英特爾 (INTC) 2002 Q4 法說會逐字稿

Good day, everyone, thank you for holding and welcome to the Intel Corporation's fourth quarter earnings conference call.

Today's call is being recorded.

At this time, I'd like to turn the call over to the manager of investment relations, Alex Lanky.

Please go ahead, sir.

Welcome to the Intel fourth quarter earnings conference call.

Attending from Intel are Andy Bryant, Chief Financial Officer, and Paul Otellini, president and Chief Operating Officer.

I'm Alex Lanky from Intel Investor Relations.

I'm filling in for Doug Lusk today.

I'd like to remind everyone that the earnings release and this conference call are available on our Investor Relations website at www.intc.com.

For those of you who may not have seen the quarterly earnings report, revenue in the fourth quarter was $7.2 billion and earnings per share on a GAAP and pro forma basis were 16 cents.

The fourth 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 the 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 include expectations for product mix and demand, revenue, gross margin, expenses, R&D spending, tax rate, gains or losses from equity investments, 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 January 13, 2003.

Finally, I'd like to remind everyone that Intel will be hosting its Intel Developer's Forum February 18-21 in San Jose.

Those interested in registering, please contact Investor Relations.

I now want to introduce Andy Bryant, who will discuss the fourth quarter earnings results.

Thanks, Alex.

Although challenged by slow growth and a difficult economy, 2002 was a year when Intel made significant progress along several fronts.

Business performance generally followed seasonal patterns and ended the year on a high note.

Fourth quarter results were better than we anticipated in our October earnings release and December update.

Despite the economy and what was for many industries a weak holiday season, order patterns held through the end of the quarter, resulting in better than expected demand and boosting both revenues and profits.

Unit shipments were higher in the third quarter in most product categories: microprocessors, chipsets, motherboards, flash memory and connectivity.

Revenue of $7.2 billion was above our forecast range of $6.8 to $7 billion and at the high end of seasonal balance.

We achieved quarterly and annual growth of 10% from the third quarter and 3% from the fourth quarter of 2001.

During the previous five years, sequential revenue growth in the fourth quarter has ranged from flat to up 13% with an average of 7.4%.

We saw good sequential growth in all operating segments.

The largest portion of the dollars came from the Intel architecture business, who's revenues of $5.9 billion were 83% of total revenues and up 10% from the third quarter and 2% from a year ago.

Revenues for the wireless communications group, comprised mostly of flash memory, were $662 million and up substantially.

The 13% increase from the third quarter and a 28% increase from a year ago.

The networking business grew 13% from the third quarter and was up 8% from a year ago.

Revenues were $544 million.

For the year, Intel architecture revenues of $22.3 billion were up 4% over 2001.

Wireless revenues of $2.2 billion were flat while networking was down 19% with revenues of $2.1 billion.

Gross margin of $3.7 billion grew by 16% from the third quarter and 3% from a year ago.

For the year, gross margin dollars of $13.3 billion were up 2%.

Profitability was better than we expected in October, when we anticipated gross margin of 49% plus or minus a couple of points.

Gross margin percentage of 51.6% is an increase of 2.8 points over the third quarter.

Most of the improvement from both the third quarter and our outlook in October came from higher volume and product mix.

Compared to the fourth quarter of 2001, gross margin was approximately flat.

For the full year, margin was up over half a point.

Operating income, excluding acquisition-related costs of $1.6 billion, was an improvement of 46% over the third quarter and approximately flat with last year.

More than 100% of this income came from the Intel architecture business, whose operating profit of $2 billion was 34% of its revenues.

Up 42% from the third quarter and 10% from the fourth quarter of last year.

For the year, Intel architecture achieved operating income of $6.6 billion, an increase of 5% over 2001.

In the wireless group, higher excess capacity and inventory reserves caused larger losses, from the loss of $30 million in the third quarter to one of $98 million in the fourth.

For the year, the wireless group had an operating loss of $294 million.

Intel's networking business lost $168 million compared to $177 million in the third quarter.

For the year, the networking group lost $622 million.

Earnings per share, excluding acquisition-related costs of 16 cents, were up 45% from the third quarter, earnings per share of 11 cents, and 7% from the fourth quarter of 2001, earnings per share of 15 cents.

Spending was on target at $2.1 billion, approximately flat with the third quarter and 5% higher than the fourth quarter of 2001.

Spending for the year was $8.4 billion, flat with 2001.

The workforce totaled 78.7 thousand people at the end of the year, down 6% from 83.4 thousand at the end of 2001.

The total for interest income, other income and gains or losses on equity investments, was a loss of $117 million, a greater loss than the $90 million we expected in October.

Within this category of income statement, interest and other income was $54 million.

Impairment charges were $177 million, higher than the $130 million we expected in October.

Average shares were calculating diluted earnings per share were $6.7 million, 52 million shares in the third quarter.

During the quarter, we repurchased 59 million shares at a cost of $1 billion.

On the balance sheet, inventories increased by $205 million to $2.3 billion, most of which was work in process and finished goods.

Cash, short-term investments and the fixed income portion of the trading assets ended the year at a total of $12.2 billion.

This was an increase during the quarter of 12% or $1.3 billion.

After $1.2 billion for capital spending and $1 billion for stock buybacks.

For the year, cash investments increased by $1.1 billion.

After $4.7 billion for capital spending and a stock buyback of $4 billion.

In summary for the fourth quarter, revenues of $7.2 billion, gross margin of 52%, spending flat and number of employees down, operating income, excluding acquisition-related cost of $1.6 billion, or 22% of revenues, earnings per share of 16 cents.

As we turn now to the outlook for the first quarter, please keep in mind that all forecast data excludes the effect of any new acquisitions that may be completed after January 13.

Let me also remind you of our previously announced plan to discontinue the practice of providing income, excluding acquisition-related costs, and the associated earnings per share figure in our earnings releases, beginning in April.

Amortization for acquisition-related intangibles and costs is expected to be approximately $85 million in the first quarter and $300 million for the full year.

We expect revenue in the first quarter to be between 6.5 and $7 billion.

The mid point of this range would be a decline from the fourth quarter of 2002 of 6%, which would be consistent with seasonal patterns.

Revenue has been down in the first quarter of each of the previous five years, with declines ranging from 3 to 23% and averaging 9%.

The 23% in 2001 was atypical.

If you exclude that year, the average decline would be approximately 5%.

Gross margin percentage in the first quarter should be 50% plus or minus a couple of points.

Changes in the first quarter are most likely to depend on changes in revenue.

Spending should be between 2 and $2.1 billion This compares to $2.1 billion in the fourth quarter.

Depreciation should be $1.2 billion.

Our estimate for gains and losses from equity investments and interest and other income is a net loss of $125 million.

This includes a net loss on equity investments of approximately $170 million, primarily as a result of impairment charges.

For the rest of 2003, we anticipate a year to be driven by the pace of recovery in our industry and in the economy.

While planning for seasonal performance, we have the flexibility to respond quickly to a change in business conditions with the right products and capabilities.

We expect growth margin percentage for 2003 to be approximately 51% plus or minus a few points.

The 51% mid point is one point higher than 2002 gross margin at 50% and 2 points higher than gross margin of 49% in 2001.

Gross margin should benefit from the productive use of .13 micron and a successful ramp of 300 millimeter wafers.

We don't include specific goals for spending for number of employees in the annual outlook, but don't anticipate growth in either category.

We plan to absorb no attrition from our workforce and to move employees from lower to higher productivity areas.

As has been the case throughout Intel's history, we are proceeding with high levels of research and development to advance our technology leadership.

R&D spending in 2003 is projected to be $4 billion, the same level as 2002.

Most of this will go to advanced technologies with future fab processors, future micro processor designs and a variety of communication initiatives.

We're targeting capital spending at $3.7 billion plus or minus $200 million.

This is down from $4.7 billion in 2002.

In 2003, the efficiencies of 300 millimeter mean that Intel will spend approximately $1 billion less than we would if we were to build equivalent [indiscernible] on the 200 millimeter processors.

Decreases in spending and construction and 200 millimeter processors are driving the reduction from 4.7 to $3.7 billion.

That has been the case for the last two years.

Office building projects are essentially on hold.

And 85% of the capital budget is from manufacturing capability.

Roughly half of the entire budget was the FAB manufacturing equipment with the largest portion invested in equipment for 300 millimeter wafer process.

One quarter of the budget will go to land and construction and the remainder is allocated for other equipment.

Intel's FAB investments continue to shift toward technologies with 300 millimeter wafers and processes of 90 nanometer and lower.

Over 90% of our 2003 budget for fab capital spending will be directed at these technologies.

Investment in 90 nanometer and smaller technologies should grow from approximately 50% of capital for fab in 2002 to over 90% this year.

At the same time, investment in 300 millimeter should grow from approximately 75 to 90% of the same total.

The transition to 300 millimeter lowers capital spending as it increases capital efficiency.

Depreciation for 2003 is expected to be $4.9 billion, an increase of 5% over depreciation in 2002.

Our estimated tax rate for 2003 is approximately 30.5%.

This is higher than the 2002 rate of 27.4%.

The primary reason is that as we ramp our 300 millimeter fabs, more of our products are made in the U.S.

As a result of this transition, we expect more of our income to come from higher tax jurisdictions.

I would like to close with a few reflections on 2002.

It is challenging to create shareholder value in a no-growth environment, but challenges bring opportunity.

The challenge and opportunity, when times are slowest, to make sure the company is in the best financial condition possible.

These are a few of the significant financial accomplishments in 2002.

General accepted accounting principals, net income of $3.1 billion. 16 consecutive years of profitability.

Operating income, excluding acquisition-related costs of $5 billion or 18% of revenues.

Another year of progress in containing discretionary spending, which declined in dollars and as a percent of sales.

In absolute dollars, discretionary spending has declined nearly 50% from the peak in 2000.

A 6% reduction in the size of the workforce accomplished without major layoffs, reducing the size of the workforce by 13% since peak employment in March of 2001.

Savings of hundreds of millions of dollars as we continue to reemploy resources with higher productivity and strategic importance.

Our standing credit controls in an environment that presented the challenges of growth in emerging markets and economic adversity in other areas.

The quarterly day sales outstanding range from 34 to 37 days.

Bad debt expense averaged less than $5 million per quarter.

Discipline on inventories, which gained below levels of past downturns with lower reserves and no significant write-offs of microprocessor inventory.

An increase in cash balances of over $1 billion.

A balance sheet with 191 million fewer shares outstanding, debt 4% of equity and cash and near equivalents of $12.2 billion.

Seven consecutive years which cash from operations has approached or exceeded $8 billion.

And return on reinvested capital is among the highest in the industry.

Although these accomplishments deliver measurable financial progress now, we expect the full value to be realized in the months and years ahead.

I've said this before, but it bears repeating, while we can't control the economic environment, we're convinced we are taking the actions that have put Intel in an outstanding position when the economy recovers.

With that, I turn it over to Paul for additional comment on the business.

Thanks, Andy.

The fourth quarter was a strong quarter for Intel with growth in all of our major businesses, with revenues up 10% above Q3.

For 2002 in total, we executed well against our plans and grew faster than the industry, gaining market segment share on microprocessors, chipsets, graphics, gigabit Ethernet, network processors and micro processors, which go into PDAs.

Our microprocessor market segment share now stands at levels equal to the highest we've seen in the last five years.

On a geographic basis, a seasonally strong Europe led our revenue growth.

Europe saw strength in emerging markets, particularly Russia, small and medium business, and an improvement in sales into the corporate segment toward the end of the year.

Asia Pacific at $2.7 billion in revenue, set an all-time record and climbed to 38% of our revenues, up 10% from Q3 and 13% from last Q4.

Within Asia Pacific, sales in China continued to be strong along with chipset sales into Taiwan.

We also saw strength in south and southeast Asia as we step up our sales efforts in these emerging markets.

Our sales in the Americas were up slightly over Q3, with growth in North America slightly offsetting weakness in Latin America.

Japan bounced back from a soft Q3 and was also up over Q4 of last year.

Japan saw strength in both microprocessor and flash shipments.

Japanese phone manufacturers are leading the way in adopting features like camera, color and smart phone capabilities, driving up flash density per phone.

Sales out through our distribution channels were our best in 11 quarters.

This channel continues to perform well for us throughout this business cycle.

Our Intel architecture business had a solid quarter.

Microprocessor units grew to an all-time record.

Chipset volumes are above Q3 and motherboard shipments were also achieved an all-time record volume.

Average selling price of microprocessors, which go into PCs and servers, was above Q3.

Overall weeks of inventory of microprocessors declined from Q3 and our inventories are well within our guidelines.

We began volume shipments of our Pentium 4 microprocessor at three gigahertz this quarter.

This is the industry's highest performing PC microprocessor and includes an innovative technology called hyper threading, which adds substantial performance increases over those we are achieving with higher clock frequencies.

In the first half of 2003, we plan to incorporate hyper threading into a larger portion of our performance processor lineup.

We will also increase the speed of the Pentium 4 processors' front side bust to 800 megahertz in the first half, while on a support for serial ATA drives and faster DDR memories.

Our mobile microprocessor business was also up in Q4 and was stronger than we anticipated, which contributed to our overall improvement mix this quarter.

In December, we began our first revenue shipments of the bannius microprocessor in anticipation of its introduction in March.

Last week, we announced the brand name Centrino, which will encompass the chipset and wireless components.

We're very excited about the new product line and its ability to capitalize on the rapidly growing demand for wireless computing.

Our server business achieved record volumes in Q4 as well.

Much of this was driven by new products as we launched 20 new Xeon family-related products in Q4, including seven new microprocessors with speeds up to 2.8 gigahertz.

During the quarter, the Itanium 2 processor continued to set records, including the world's highest four-way results in database, application and secure web server bench marks.

Oracle and SAP went into production with their enterprise applications and we expect the softer momentum to accelerate in the first half of 2003.

We also firmed up our plans to launch the third generation Itanium family product, code-named Madison, in mid-'03.

In flash memories, our revenues, density, ASP and volumes were all above Q3.

The volumes were the highest in eight quarters.

Our "X" scale processor achieve achieved a new record for revenue and units.

Our design win momentum continues, with Dell announcing a new PDA line, and a number of other companies announced eight smart displays at last week's Consumer Electronics Show.

We introduced a new product that includes "X" scale processor and flash memory in a single package.

And we are in the process of testing samples of our upcoming Manitoba product, which integrates "X" scale, flash and 2.35 "G" base-band silicon onto a single chip.

The Intel communications group saw strong unit growth in [indiscernible] connections, and we continue to lead the industry in the gigabit Ethernet transition.

While the overall communications landscape remains challenging, we continue to introduce new products that will help our customers win the next generation of end user deployments.

We launched a new network processor, optical components for 10 gigabit Ethernet and fiber channel and new products for network storage.

The Intel Communications Fund also began investing in fiber companies such as Comada, to further accelerate the deployment of 802.11.

Lastly, we remain ahead of the industry in the conversion to 300 millimeter process technology and the deployment of 90 nanometer silicon technology.

Our plan is to qualify our 90 nanometer capability mid-'03, and we've completed the designs of Prescott and Delfon, our next generation desktop and mobile processors that, will be the ramp vehicles for that process.

Looking back, 2002 was a tough year for our industry.

Against this backdrop, Intel focused on technology leadership and operational excellence.

We gained market segment share in most of the businesses in which we participate.

And have turned in 64 straight quarters of profitability.

We believe we are well positioned as we enter 2003 and that our investment strategy of new products and new technologies is the right course for our long-term growth.

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

Thanks, Paul.

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

In the interest of moving along and giving other questioners a chance, please limit yourself to one question.

At this time, if you have a question, please signal us by pressing star and then 1 on your telephone keypad.

Again, that's star 1 to ask your question.

And we will pause for just a moment to assemble the roster.

First up, we have a question from Mark Edlestone of Morgan Stanley.

Good afternoon, guys, and congratulations on the results.

Paul, I had a question related to microprocessor units in the quarter.

Can you give us a sense as to how much faster units grew sequentially in Q4 than they did in Q3?

How they grew -- you mean over the course of the quarter?

No, just if we look at the quarter as a whole, how much faster did they grow in Q4 versus Q3?

And I'm specifically interested in the non-Xbox-related profit uses.

Almost all the growth was non-Xbox, and principally in PC and server based microprocessors.

Okay.

And given that, how much faster that that total grow in Q4?

If you go off the revenue numbers, you can work backwards, Mark.

That's the simplest way to do that.

Revenue growth for that was about 10%.

Okay.

I -- you had ASPs up there as well, right, Paul?

That's right.

Okay.

And when was the prior record in microprocessor units for the company?

A while back, probably eight or nine quarters.

Great.

Thank you very much.

I don't have it at the tip of my fingers.

Thanks a lot, Paul.

The next question is from Joseph Osha at Merrill Lynch.

Yes, good afternoon.

Just one question.

Paul, can you tell me what exactly is included in the Centrino platform; is there graphics in there?

And also is there a gigabit Ethernet specified as part of that?

And also as a follow-on, I am correct in understanding that OEMs must buy that complete suite of products in order to use the Centrino brand?

Just wanted to confirm there.

Without preannouncing the product, it will include the microprocessor formally known as [indiscernible], it will include a chipset that initially doesn't have graphics, but will be supplemented by a graphics version very shortly after introduction.

That's ODOM, Paul; that correct?

The initial chipset is code-named ODOM, that's right.

And it includes the 802.11-B silicon in the PCI form factor at launch.

The second part of your question was on the use of the brand name Centrino.

Yes, you're correct, while OEMs can certainly buy any or all pieces of -- of silicon that I described, it -- it is required that they use all of those pieces in order to take advantage of the Centrino brand name and the marketing programs associated with Centrino.

Okay, you didn't mention [indiscernible] or Ethernet.

Is that not part of the specified platform?

Correct.

Thank you very much.

Next up, we have a question from Jonathan Joseph, Salomon Smith Barney.

Yes, good afternoon.

I wanted to ask you, Andy, with regard to your guidance of gross margin for the new year, 51%, can you -- is there any takeaway there for investors with regard to your anticipated top line growth or perhaps utilization rates in that guidance?

You know, Jonathan, not really.

It's -- it's not precise enough to give you any real guidance to be honest.

Right now, you know, we saw essentially two years of flat revenue, I mean seasonality, but if you look at the numbers, it's kind of flat.

We forecasted Q1 to be seasonal.

If you look at those years, what you saw was margins from -- going from 49 to 50.

What I'd like to do is make sure we find a point to stay in the same environment.

My range is wide enough, saying we'd get growth with a few points and we will get to the high end with revenue growth.

By the way, if we see revenue shrinkage, we will get to the low end.

What we're really trying to do is manage the company very cautiously.

You know, keep head count flat to down, spending flat to down.

Try to find a point of margin in this environment and then get a lot of upside if we see revenue growth.

That's okay.

A little bit of a complex follow-on, but capital spending is reduced this year, yet -- and partly as a result, I think that efficiencies seem to be improving pretty nicely as you mentioned in 300 millimeter.

Would there be cost savings from the greater efficiencies in 300 millimeter?

And possibly some upside to gross margin as a result?

Let's see, the answer to the question is yes, we see the cost efficiencies.

The key to recognizing that into the P&L, of course, is utilization of the factories, which means you need to see enough unit growth to actually meet that capacity or have to deal with excess older capacity, which I don't plan to do.

I plan to keep it in case I need it.

My more complex answer to your complex question is that's why I say if we can see some revenue growth, we'll see real benefits into the P&L because we have the technology in place.

We have it positioned and can take advantage of it and drive costs much faster.

So, the variable is really top line.

The variable really is top line and use of those factories, yes.

And -- again, what we've been trying to do over the last two years is position ourselves so that when the recovery comes, you know, we have the technology in place both in products and in the factory space.

We think we're actually there.

You know, we're gaining market segment share, the products are ready. 300 Millimeter is coming along nicely. 90 nanometer for later this year.

Everything is happening the way it should.

The question is when does the market return?

Great.

Thank you.

Next up we have a question from Nimal Vallipuram at DKW Securities.

Yes, this is for Paul.

Paul, if you look at the demand, overall demand for the PC market, we are in a situation where for the last few years we have not seen much of a demand for the PC market worldwide.

I'm not asking for a full internal forecast, what you're expecting, but if you can give us subjectively what you expect from corporate market on the basis for the near term as well as for this year.

Well, Nimal, you know, I want to maybe make a couple of comments.

This was a record quarter for us in terms of microprocessor shipments.

So, it's not exactly anemic out there in terms of consumption of microprocessors, particularly as they go into PCs and notebooks and desktops.

The only help I can give you going forward is really to reiterate what Andy said about the first quarter guidance, which is, you know, more of a seasonal outlook, a normal general outlook, which would imply that units would be down a bit in Q1, which we have seen nine out of the last 10 years or something like that.

I really can't help you much in terms of the granularity for the second half of the year, except to say we're ready for whatever comes at us in terms of the overall demand and capacity situations.

Thanks, congratulations for a good quarter.

You're welcome.

Moving on, we will take a question from Dan Niles at Lehman Brothers.

Great.

Thank you.

Congratulations on a solid quarter.

Paul, this one is maybe more for you.

Can you kind of talk a little bit about the corporate demand environment, versus consumer?

In other words, you talked about, I think already, that portables were a little better than you thought and so were servers, those, obviously, tend to be more as a percentage basis, bought by corporations and your ASPs were up sequentially, which would tend to imply the corporate demand was good.

Can you give us whatever color you can on that?

And then, I guess on seasonal patterns going into Q1, my recollection is that Q4 usually ASPs get pressured by a greater mix of lower-end consumer stuff and in Q1 that, let's off so your ASPs have some room to move.

Can you comment on that, too?

I'll try, Dan.

I think that what I would characterize what we've seen with the corporate purchases at this point in time as being anecdotal.

There is not a clear pattern that -- that it is returned although we have seen a number of fairly large spot purchases.

At very large corporations, and I said in my commentary that we saw some upside in December in Europe from corporate purchases, as well.

So, there are -- there are some, you know, points of light out there, although nothing that I would say, at this point, indicates a clear pattern.

And when you look at the install base, it is aging.

And it's aging about three years at this point in time, it is approaching a point where you would normally expect those machines have to be retired just out of a cost basis, cost and productivity basis.

So, I don't think that really drove our results in Q4.

The ASP in Q4 was really driven by a slightly more favorable mix shift in total in all geographies toward Pentium and away from Celeron, particularly -- we also got some lift out of the 3 gigahertz introduction in the quarter and sold very high volumes of that product.

Okay.

And then maybe a follow-on to that, to your aging -- your comment on an aging installed base.

You know, how relevant do you think it is that, you know, on June 30, Microsoft stopped support of windows 98 and NT 4X in terms of giving maybe some kind of kick to get corporations to think about what they want to do with that installed base, how relevant...

I think every company has to answer that on their own, Dan.

I can tell you it was a factor in Intel's decision to buy quite a few PCs starting again this year.

Great, thank you very much, Paul.

Next up is Grant Sinaca, Sinaca Capital Investments.

Yeah, hi.

I just was wondering if you could talk a little bit about the mobile processor side that I assume helped enrich the ASPs and mix?

And talk a little bit more about Asia, which seems to be breaking out at the seams.

Thanks.

Let me do them in reverse order.

Asia has been breaking out of the seams all year and probably most of the prior year, as well.

It's been one of the clear positive signs of growth for us in the last 18 months or so.

And I don't see see that abating, it's driven by a number of structural changes in the industry and in some cases, offshore manufacturing shifting increasingly to China.

Mostly driven by local consumption in Asia Pacific, driven again by China.

I don't see that slowing down at this point in time.

In terms of overall mobile growth, we've seen a move towards mobile in the corporate space and in consumer space over the last year.

Albeit a slow move.

I -- I -- as I said earlier, when I look forward into 2003 and think about the introduction of the Centrino architecture and products based upon that, we get very excited because it taps into the mobile conversion and it taps into wireless, both of which are probably the two hot spots, no pun intended, in the market today.

Have you done any studies or seen any that give you any indication at all on what the incremental demand could be from having 802.11 and other features in '03 and '04?

It gets into how seamless its integrated and how easy it is to connect to the various access points around the world.

I neglected to mention earlier on, when I was answering the question -- on Joe's question on Centrino, that we will be shipping quite a bit of software with that -- with those silicon chips and that a lot of that software is focused on discerning signals and checking access points in a fairly seamless fashion.

Thank you, congratulations.

You're welcome.

Moving on to UBS Warburg's Thomas Thornhill.

Mr. Thornhill, your line is open, please go head.

Thank you.

Shifting away from the IAG group for a minute, looking at WCCG, the flash business, surprised to see the loss increase on an increase in revenue.

And can you speak to the trends in the operating profits on that -- those -- that sector and ICG as we go through 2003 and a follow-on on that one.

I can certainly do part of that, Tom.

What you saw in the fourth quarter in WCCG, you saw underutilized capacity, which we anticipated going into the quarter.

An inventory reserve on a couple of specific products, which, you know, basically had limited demand and we had to write off.

And finally, we, again, in an effort to clean up a line of capacity, we stopped production in one of the older fabs in the flash product lines.

So, what you really saw was a one-time clean-up.

I don't think you will see it repeated, but certainly it cost me not quite $100 million, but nearly $100 million in writeoffs and inventory for -- writeoffs and inventory or costs of sale in the quarter.

Going forward, I would hope we can do better than that.

While we don't generally provide forecasts, you know, by group, if you go back and look at year after year after year of the segment reporting, what you will discover is the flash group, the WCCG group typically gets hit the hardest in the first quarter.

So, I'd expect to see some struggle somewhat there and hopefully we will continue to see strength and growth through the rest of the year.

We think the cell phone business is coming back, we see colorful screens, cameras, PDAs, a lot of good things are happening.

Without giving a specific forecast, I'm pretty optimistic there for the year.

I think we have one quarter we have to get through, just due to normal seasonality.

In terms of the communications group, you know, we've seen, you know, a sharp drop year-over-year in revenue, we continue to downsize, we continue to refocus.

You know, we saw a little bit of growth in revenues, a little bit of growth in profit in the fourth quarter, but certainly it continues to be very, very tough business, it continues to be the most depressed environment and I think they will continue to struggle probably beyond the point we see recovering in the other businesses.

Does the -- it is a follow-up, does the -- does management strategy include spending between 25 and 33% of the profitability of IAG to support those businesses over the next year?

As they have over the last two years?

Well, I don't want to get into providing specific pretax forecasts for those business, we actually remain committed to the communications business and to the wireless communications business.

We will continue to invest.

One of the things we have to do, Tom, with all the new accounting rules, is we have to do a very detailed analysis of those businesses.

We have to look at the assets employed.

We have to look at the returns expected.

Those businesses both have good futures if we ever get to a future.

So, no, we're not slacking off, we're still committed to it.

Thank you.

Adam Parker, Sanford Bernstein.

Hi, how much will the microprocessor up sequentially and how does it apply to the Q1 guidance here?

You know, it's a level of detail we don't provide.

All I would say is the ASPs were up on microprocessors, with the PC sector some.

As Paul said, better mix.

But, again, it's not a major factor.

The bigger factor is the fact that we saw growth in units and I really can't do much more for the next quarter as I'm going to tell you, Paul said units will be down a little bit.

We said revenue would be down a little bit and said margins would be down a little bit.

Shifting gears, then, in the past couple of calls, you mentioned that managing inventory is crucial and talked about the underutilization charge, I'm looking at given your, you know, given you're going to have reduced stability with the processors as it ramps up and the seasonal slow down of PCs in Q1, can you quantify the impact on margins you anticipate the underutilization charge will have on 2003 or the first part of the year?

It's -- to directly answer your question, I won't give you that much.

What we saw was an increase in underutilization charges, primarily in the flash factories in the fourth quarter.

That's getting better in the first quarter.

Remember I took a factory, one of the older flash factories offline?

Right.

We're continuing to, you know, we've drawn inventory down some in the fourth quarter, so, we underutilized some factories for that.

Like Paul said, we saw a a nice response in products, which used more of the capacity.

The underutilization charges actually get better in the first quarter.

On the other hand, I have start-up costs for the 300 millimeter factory in Albuquerque just about offsetting it.

The way you look at the first quarter is that utilization, start-up costs on a new technology and a total change in margin is directly related to the change in revenue.

Okay.

Just one other question, then, if I can.

Can you just talk about, you know, your advertising strategy in general?

It's overall expense and, you know, what the boost in incremental revenue you get as a result of that advertising?

What you mean in -- in the first quarter?

Or just -- you know, just as a strategy, I mean it seems like the more we see, you know, advertising -- I wonder what the overall expense is and assuming you guys do the math on the incremental boost in revenue you anticipate over some period of time as a result of that expense.

Well, you know, there's two levels of advertising, right?

There's the very broad MDF program, which is Intel Inside.

That has an impact mostly, not in just advertising our products, but in providing a vehicle or a very loud voice of the computer industry to end users and IT buyers, and I think it's been one of the most fruitful technology campaigns in the world, simply because of the scale and ability to tell people about new technology.

Our direct advertising, Intel direct advertising, the thing on television, is more focused on point products.

We want to convert a market or a technology from an older generation to a newer generation or want to advertise in the upcoming Centrino launch, that technology and the move to wireless notebooks.

We believe this kind of direct advertising is key to converting from our older product lines to our new product lines and we've been doing it for a decade.

Okay.

Great, thanks a lot.

We have a question now from Scott Randall at SoundView.

Great, thank you.

Andy, can you talk a little bit about the unit cost structure?

I think at the November webcast, you shared data showing declines anticipated but at a slower rate than you originally thought.

Has that volume moved at all between then and now?

Not really, Scott.

What would mostly make that bar move differently is a change in economic environment, a change in unit demand.

You know, again, what we saw in Q4 was a pleasant surprise, but didn't change our overall opinion about demand in the markets.

Sure.

And secondly, to the extent you've got visibility this, what would your view in entering the channel be right now?

Our own distributor channel inventory is in very good shape.

I think you need to ask our customers about their own PC inventory in the channel, but my sense is that it's nothing extraordinarily large.

If feels like a fairly healthy overall inventory environment.

Great, thanks, Paul.

Charles Boucher has our next question from Bear Stearns.

I just was wondering if the, you know, Andy, the guidance on gross margin, if you were -- if your projections are based on a typical seasonality that you'd normally see throughout the year and sort of a related question, I guess, just as you look at the year, you had a pretty steep, you know, speed upgrade curve in 2002 and at least least based on the current road map it looks like it won't be quite as rapid in terms of speed improvements.

How do you offset that to maintain relatively stable pricing?

Anything you plan to doing to handle that?

Wonder about those two questions.

Let me take the first one and Paul can work on the second one.

In general, again, if you look at the lid point of the gross margin range, 51%, pretty much assumes seasonality.

A lot of variables, I really had to revise it both of the last two years.

You know, the economy can change fairly dramatically inside a calendar year, which gives different answers.

But what I've assumed is, you know, assumed expected seasonality, a relatively wide range.

If we see a pickup, it will be at the upper end of the range.

If we actually see further declines, it will be at the lower end of that range.

And, Charles, the second question is kind a difficult one to answer because we are not pulling back on the -- on the -- if you will, the gigahertz throttle at all.

Our intention is to move that as fast as possible, hence the direction of aggressively ramping to 90 nanometer silicon.

I gave you an update on the two lead vehicles in my commentary.

So, that's the path to higher clock frequencies over time.

Having said that, you've seen, in the last quarter, with the introduction of hyper threading on to the Pentium 4 desktop product line, sort of the first footprint in the sand of what we will be doing with our architectures.

We will continue to add features beyond clock enhancements to the architectures that add performance or other kinds of value-added features.

Larger caches would be one element.

Different kinds of micro architectures like hyper threading and other.

I talked about security coming into the microprocessor over time at our last developer's forum.

Similarly, you will see -- a comparable level of activity in our chipsets as we move to expand more of the functionality on the motherboard and embed that into the chipsets like we're doing with Centrino.

I think you need to look at our road maps a little bit differently as you think about how we approach the various compute platforms that we compete in.

Okay.

Thanks, Paul.

This question comes from Thomas Weisel's Max Sherin.

Yes, thank you.

Could you give us an idea of the demand that you saw in the white box market?

And also by geography there?

And also, I know the distributors play a big role there, can you give us a percentage of sales on distribution through the processor business?

Thanks.

The -- really the only data I can give you is the comment I made in my commentary on the channel business being very robust, strongest sales out through the channel in -- in 11 quarters.

As you know, we only report sales out, not sales in in our revenues.

And the -- it's hard to say, but the -- the large share of that, I think in excess of 75%, is going into the white box market.

People -- there are other sources that buy in that particular channel, but principally into unbranded systems and that's remained quite strong, particularly in Asia Pacific, where we saw quite a bit of our growth this quarter.

Okay, thank you.

This question now from Hans Mosesmann at Prudential Securities.

Yes, first a clarification, what's the tools portion of your Cap Ex for '03?

And then the question is: Is the processor in the Centrino, is that a Pentium 3-based core?

Thanks.

You do the second, I will do the first.

When you say tools-based portion, you're asking about FAB equipment versus other, FAB equipment is approximately half of it.

Buildings, land construction, you know, finishing the fab in Ireland is a little over a quarter.

A little less than a quarter goes to assembly and test, which, again, you know, is some equivalent there.

It goes to engineering, logic analyzers, you know, equivalent for fab, for not fabs, but for the labs, PCs, service for the company.

And on [indiscernible], it is an entirely new core.

We took advantage of some of the infrastructure that was out there for Pentium 3 in terms of how we designed it, but it's also -- it's -- it has is own packages, it looks more like -- it looks like something between a Pentium 3 and Pentium 4 in terms of architecture, but was designed and optimized for low power, very high performance mobile environment.

And is that a .13 micron --

The initial version is .13 and the product I described in my commentary called Delfon, which we completed design on, is a 90 nanometer version of that -- of that chip.

Thanks a lot.

We will take two more questions, please.

All right.

The first of those will come from Tim Mahon at CS First Boston.

Yeah, it might be for Andy.

Andy, Ron Smith said a few months ago that Intel was expected to increase prices in flash.

I'm just curious if you could give us an update?

How is that going?

Is it generally across-the-board?

And then I just have a quick follow-up.

Yes, increases in price were sent to the customers, recognized a lot of that business is under long-term agreement.

We honor the agreements already in place.

The price list was a little bit higher.

As you expect with any increase in price, it isn't welcomed with open arms, but we're sticking to our guns and what we really believe is it's an environment where if you look forward, look again at the color screens, cameras, the PDAs, the technology is evolving and leading edge flashes is going to be and is in tight supply.

As a result, a price increase is appropriate.

And, Paul, maybe if you look at 2002 and maybe where 2003 is going to shape up from the notebook perspective, do you have an idea of the number of notebooks you expect to be sent in light versus the ones to be desktop replacement?

Oh, gosh.

I think virtually all of the corporate notebooks are thin and light and what we see in terms of the desktop replacements has been principally what's sold in retail.

Our experience is that as people buy their second note -- go to buy their second notebook, they buy thin and light machines.

You've learned and tired of lugging a large, heavy machine around.

So, I think this is more of an experience curve.

I would expect even in the retail markets, you will see a shift toward thin and light over the course of next year, particularly with the new machines.

Are you thinking like 60/40?

The majority of the machines are based upon mobile technology and in the thin and light category today.

I would expect it to grow over time.

Okay.

Thank you.

We will take the final question from David Woo at Wedbush Morgan Securities.

Yes, thank you.

One question for you, Andy and one for Paul.

Andy, can you talk about the -- you have had substantial writedowns of your portfolio in the last couple of quarters and you guided another $170 million in Q1.

Are we getting to the end of those kind of writedowns for the nonoperating income line in '03, beyond Q1?

For Paul, the -- it looks like with Delfon out, essentially the Pentium 4-M won't be a meaningful slot in your portfolio of products beyond the second half of 2003.

Am I looking at it right -- correctly?

Okay, the portfolio, first of all.

I need you to understand this is probably the hardest thing we have to try to forecast.

Yes.

I understand.

It is a function of the company, it is a function of the venture environment, it is a function of whether the companies are going to meet milestones or not, find cash or not.

This is mostly private portfolio.

So, there is not a public market for this.

So, when you look at my forecast, you should recognize, you know, it's a -- it's a best guess of events that have yet to occur and try to figure out where they come out.

Now, after I say all that, the answer to the question is I don't know.

You know, certainly we've written down a substantial portion of the portfolio.

There is enough value left on the books that we continue to see knowledgeable numbers for a few quarters.

It's really more a function of what's happening in the technology environment, what's happening in the industry, do we start to see a recovery in the business?

If we do, you can see these things get smaller and fast.

If we continue in the environment where funding is scarce, where companies are having difficulty getting customers making breakthroughs, it will continue to be a sizeable number for a while.

And on the -- the mobile -- if you will, the conversion from Pentium 4 to bannius and Delfon, David, once we launch the product in March, we will move as rapidly as possible to move as much of our performance based microprocessors in mobile to the new technology in mobile.

Our intention will be to use both brands over the course of the year.

The bulk of the volume because of timing will be bannius.

We believe it is a superior product and have very strong design wins and indication from our customers that they will ramp it aggressively.

Thank you.

We would like to thank everyone for listening to today's conference call.

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

Pacific time tonight.

Those interested should call 719-457-0820 and reference passcode No. 735793.

Thank you.

That concludes today's conference call, again, thank you all for joining us, have a good day.