium and value processor lines beginning in Q1 and quickly accelerating throughout the year.

Our strategy is designed to further strengthen our competitive position and allow us to take advantage of increasing levels of demand for our products.

I will provide more details behind this transition at our upcoming analyst meeting in November.

Moving to our communications-related businesses, revenue for our wireless group declined slightly primarily due to lower flash ASPs.

Flash units were up slightly as we saw flash demand improve in September.

Densities continue to climb, led by our StrataFlash products, which now represent over 50% of our bit shipments.

Shipments of our .13 micron version for cell phones, which began shipment in Q2, increased substantially in Q3.

Earlier today, we announced a new StrataFlash memory for cell phones that combines code execution flash, ram memory and up to a gigabyte of storage in a single package.

Our Xscale processors set a new revenue record in Q3, and at IBF we discussed our new generation Xscale processor which will have a number of new technologies to improve multimedia performance, battery life and high quality digital images.

Revenue in our communications group rose 7% sequentially, led by higher demand for wireless and wired Ethernet products.

Demand in telecom related products remain soft.

During the quarter, we detailed our plans for extending PCI Express, the next generation interconnect technology for computer I/O into our modular product lines.

We also announced a new line of telecom boards based upon Intel CPUs and network processors.

These products conform to the advanced TCA specification, which is increasingly being adopted as a standards-based alternative to propriety telecommunications equipment designs.

In summary, our computing businesses continue to improve in emerging markets, Western Europe, Japan and in the U.S. consumer channels.

However, even with record unit volumes, we have seen only modest growth in PCs in the U.S. corporate market segments.

We are expanding our growth beyond our core microprocessor business with records in many of our computing-related businesses like motherboards, chipsets and WiFi silicon.

Lastly, our investment strategy is delivering growth Virtually all of our revenue growth is coming from our new products, and our capacity strategy is in place to profit from the continued growth in demand.

With that, let me turn the meeting back over to Doug.

Okay.

Thanks, Paul.

We will now open the conference call for Q&A.

We will attempt to take questions from as many participants as possible.

To help in this process, we ask that you please limit yourselves to only one question and no more than one brief follow up.

Thank you.

Operator?

Thank you.

If you'd like to ask a question today, using your telephone keypad, press the star key followed by the digit 1.

If you're using your mute function, please release that prior to signaling.

Again, that's star 1 for your questions.

We'll hear first from John Lau at RBC Capital Markets.

Thanks.

Banc of America, actually.

You mentioned the strengths specifically in the emerging markets.

I was wondering if you could give us more color on these markets?

For example, in China, is the CPU sales continuing to be more P-4 oriented or is the Celeron growing fast in those markets?

And within the China market, because there's not a lot of visibility, what percentage of the China markets are slated toward notebooks?

Thank you.

I don't know if I can give you a lot of detail on that, John.

In general, in China and in the emerging markets of Europe, the processor mix is about as it is in the United States and Western Europe.

That is the mix of premium to value products is about the same.

In Latin America where the economies are in a bit more trouble, there is a slightly higher mix toward Celeron, but not terribly so.

In terms of notebooks, we are seeing notebook growth in China, but it is from a relatively small base.

The business is there.

The microprocessor businesses there tend to be desktop and server first and then notebooks growing.

Great.

And -- and then just a follow-up, with regards to the notebook market, as you exit Q4 in 2003 this year, what percentage do you believe will be the Centrino based as opposed to P 4 M?

Thank you.

Well, the bulk of our premium processors in notebooks, actually in the year, will be based on the Pentium M processor.

The vast bulk of those are Centrino.

Would that be in excess of 50, 70%?

What's the --

When I say the bulk, I'm saying about 50.

In both cases.

Okay, thank you.

Our next question comes from Dan Niles at Lehman Brothers.

Thank you and another nice quarter, guys.

I guess for either Paul or Andy, your guidance for the fourth quarter, which I guess at the midpoint is up about 7%.

Your ASPs have been doing well I think for the last four quarters or so, you know, they've been either flat or going up sequentially.

You've got another nice mix it sounds like into portables, moving to more of the Centrino offering in the Q4.

Can you give some comments or thoughts around ASPs and maybe touching with that, do you think, you know, your portable growth will continue to be faster than, you know, your overall desktop growth given some of your comments, for example, around U.S. corporate segment demand on desk tops?

Sure.

Sure, I'll talk.

I won't tell you much.

I had to try.

You know, we don't make ASP forecasts.

You know, but we have had the conscious effort here to try to find features that are valued in the marketplace to help us in this space.

Centrino, hyper-threading.

We've seen improvement in mix with Centrino and mobile.

It's a premium with being Centrino and also a mix with mobile versus desktop.

I think those things continue.

You will continue to see us focus on it.

You will continue to see us recognize its importance.

So, you know, I won't give you a forecast, I will tell you it's foremost in our minds and we will continue to do our best to work in that space.

Okay, and can I have one follow-up, Andy?

Sure.

Can I -- you and I -- we've mentioned on these calls in the past and that sometimes your Q4 guidance will partly depend on what happened in late September versus October, in terms of sometimes you will ship processors right in late September or it will be in early October and that will skew third quarter versus fourth quarter and the mix of the revenues.

I mean have you seen, in terms of what you're looking at for your business, do you see anything that seems to indicate, you know that a lot of stuff that could be shipped in early October got pulled into September, in the sense of inventory or anything like that?

Or does it look clean to you from a seasonal perspective?

I think things are pretty clean, Dan.

We didn't say much in September because we did the, you know, preannouncement a week or two before that.

As you can see, though, we were a little above the top end of our range we gave then.

The business built nicely, you know, through the quarter and -- and you get no senses of inventory build.

You get -- actually, you get the sense that it's almost too good of orderly markets.

You don't get a sense of a lot of double ordering.

You don't get a sense of people buying irrationally from us.

Customers say here's what we need, we're able to apply it, and we think it all through.

So, we see nothing unusual.

Like I said, almost an unusually normal set of behaviors right now.

Okay, great, thank you very much.

Moving on now to Adam Parker at Sanford Bernstein.

Hi, Andy, you said you were going to continue to drive efficiency at capital spending.

What would have to happen for Cap Ex to be less in 2004 than it is here in 2003?

You know I can't get into the 2004 number yet.

I have shown charts that tell you I do expect my capital -- depreciated percent of cost to continue to get better to the 300 millimeter ramp.

You know, what you'd have to expect is to see us continue to -- and in my opinion, for the next several years, to have depreciation as a percent of costs be improving.

And I say depreciation, I mean that's almost directly related to capital, but just from timing differences, so, you have to be a little careful there.

But it's foremost in our minds.

We've -- we spent a lot of money get to a point where we thought we could become more capital efficient and we're not going to give up on it.

Okay, so when are you guys going to provide color on Cap Ex for '04?

We do it in January in the conference call.

Okay, I have a different question, I'm sorry, I wanted to go where I wanted it to.

Do you expect -- maybe this is for Paul, you know, are the -- do you expect any incremental microprocessor price cuts in the next couple of quarters as a result of AMD’s product launch?

Or is it just as if it's not happening?

We are -- I expect no -- I expect no pricing action as a result of their actions.

Okay, great.

Thanks, guys.

We'll move next to Eric Gomberg at Thomas Weisel.

Hi, nice quarter.

You talked in your comments about geographies that you saw signs of a corporate spending pickup in western Europe.

I'm wondering as you look in the U.S. and in Asia, what signs you may be seeing there.

On corporate markets?

Yeah, if -- if you've seen any type of spending pickup --

Yeah, I -- the Asia market is -- is -- currently in Japan, some of the business upside we're seeing year-over-year is corporate.

And in Asia, Asia Pacific in general, it's very hard to look at that because there's no upgrade per se, it's a lot more new installations.

Western Europe, I made a comment on, which, as we've seen a -- a small increase in -- in corporate sales, particularly around notebooks in Europe, in the western parts of Europe.

And in the U.S. it still remains antidotal.

I saw what I would characterize as a modest change in corporate purchases, and even there, there seems to be a move toward notebooks versus desktops.

Okay.

Just one quick follow-up, Chinese New Year comes a little earlier this year.

I wonder if any pull-ins into the month of December are at all contemplated in your guidance?

Not per se.

Okay, great, thanks.

From JP Morgan, this is Chris Danely.

Hey, guys, nice quarter.

I know you can't probably give any back numbers, but can you just talk about sort of where utilization rates are?

And how much more room is left to go?

And what it can do to gross margins versus ASPs and mix, et cetera?

Utilization rates in logic factories are very high.

I'd say most of the factories are producing about the level we think they can.

And the flash factories, you know, we're still end loaded and still have a small, but a little bit of excess capacity charge there.

You know, I can't directly answer your question on what the -- about the margin.

Margins are so complex in terms of percentage.

You have product mix between flash and networking, motherboards, chipsets, processors.

So your mix is going to make a difference even with processors, desktop and server and mobile you will get different reactions.

What I can tell you is, you know, we're ahead of where we're expected to be on cost unit right now.

The full effect of 300 millimeter is out in front of us.

It should be more benefit from that.

So, yeah, I actually believe I can still drive costs down.

The only thing that derails us is if the factories get empty.

As long as demand stays up, we deploy a new technology, we will see cost improvements.

So, utilization rates on processors are basically more or less maxed out?

You hate to say maxed out because, you know, the manufacturing organization here always manages to rise to new levels for you, but yes, they're running pretty full.

On the older factories.

On the older factories.

Sure, can you do anything with flash lagging here to switch over or are you just going to start to, you know, ramp the 300 millimeter in the processors?

Are you really going to ramp 300 millimeters in the processors is the right answer.

The ability to move between -- you have very little bit of ability to move between say some chipsets, say some flash, you know, manufacturing processes, but mostly the answer is we have to ramp the 300 millimeter processors.

Thanks.

We have a question now from Mark Lipacis at Prudential Securities.

Great, hopefully you can give me a clarification.

Andy, I thought in your scripted comments you talked about some inventory build in the channel and then -- and then later I thought you said there wasn't an obvious build and I thought Paul said also the same thing.

Could you just clarify that for me?

Thank you.

Sure.

I -- I communicated ineffectively.

When I say in a channel, I'm talking about our distributors and we count that as our -- it isn't when we talk about the sales channel we're going beyond our distributors.

So, I think if our distributors and our warehouses, our inventory, we put a little bit more in each of those places.

In terms of the people they sell to, which are the real people who take the product and put it into the system, we see no evidence of inventory build in those channels.

Okay.

And then -- and then when you talk about -- and when you talk about inventory builds, are you basically lining that commentary against, you know, I guess I would expect to see a normal inventory build in Q3, or a normal seasonal inventory build in Q3.

So, are you baselining that against -- against that or just on an absolute basis?

Thank you.

Yes, the inventory we put in the third quarter is consistent with the revenue growth we should see in the fourth quarter.

So, you know, it's plan and calculated to make sure that we're comfortable we can meet our customers demands.

It is absolutely nothing unusual there.

Thank you.

Moving on now, we have a question from Mark Edelstone at Morgan Stanley.

Good afternoon, guys.

And congratulations on a truly phenomenal quarter.

Andy, this is the first quarter I can remember in a long time where there was no stock buyback.

Can you give us a little bit of insight there?

Not much, Mark.

You know, we have a long-standing history of not explaining, you know, historically looking back, and prospectively, legally not even allowed to speculate about what we may do.

The answer is no.

You can have another question.

I can't tell you anything about that.

I appreciate that.

The other thing I noticed as well is that you guys have done a great job of managing head count and driving revenues higher, your revenue per employee now is almost back to where it was in 2000, roughly around $400,000 per head.

What's the thought process going forward?

To continue to drive that higher?

And what do you now see as kind of the realistic revenue per head target for the company?

You know, we don't have -- we haven't set a next year revenue per head target, but we have taken an overall goal of minimal growth in terms of head count until we see what I will call the next increment, and you can determine what the next increment is of revenue.

So, I'm doing what I can here to build a plan that's kind of flat to spending, maybe up in a few very small strategic locations, and head count to be relatively consistent with that.

So, I'd like to see another year of kind of staying where we are and letting the business continue to grow.

Great, thanks a lot.

We have a question now from Ben Lynch.

Yeah.

Sorry to sort of focus on a slightly negative area given how good the quarter's been for you, but the WCCG business looks pretty inconsistent with positive data points we've got in cell phone.

A lot of other people have seen strength from late Q2 all through the quarter, you guys say you've seen sort of a pickup only in September.

And if you say units were slightly higher and revs were down 3%, you could say maybe blended ASPs were down 5% Q on Q. Does this mean the mix hasn't improved?

Or like to like megabit ASPs were down quite a bunch?

Or could you maybe explain what's going on in terms of the market and your share?

Are you still suffering from your failed price hikes, et cetera?

I think you explained it pretty well for me.

There is no question we've been struggling now for nine months.

You know, I don't have precise data.

We feel, based what what we've read that we lost share again in the third quarter.

We were encouraged to see, you know, actually unit uptick.

We were encouraged to see a little better news in the back part of the quarter.

You know, we believe that -- and our history has shown -- the fourth quarter tends to be the best quarter for cell phone sales, and we sell into that market.

So, I'd like to think we will see a little bit of positive in the fourth quarter.

But, you know, we're still in the hole we dug ourselves and it's a long way out and we're still working on it.

But I think neither Paul nor Craig or I have slashed the performance and we're going to continue to push on it.

It feels like it's still an area of focus that you have not necessarily turned the marketshare corner there yet?

Absolutely feels that way.

Great, thank you.

Next up from CS First Boston is Michael Masdea.

Great.

Thanks.

Congratulations on the quarter.

Andy, any sense we can get from you in terms of magnitude of the ASP on the MPU side?

Or is it close to being flat to slightly up?

Or is there a material uptick an ASP this quarter?

You know, as Paul said, slightly up is basically it means it's noticeable, but it's not a major driver of margin improvement, but it's a real help.

It's sort of -- I'm not saying anything other than slightly up.

And it if I could real quickly, you guys came out with a way to get out some of the competitive natures of Nan coming into the market with your product you announced today.

Is there any reason we shouldn't think about Intel getting into the Nan market?

Is there any reason we haven't seen that so far?

We have no plans, really, to go there right now.

We're looking at exactly just the technologies to take advantage of where we are strong to deal with the foreign factor and data requirements across the form platform.

So, no near-term plans.

But there's no reason you couldn't if you decide it makes sense, correct?

Technically, no.

Great, thanks.

This is Nimal Vallipuram at DRKW.

Yes, first of all, congratulations on a good quarter.

This is a question for Paul.

You indicated that the third quarter has been one of the best quarters for so many years and you did give some reasons why that was the case, and if you take a step back and if you look at what happened to your business after -- after going through such a long, lean times, is there any other broader reason why the third quarter was so strong?

And I have a follow-up question.

I don't think there's any intrinsic reason.

I mean we've got continued growth in emerging markets.

That has been our story throughout the downturn.

Would be one vector.

You could argue in that that the second quarter may have been slightly depressed from SARS in hindsight, and we saw a bit of a snapback, but that may be hindsight rationalization.

What we saw was the business in Asia was certainly better in Q3 than it was in Q2, and in Q2, we thought we saw no impact from SARS.

The second trend would be a -- a -- a good portfolio of new products.

You know, the Pentium 4 with hyper-threading is selling very well.

I gave you some statistics on that.

And Centrino Mobile Technology has really taken off at or slightly above our expectations.

So, I think it's a technology story, it's a product story in the mature markets and in the emerging markets, it's just a continued boom.

So, just a follow-up on that, I'm not asking you to forecast what the numbers are going to be beyond the fourth quarter, you mentioned in your presentation that you believe this might be -- or to the effect that this might be the beginning of an upturn or something to the effect.

I mean are you seeing -- internally in Intel are you putting that as a very strong second half seasonality?

Or are you seeing that as the beginning of something beyond that?

You know, clearly the third quarter results were more than you could ever call seasonality.

It was -- the strength in the emerging markets was phenomenal and the beginning of some bright spots in Western Europe and even some anecdotal evidence of things starting to happen in the U.S.

So, it's more than just seasonal.

Certainly in the third quarter we're forecasting seasonal fourth quarter off of the new found goodness, but I -- I -- we want you understand, we're not looking at the U.S. and saying my God, the good times are here, there is no evidence of IT budgets increasing across-the-board.

There may be some more laptop buying as Paul talked -- as corporate looked at that.

So, there are signs that are evidence but there's certainly no overwhelming rush to increase IT budgets that we've seen.

All right.

Thanks, Paul, thank you.

Clark Westmont of Salomon Smith Barney has the next question.

Nice job.

I'd like to ask about the gross margins and is there the way that -- that you'd like to manage the business, is there a gross margin limit in terms of how -- if they reach some -- I'll pull out a random number -- 64% or something like that, would you then, with any incremental cost savings or other leverage, would you be more aggressive with pricing to -- to try to press or stimulate elasticity in the market or is there no -- or is that not even entered into your thinking at all?

Just help us understand how high gross margins can go.

All right, now, it's something I haven't -- I thought -- not something I spend a lot of time thinking about, Clark.

As I said a minute ago, gross margin percentage is extremely difficult to forecast because you start on the demand side, forecasting between, you know, flash and networking and then different kinds of processors and motherboards, all of which have different margin profiles.

You can get several points of margin if you just sold -- stop selling some of the product that we think we need to sell and have lower margins.

They add value to the company but not at the same percentage.

So, it becomes difficult and -- and, you know, the cost side, again, the utilization of 300 millimeter, how much we can use that due to the factory stay full, it's -- I wish I could easily answer your question.

So, I don't know what a theoretical limit is.

I do believe we haven't seen the benefits of 300 millimeter yet.

So, if those factories get full -- if you continue to have demand, there is more good news coming from cost.

We tend not to think in terms of -- especially since the first of the year, we think the factories are tight and so we can raise prices, but we tend to give our customers quotes for prices well in advance so they know how to run their businesses.

We don't opportunistically then go back and try to change that.

We try to find features, you know, things we can bring to the marketplace that their customers and then they will value and pay more for.

So, those will affect margins.

I -- I know I'm not giving you the answer.

I really don't know the answer.

If you ask me is this as high as it can go?

Are you worried about that?

No, it's really a question of can demand stay up, can we fill the 300 millimeter factories.

Can we find features they value?

Can we continue to drive the Intel architecture business at high rates?

Those are the things we're trying to do.

Fair enough.

Maybe I could ask a follow-up.

With -- at the May analyst meeting you had shown a slide that had your mix of 300 millimeter production for your CPU production.

Could you update us where you stand today in terms of that mix in the third quarter and how you see that fourth quarter unfolding in next year?

You know, I haven't updated it.

I'd have to go back and get the data.

While, you know, I do know in the 300 millimeter factories we have begun production, so, we started wafers in the third quarter.

We're ramping the wafers in the fourth quarter.

So, I'm getting more production as -- as we expect to sell product off of the 9 nanometer, 300 millimeter product this year.

So, you know, I need to go back and get precise data, but the processors working products are being built.

More wafers are being started every week.

Fair enough, okay, thanks.

There's a question now from Hans Mosesmann at SoundView.

Yes, thanks.

There are rumblings out there that the 90 nanometer node is having issues regarding thermal aspects of the desktop Prescott product that you're going introduce shortly.

Can you comment on that?

And can you indicate if there is an issue with it, is it with active power or leakage?

What could it be?

Well, there are a lot of rumors running around.

I'm not going to dissect the process on this call, but let me certainly start by saying that the process is very healthy.

We ended up changing the thermal target for systems for Prescott as we got silicon out and ran that against our models and against what we think the likely speeds are going to be on that product.

And that required us to change the thermal envelope at the systems level, slightly.

When we do this, we do this looking out over a long period of time so that we aren't faced with a headroom problem halfway through our product life cycle.

So, what you -- when you -- if you add those facts together, you get -- you can take one picture, I think you get a different picture if you look at it from the headroom that it is likely to give us over the life cycle.

Both products are functionally very sound.

We're finishing up the last bit of circuit work to -- to make sure we can enable very high volume manufacturing across both our desktop and -- and mobile product lines from performance to value processors.

We intend to use this technology in the scale that is coming online very aggressively across the bulk of our PC product lines.

Great, thanks a lot.

Next up, Michael McConnell, Pacific Crest.

Back in August you had talked about some supply constraints.

Was just kind of wondering with the fabs running as high as they are now in terms of utilization, if you're seeing any type of allocation or supply constraints at this point?

We're actually in pretty good shape now.

We did get, as I said, sometimes manufacturing gives me more than I expected.

We got some extra parts, you know, we put about $100 million of finished goods inventory into our warehouses, we added, you know, $70 or $80 million of parts which are also finished goods in our distributor's warehouses.

So, I think I'm pretty comfortable right now.

And just on the tax rate going forward for 2004, just for modeling purposes, should we keep it at 31.5%?

I really can't give you an '04 tax rate now.

What I would -- the only help I would give you is in your modeling.

As Paul said, we do plan to use 300 millimeter heavily next year.

The wafers coming out of 300 millimeter are in the U.S. tax jurisdiction, which is our highest tax jurisdiction.

The Irish fab comes on-line next year, which is a low tax jurisdiction, but the bulk of your wafers for next year's revenue will come out of the U.S. which is a higher tax location.

Okay.

Thank you very much.

So, model -- model what you think revenue will be and if you think it's going up, more will be in the U.S. than the lower tax areas.

Okay, thanks.

From UBS, this is Tom Thornhill.

A question related to -- to -- back on the margin subject.

You're close -- getting close again to the margin levels that you achieved in '99 and 2000, we're a few points away, and this is without having the 90 nanometer or 300 millimeter ramp.

The questions are how quickly do you see the transition to 90 nanometer 300 millimeter across '04?

And what -- given the smaller die size and the higher efficiency, is it a reasonable target to get back closer to mid-30s operating margins or margins above 60% as we go through the year?

Yeah, I -- Tom, I can't say much more and I've already not said about that.

It's -- we will have every intention of ramping 300 millimeter hard next year.

We will push to utilize that capacity.

I think it will make a difference in your cost per units next year.

So, you can -- you can pretty much expect costs to improve, and the question becomes what happens with margins and what happens with overall volumes?

Well, the combination of 300 millimeter plus your head count comments would -- would lead -- would leads us to, I think, optimistic conclusion on this.

Certainly if those are the only records, I'd agree with you.

I -- you know, we know we've had a long-term battle to keep ASPs up, so there is still a lot of -- a lot of variables in there that have to be wrestled to the ground before you can look at margin and be comfortable with the forecast.

So, again, we will give you our best guess for next year in January, but the cost side of it, if we -- unless we have a volume collapse, the cost side of it, I think we will continue to derive improvement in the cost of the units, yes.

Thank you.

We have a question from Fulcrum Global Partners and Clark Fuhs.

Yeah, thanks.

Great quarter, guys.

On server chipsets, are you guys -- feel like you're increasing share there?

How did units grow sequentially?

And how much faster did they grow than servers?

We don't -- we don't usually break out the sub segments of the chipset business but I do believe we are gaining market segment share nicely in servers.

We've got a new set of designs that are out there.

There is very high volume designs in a number of manufacturers that are under way, and I made some comments in my channel section about our server program -- server channel program doing very well.

A lot of that is based upon our chipsets and our boards.

Okay.

And can you -- can you kind of walk us through your thought process and logic for the investment into Micron?

What -- you know, one of the things we are -- that we've done is new technology coming to the marketplace.

And what we wanted to do was to, you know, we talked to Micron, looked at getting a DDR-2 out, helping them if we can and if we can do that, that's fine.

There are lots of different paths it could have taken, but the key is we will continually invest in places that bring the technology that complements our technologies in the marketplace.

That's primarily what we're up to is trying to make sure that -- that the good technology is there across all phases of the platform.

Okay.

Thank you.

Operator, we will take two more questions.

All right, the first of those is from Joe Osha at Merrill Lynch.

I made it in under the wire! [ Laughter ] Pretty good stuff.

Two questions.

One, can you talk about -- as you move to 300 millimeter, how you feel you're going to -- obviously you're converting some capacity, but how you feel you're going to absorb what will be a -- a fairly substantial amount of 200 millimeter capacity as the year progresses, what strategies you have for doing that?

And secondly, although it's not your business directly, I'm wondering if you can comment on what flat panel availability or perhaps lack thereof has done to your -- to your own ability to ship microprocessors for revenue here in the fourth quarter?

Sure.

The -- you're right, we're going to move as much of the product line, the 300 millimeters as quickly as possible.

That was my comment about utilizing it across a broad spectrum of products.

As with do that, we have a couple of options on the 200 millimeter factories.

We can certainly and will certainly use some of that volume to continue to grow our chipset business in those factories as our processor volumes scale and new generations of chipsets come on.

And secondly, in at least one of the factories they've already announced plans to retool that into next generation 300 millimeter factory.

So, we have both options to play out.

In terms of flat panel displays, the -- the -- the -- there's no measurable impact, and if you measure it by our notebook shipments it doesn't seem to be retarding that in any way.

And I do understand that a very large percentage of the desk tops are now being shipped with LCD monitors.

In general, we've not seen any complaints filtering back to us with the exception of -- in some cases some retail notebook SKUs which tend to be low end SKUs on older strain technologies, but no yells and screams at this point.

Okay.

Thank you, Paul.

And our final question today is from First Albany's Gus Richard.

Yes, thank you.

Just real quickly can you talk about Manitoba and how that's -- the design wins are going?

And how you see that ramping?

Well, we're moving that product towards full production this quarter.

We -- we've been tracking design wins for a year and a half now, a number of them are scheduled to go into production in the first half of next year.

Albeit in relatively low volumes and I think this is one of these products that if -- because it starts at the high end, starts with low volumes and trickles down.

We are continuing to provide all the other necessities for volume ramp, in terms of reference designs and working to get the software stacks in place to make this as easy and seamless as possible for our customers.

Great.

Thank you.

Okay, I'd like to thank everyone for listening to today's call and we will hold an analyst meeting in New York City on November 20th.

If you're interested in attending, contact us in investor relations.

A recorded playback of this call will be available at approximately 5:00 p.m.

Pacific time tonight.

If you're interested, dial 719-457-0820 and reference passcode 328339.

Thank you and good night.