德州儀器 (TXN) 2003 Q4 法說會逐字稿

Good afternoon, ladies and gentlemen, and welcome to the Texas Instruments fourth quarter 2003 earnings conference call. (OPERATOR INSTRUCTIONS) It is now my pleasure to introduce our host, Mr. Ron Slaymaker.

Sir, the floor is yours.

Good afternoon.

And thank you for joining our first quarter earnings conference call.

With me today is Kevin March, TI's Chief Financial Officer.

This call will last one hour.

For of you who missed the release, you can find it on our website at www.TI.com/IR.

This call is also being broadcast live over the Web, and can be accessed through TI's web site.

A replay will be available through the Web.

This call will include forward-looking statements that involve risk factors that could cause TI's results to differ materially from management's current expectations.

We encourage you to review the Safe Harbor Statement contained in the earnings release published today, as well as TI's most recent SEC filings for a complete description.

Our midquarter update to our outlook is scheduled this quarter for March 8th.

We would expect to narrow or adjust the revenue and earnings guidance ranges as appropriate with this update.

We will observe a quiet period beginning on March 1 until the update.

In today's call I will review our revenue performance, and then Kevin will discuss profit performance and the first quarter outlook.

After this review, we will open the lines for your questions.

Fourth quarter TI revenue grew 9 percent sequentially and 29 percent from the year ago quarter, slightly exceeding the upper end of our most recent guidance range.

Revenue was strong across almost all of our semiconductor products end markets.

In fact, this was the first time in over 10 years that we have had consecutive quarters of double-digit sequential growth in our semiconductor business.

Semiconductor order growth was also strong, leading us to believe we have the potential for stronger growth than the seasonal norms in the first quarter.

Semiconductor revenue in the fourth quarter was up 16 percent sequentially and up 34 percent from the year ago quarter.

Sensors and Control's revenue grew 4 percent sequentially, and grew 5 percent compared with the year ago quarter.

Educational and Productivity Solutions, or E&PS, revenue seasonally declined 57 percent sequentially due to the end of the back-to-school period for graph and calculator sales, and declined 3 percent from the year ago quarter.

In Semiconductor, DSP revenue grew 18 percent sequentially, and was up 52 percent compared with the year ago quarter.

Sequential DSP revenue growth was driven by wireless.

Wireless was also the primary factor in DSP growth from the year ago quarter, although growth in the digital consumer and broadband markets was also notable.

Digital consumer DSP revenues more than doubled from the year ago quarter, and DSP revenues from the broadband DSL market was up more than 75 percent.

Analog revenue grew 16 percent sequentially, primarily due to strength in high-performance analog standard products, as well as growth in shipments of analog products for wireless applications.

Analog revenue was up 26 percent from the year ago quarter due to high-performance analog, as well as growth in servo and preamp products for storage and broadband wireless LAN products.

Other semiconductor revenue was up 11 percent sequentially and 23 percent from the year ago quarter due to growth in RISC microprocessors, DLP, standard logic products and royalties.

Microcontroller revenue was up compared with the year ago quarter and about even with the third quarter.

I will turn now to review those semiconductor products focused on targeted end equipment, starting with wireless.

Revenue from shipments into the wireless market grew 23 percent sequentially, and grew 41 percent from the year ago quarter.

Advanced feature phones using TI's latest wireless technology, including 2.5 G modems and OMAP applications processors, were the biggest factor in this growth.

In addition, revenue from complete wireless chipset modem solutions that include the digital baseband, analog baseband and power management, as well as RF transceivers, reflected strong demand from Asian ODM customers and a success of TI's solutions in this market.

TI's opportunity for higher content per phone continues to build as demand increases for our 2.5 G modem technology, complete chipset and OMAP applications processors.

In addition, CDMA represents an emerging opportunity for TI based on Nokia's success, as well as TI's entry into this market with standard chipsets.

In the fourth quarter TI and ST MicroElectronics announced that we are sampling our jointly developed cdma2000 1X chipset.

And we now have active OEM customer engagements under way.

We also announced that we're developing a cdma2000 1xEV-DV standard chipset that we expect to sample early this year, with full production anticipated by year-end.

These chipsets, along with significant engagements for our 3G UMTS WCDMA technologies provide important opportunities for TI in the next few years.

Broadband revenues grew 12 percent sequentially driven by strong demand for DSL, and was up 75 percent from the year ago period, with growth particularly strong in DSL and wireless LAN products.

Finally, the combination of catalog DSPs and high-performance analog revenue grew 17 percent sequentially, and 35 percent from the year ago quarter.

Both DSP and high-performance analog were factors in the growth, with each contributing double-digit sequential growth in the quarter.

TI's high-performance analog sequential revenue growth has accelerated in each of the last three quarters, as our investment in this important area are producing significant and profitable results.

Customers are embracing our new products, such as the ADS 5500, a 14 bit 125 megasample per second data converter that was announced in early December.

This part offers an unmatched combination of speed, performance and low power consumption for applications such as wireless base stations and tests and measurements.

The level of customer enthusiasm for our latest products leads us to believe that we're on the right track in high-performance analog.

Before I turn it over to Kevin, let me do a quick snapshot of 2003 highlights.

TI revenue grew 17 percent to over $9.8 billion for the year.

Semiconductor was the biggest factor with 20 percent growth.

DSP grew 36 percent, and was about 35 percent of Semiconductor revenue for the year.

Analog grew 13 percent, and was about 40 percent of Semiconductor revenue.

In the targeted end equipment, wireless grew 32 percent, and was about 35 percent of Semiconductor revenue.

The combination of catalog DSP and high-performance analog grew 24 percent, and was about 15 percent of Semiconductor revenue.

Broadband grew 71 percent, and was about 5 percent of Semiconductor revenue.

These results encourage us that our focus on customers is allowing them and TI to be successful.

We believe we are engaged in the most exciting parts of the semiconductor industry, and are successfully addressing the most promising end equipment with the world's best customers.

At this point, I will ask Kevin to review profitability and our outlook.

Good afternoon everyone.

TI's fourth quarter gross profit was 1,194,000,000, or 43.1 percent of revenue.

Gross margin was up 2.4 percentage points compared with the third quarter, and up 7.3 percentage points from the year ago quarter.

In a sequential comparison, the change in gross profit is a mix of higher profit in Semiconductor and lower restructuring charges, partially offset by lower seasonal profit in E&PS.

Compared with the year ago quarter, profitability is up due to higher revenue.

Semiconductor gross profit grew by 184 million sequentially to 1,083,000,000, or 44.2 percent of revenue.

The sequential margin increase of 1.8 percentage points reflects increased utilization of fixed cost manufacturing assets.

Gross profit in Sensors and Controls increased to 95 million, or 37.8 percent of revenue.

Gross profit in the E&PS seemingly declined from the third quarter to 37 million, or 49.2 percent of revenue.

TI's total fourth quarter operating expenses of 756 million in the fourth quarter was down 25 million from the third quarter, reflecting lower restructuring charges and a $23 million in process R&D charge incurred in the third quarter.

Although we have proactively increased our R&D spending in 2003 to further strengthen our wireless development activities, we remain committed to tightly controlling SG&A expenses.

TI's SG&A was 11.1 percent of revenue in the fourth quarter, the lowest level we have operated at for many years.

For the fourth quarter TI's operating profit was 438 million, or 15.8 percent of revenue, up 76 percent from the third quarter.

Semiconductor operating margin was 17.7 percent, up 5.2 points from the third quarter.

Sensors and Controls' operating margin grew to 25.4 percent, while E&PS operating margin seasonally declined to 15 percent of revenue.

For TI other increment expense, including interest income, investment gains or losses and other items, produced income of 131 million in the quarter.

This included a $97 million gain from our sale of Micron stock in the quarter.

Interest expense on loans was $8 million in the quarter.

The Company's effective tax rate was 20 percent in the quarter, exclusive of the tax impact resulting from the recognition of the previously reserved $62 million tax benefit associated with the Micron stock transaction.

This rate was lower than our prior estimate of 26 percent due to changes in our estimate of the Company's foreign tax liabilities that were made during year-end accounting reviews.

This was resulted in a $37 million tax reduction in the fourth quarter.

Net income grew to 512 million, or 29 cents per quarter.

The total earnings contribution from the Micron stock transaction during the quarter was 7 cents per share.

And 2 cents resulted from the tax reduction associated with the foreign tax liability adjustment.

The quarter's results included $20 million of pretax amortization of acquisition related costs.

Total cash of 5,664,000,000 was up 1,126,000,000 from the prior quarter, and 1,522,000,000 from the end of 2002.

Cash flow from operations increased to 1 billion 67 million, up $557 million sequentially, and 323 million from the year ago quarter.

Capital expenditures of 272 million in the quarter increased 39 million sequentially, and 36 million from the fourth quarter of 2002.

For the year capital expenditures were 800 million.

Depreciation of 370 million increased 14 million sequentially, and decreased 27 million from the year ago period.

For the year depreciation was 1,429,000,000, down 145 million from 2002.

Accounts receivable at 1,451,000,000 decreased sequentially by 71 million due to seasonally lower E&PS revenue.

Days sales outstanding declined to 47 days at the end of the fourth quarter from 54 days at the end of the third quarter.

Inventory in the quarter in the fourth quarter decreased $11 million sequentially to 984 million due to the seasonal decline in E&PS revenue.

Days of inventory declined to 56 days at the end of the fourth quarter, from 60 days at the end of the prior quarter.

TI's orders in the fourth quarter were 3,076,000,000, up 16 percent sequentially.

Semiconductor orders were 2,744,000,000, up 20 percent sequentially.

Semiconductor's book to bill ratio was 1.12, up from 1.08 in the third quarter.

Turning to our expectations for the first quarter, we currently expect total TI revenue to be in the range of 2,720,000,000 to 2,950,000,000.

Semiconductor revenue should be in the range of 2.4 billion to 2.6 billion;

Sensors and Controls in the range of 255 to 275 million; and E&PS in the range of 70 to 80 million.

Earnings per share are expected to be in the range of 16 to 22 cents.

Restructuring charges in the first quarter are expected to be about 10 million, and should remain at about this level through each of the remaining quarters of 2004.

The effective tax rate for 2004 is expected to be about 30 percent, reflecting our expectations for higher profitability in the year ahead.

Similarly, with higher revenue levels and profitability anticipated in 2004, we expect to cross the Company's minimum financial performance thresholds for TI's employee profit-sharing program.

Profit-sharing is one of the Company's variable compensation programs and helps TI to retain employee talent, but maintain competitive compensation practices over the course of the cycle.

The Company has not met our profit-sharing criteria since the year 2000.

And we look forward to once again delivering the results that will provide profit-sharing to the broad base of our employees.

Accruals for profit-sharing will be made quarterly in each of the quarters of 2004, based on our estimate of the Company's full year financial performance.

At this point, profit-sharing accruals are estimated to be approximately $60 million per quarter, about evenly distributed across cost of revenue, R&D and SG&A.

For 2004 we expect R&D to be about 2 billion; depreciation to be about 1.4 billion; and capital expenditures to be about 1.1 billion.

Capital expenditures will be up from the 2002 level, but are expected to remain below our depreciation level for the third consecutive year.

Depreciation should drop by about 30 million in the first quarter, and then build through the quarters at 2004 in a similar profile to our 2003 depreciation expense.

In summary, we're encouraged by the strong momentum in our markets and our business operations.

It is notable that TI's operating margins have returned to double-digits again at almost 16 percent revenue, and a significant headwind for further expansion as revenue continues to grow.

We made a lot of progress in 2003, and are increasingly optimistic that 2004 will be even better.

With that, let me turn it back to Ron.

At this time I will ask the operator to open the lines up for your questions.

In order to provide as many as you as possible an opportunity to ask your questions, please limit yourself to a single question.

After our response, we will provide you an opportunity for an additional follow-up.

Operator?

(OPERATOR INSTRUCTIONS).

John Barton of Wachovia Securities.

Ron, I wonder if you could elaborate a little bit on the attach rate for OMAP?

You talked about that being a driver, and the ability to continue to drive dollar content in the handset.

If you could give us an idea of the tax rate today and what your forecast are, I think, over time, I would appreciate it.

I guess probably the best way I would describe it is that OMAP attach is still a very low percentage of our overall, say, unit sales.

For example, compared to digital baseband units, OMAP units would be a single digit percentage of the total.

So still a very low penetration, or attach, level, and has tremendous opportunity for growth as smart phones continue to deploy and as 3G becomes a more significant factor over time.

Again, probably the most significant driver of our OMAP revenue at this point would be the 3G deployment in Japan, where TI's OMAP applications processor is in most of these forma (ph) phones.

But it also has penetrated into 2.5 G smartphone applications, as well as PDA.

I don't have a specific time line of expectations to provide, John, other than certainly we expect that penetration of applications processor in general, as well as TI's penetration into that market, to continue to build.

As a follow-up to that, could you give us a basic feel for the revenue levels today for OMAP?

We're not breaking that out specifically, John, other than to say it is what we would describe as a significant part of our wireless revenue.

But we're not breaking it out any more specific than that.

Andrew Root of Goldman Sachs.

A quick question, or actually to finish on the OMAP question, Ron.

Last quarter I think you gave $1 contribution to year-over-year growth from OMAP.

Could update us on that for this quarter?

Andrew, we're not specifically going to provide that level of detail every quarter.

It was certainly, as it was last quarter, a significant factor in the growth compared to, say, fourth quarter a year ago.

But the reason we gave that number specifically last quarter was to help you understand that our wireless content per phone has multiple growth drivers.

I think so much attention had gotten focused on 2.5 G as the only content adder to the handset over time that we wanted to make the point that we actually have several significant growth adders in the content, including 2.5 G, including OMAP applications processors.

And also just the increasing importance of chipsets to our wireless revenue base as well.

My question was on incremental gross margin as well, if I could just quick follow up.

It dipped a little bit.

And if you can just remind us what you are expecting -- what percent of each revenue dollar to fall through to gross margin?

And in this quarter specifically, if there's anything in particular that caused the deceleration, or just quarterly noise?

I am thinking specifically about wafer costs from foundry and what you are expecting there?

Andrew, I would characterize that as probably just the quarterly noise that we're talking about.

You have always got things going on from quarter to quarter that will change a little bit.

For example, we had $14 million increase in depreciation from third quarter to fourth quarter.

So that will affect the drop through a little bit.

You'll get -- with the R&D levels that we have got -- the new product that we're introducing, from any one quarter, we will have more or fewer new product ramping to production, and this has some yield effects on that.

Again, you'll have noise level from quarter to quarter, as you mentioned there.

It is best to look at that over longer periods, over a year period of time.

As we go forward, we would expect our follow through on our analog products to be pretty high, given that we source all that internally, and we continue to have open capacity in the analog factories.

On the more advanced lithographies and where we're using foundries, naturally the fall through will be lower than what we would have for internally sourced, given that it is our purchase price from the foundries.

In addition, we will have more fall through in the future as we begin to see more strengthening in the pricing of our commodity, which we have not -- we're only just now beginning to see some turnaround in the pricing in the commodity spaces.

So over time, we will see some pretty healthy fall throughs as we go through into the future.

And Andrew, I think if you look at year-over-year fall through, I believe our fall through this quarter was just below 70 percent to gross margin.

I think that is pretty consistent with where we were last quarter.

Probably where you see more up-and-down, as Kevin pointed out, is when you look at it on a sequential basis.

Michael Masdea from Credit Suisse First Boston.

Good quarter guys.

First question for Ron.

You mentioned that you're pretty confident we are going to have a better than seasonal first quarter here in semis.

Can you help us understand if that is a unique kind of TI phenomenon?

Do you think a lot of ASP driving a product cycle, or is that kind of an underlying phenomenon that you're seeing from the customers?

I believe what we're seeing is pretty broad-based.

And as we commented even about fourth quarter, the growth we saw went across almost all of our semiconductor product lines.

In fact, must of the major product lines were all in double-digit sequential growth in fourth quarter.

And it is that general strength that we're seeing across our business that we believe carries forward into the first quarter as well, or has the potential to carry forward at the same time.

So clearly that is partly market-driven.

At the same time, we do believe that we're gaining share, especially in both DSP as well as analog.

Maybe a quick follow-up for Kevin.

On the profitability side, you guys have a number of levers that are hitting potentially in '04.

Can you help us prioritize between what is happening with depreciation and mix and commodity pricing, and fill in the fabs, etc.?

Sure, Michael.

One of the biggest things we got obviously going from fourth quarter into first is -- the good news is we expect growing revenues and profitability.

Along with that comes a higher tax rate.

So we will have to deal with that from a fourth to first transition.

That is the 30 percent tax rate I mentioned earlier.

In addition to that, we're reaching thresholds now where we've got profit-sharing accruals.

And that is going to be a $60 million fourth to first transition that we will be managing through also.

As we go through the balance of the year, we will continue to source externally, as I mentioned a few moments ago, on the advanced lithography as well as internally.

And our analog will continue to source internally, so we will get some good follow through on that.

One of the main things that we're looking forward to also as we go through the next couple of quarters is we're finally beginning to see pricing strengthen a bit in the commodity spaces.

And that pricing should fall straight through to the bottom line at 100 percent for each penny of increase that we get there.

So we will be looking forward to that expanding our profitability at a faster rate perhaps than what our revenue actually expands.

Cody Acree of Legg Mason.

Good quarter guys.

Maybe I could just follow up there for minute, Kevin.

Is there anything in the first quarter outside of the changes in the taxes, the changes in the profit-sharing, that would lead toward a guidance range and revs which are basically flat to up, but yet an EPS level that gives the possibility of a declining EPS?

Is there anything in that other than just those numbers you have gone through that would account for that lower end of the range?

Cody, those would be the two, if you will, hurtles we have to overcome in transition.

We do have, I will remind you on the depreciation, we expect about $30 million working in our favor going from fourth quarter to first quarter.

But the tax rate and the profit-sharing are the two main points that we will be confronting as we move into the quarter.

And then you talked about prices and utilization rates.

Can you maybe handicap those, as you look at Q4 what kind of acceleration have you seen in prices?

What kind of acceleration have you see in utilization rights?

Where are they sitting today in Q4?

And where do you see them over the next three to six months?

Cody, I would -- we're not really breaking out our utilization rates, but I would try to add some color to it for you to remind you, again, that on the analog side we continue to have open capacity in our advanced lithography consistent with our foundry strategy, to the extent that we need more than our internal capacity -- we source that from the outside, which we have been doing.

So our advanced lithography is pretty full.

In the commodity spaces, we own all of our capacity on that.

It is an old factory, old equipment, very depreciated.

So very low cost to operate that.

So from a my pricing standpoint, we began to see some pricing move in our favor in fourth quarter, and it is continuing to hold nicely.

And we're finally seeing that begin to turn in the right direction in the commodity spaces.

From the high-performance analog standpoint, we're also seeing demands there's such that prices have always been pretty strong there, but they are even moving up a little bit in the high-performance analog area.

As far as many of the focused end equipment, we have our normal scheduled price reductions that we do on advanced basis with our customers as we travel through each of the quarters.

So overall, again, I think what we're anticipating is that our opportunity for profit expansion should outpace our opportunity for revenue expansion as we move forward.

Cody, one thing I would just add, and maybe reinforce what I believe Kevin said earlier, was depreciation is really going to be working in our favor, frankly, probably over the course of the next couple of years.

And this is pretty different than, for example, if you go back to the year 2000 and kind of the peak environment we were in back then.

We had depreciation, I think, in the year 2000 that was increasing 20 percent, going up another 30 percent in '01.

With depreciation in '04 expected to be reasonably even with what it was in '03, as revenue grows, that depreciation line will be a contributor to profit margin.

And at least based upon our forecast today, we would expect depreciation levels at an absolute level to decline further in '05.

So again, for the next couple of years depreciation is probably going to contribute to margin expansion versus maybe where it was last cycle.

Ty Nodyan (ph) of Susquehanna.

I was just wondering in terms of the handset, although we went through a very strong period in fourth quarter, I was just wondering if you have any color into the Q1, and also the inventory that you see in the channel today?

What I can say from a component inventory perspective is that we believe most of our customers are at, or below, their targeted levels.

From a handset channel inventory of end equipment handsets, we don't have any direct visibility into those channels.

I guess I would refer you to our customers there.

As we have said before, probably the most significant color as to what drove handset growth was just the trend toward advanced feature handsets, where these handsets tend to the chock full of TI parts.

So whether it is a 2.5 G component, or whether it is an OMAP processor, the trend toward higher featured handsets is working in our favor.

In addition, chipsets where we get to sell a more broader range of components into the handset, as that revenue has expanded with strength in the ODM customer base, that has been a benefit to TI as well.

And also we have been hearing that out in the channel there is a shortage for the power management.

I was wondering if TI sees any of that, and if you see any impact -- impact your revenue or bottom line going forward?

I believe we are pretty well-positioned to support power management products into the handsets.

So if you're aware of those opportunities, just send them our way, please.

Just one last question.

In terms of like the Alcos (ph), I was wondering how you see that as compared to your DLP?

And do you see that as an impact in the near term?

I'm sorry, I didn't -- did you understand the question, Kevin?

Yes.

Go ahead.

On the low cost for imaging for television versus DLP, obviously we look at every competitor as a viable person to pay attention to.

But right now our DLP is very well accepted in the marketplace.

It is growing very well.

We have enjoyed not only very strong market acceptance through the Samsung products, but a number of other major manufacturers have introduced models last quarter and this quarter.

Among the names would be RCA, Thompson and Optima and a few others.

One of the things that I would just add is that the whole projection for TVs is a complicated system solution, and not just a single part.

And it takes time for the systems to be developed for that.

But we feel very confident about the position of DLP today and going forward.

I would just add that DLP actually has some just very fundamental advantages over low cost technologies, specifically the switching time that we can move the mirror.

And why that matters, as Kevin was saying, in system costs is it means we can implement a DLP based television with a single DLP devise, whereas in low cost technology basically requires three low cost panels to implement a digital television.

So again, we believe we have some fundamental performance advantages that translate to cost advantages inside of these televisions.

Adam Parker of Sanford Bernstein.

Can you talk about your PC peripheral business at all?

How has that been growing relative to your overall business versus your competitors and so forth?

And what are some of the recent dynamics you have seen in the PC space from your angle?

What I would say is that I can talk about what we're seeing.

I will let you do the translation to how we're doing versus our competitors.

But certainly what we sell into, call it the PC and peripheral space, is probably more weighted toward peripherals then the PC itself.

And what I would say is this kind of mix, just from the standpoint of seasonal trends, in the PC itself connectivity products, power management products, have done -- demand was very strong in the fourth quarter.

And that I would translate to comparison to the third quarter.

In areas like hard disk drives, we mentioned servo preamp revenue was strong.

In areas like printers, we saw a sequential decline, but that is pretty normal as a seasonal pattern for third to forth quarter.

So I think that space I would describe in general it did well, and probably overall did above the seasonal norm.

Okay, above.

How are separately here -- can you just talk a little bit more about the wireless business?

I know you discussed your several moving parts, and there is a lot of offsetting factors.

But can you help us out with what is embedded in your guidance in terms of units versus pricing for the quarter?

Or maybe an other way of asking this, should we assume that wireless units are therefore going to decline less than normal seasonality, given their importance to your revenue?

What I would say is that the wireless business overall probably, like many of the different pieces, is expected to continue to do well.

I don't want to try to forecast the unit versus overall trends in first quarter, because that is just a granularity that we don't what to provide in terms of our outlook.

So we're really try to keep our outlook at the semiconductor level, give that guidance, and not try to get specific expectations on all the different moving pieces.

The other thing I would add, Adam, is that if you look at our wireless revenue, normally the forth to first transition is down a few percent, which is probably more muted than what handsets would be overall.

That's true, Ron, but historically you've also had a more muted upturn in Q4, which you don't seem to be experiencing now.

So I was just wondering if there's a different shape to it, or if you're just going to beat all the time on wireless?

Time will tell on that one.

Thomas Thornhill of UBS.

Ron, in the wireless space there seemed to be several drivers that are giving you leverage, the transition to 2.5 G, the OMAP and the ODM chipset business.

Can you rank quarter their impact during '03?

And then also give us some idea of the rank order of their impact in '04?

I am trying to get a picture of -- you know, 2.5G. drove things in '03.

Will it continue in '04, or will the real driver shift more toward ODM or more toward OMAP?

I think I would agree that in '03 2.5 G. was the -- was probably the most significant piece, or driver, of wireless revenue growth.

Chipsets helped, OMAP helped, but 2.5 G was the most significant.

OMAP, as you probably understood, came on more in the second half.

So just by the fact that it ramped later you would expect, even if it were to carry the same, call it, current run rate into '04, it is going to a bigger factor in growth in '04 likely than it was in '03.

Chipsets, I think we commented that chipsets were relatively weak both in second quarter and in third quarter, as customers in the Asian market were working through excess inventory of handsets and things like that.

So to the extent that we don't have the inventory related depression in '04, I similarly would expect chipsets to be a bigger factor in '04.

I don't have a specific forecast on what our expectations are for 2.5 G, although it certainly has room to continue contributing to higher TI content for phone on average.

And we would expect that it will be a factor in '04 as well.

But if you're going to rank order those for '04, what would the rank order be?

And what percentage of the mix is ODM versus OEM at this point?

I don't want to try to provide the rank order projection in '04.

We will just have to see how that develops as it goes across the year.

Chipsets last year in '03 were probably right around 20 percent of our wireless revenue, Which is a pretty good representation of the ODM part of our customer base, as most of those chipsets, although not all, most would be folded into the ODM customer base.

So that is probably a good proxy for the ODM percentage of revenue anyway.

Nimal Vallipuram of DRKW.

Most of my questions have been answered, but I have one question.

Kevin, can you discuss what the peak margin is going to be for TI in this cycle, vis-a-vis the last cycle?

Sure, Nimal.

From the peak margin standpoint, if you recall during the last cycle we reached a peak margin on a GAAP basis of probably about 22 percent.

And while our revenues continued to grow, our margins didn't expand beyond that.

We were probably mainly driven by the fact that we were invested heavily in capital, depreciation was growing, and so on.

What is going to be different on this cycle here is that our peak margin should be able to exceed our prior peaks once our revenues -- our revenue passes our prior peak, because we don't have the hurtle of the depreciation.

In fact, our depreciation is going to be even lower as a percent of revenue as we go through this cycle.

So I think to summarize that, Nimal, the peak margins should be higher than they were previously, and not capped like they were previously as our revenues grow.

But we're not really forecasting the exact amount as to what that will wind up being.

It should be a few points better than where we are at right now, or where we were at rather.

Is it at all possible to give us an idea when you expect that to be happening any time, near time, or mid-term, or longer-term?

We are ready enjoying a pretty rapid growth or expansion in our margin rates quarter over quarter.

Perhaps the last missing ingredient that we're seeing right now, which is beginning as I mentioned a little earlier in the call, was that commodity pricing -- last time at the peak, commodity pricing was also at its highest level.

This time commodity pricing is just now coming off its floor and beginning to show some strengthening and improving.

So that is one of the remaining elements to contribute to our margin expansion going forward.

And I think you would probably be better positioned than I would on trying to forecast how those prices would move in the commodity space, Nimal.

Mark Edelstone of Morgan Stanley.

A question relates to distribution.

What was the growth in POS sequentially in the quarter, and did inventories do quarter to quarter?

Mark, distribution revenue grew -- I'm sorry, let me break it down.

Resales grew revenue, which would reflect our shipments into distribution, grew even faster.

So the shipments into distribution, or the revenue from those shipments, was up pretty consistent with our semiconductor growth rate overall.

So from the inventory standpoint, the distributors did put in place some inventory, which reflects, one, that I'm trying to address specific areas where they had gotten down actually into shortages.

And then secondly, just reflect their view of a more robust market going forward.

So again, shipments in exceeded the resales somewhat, and distributors put in place some proactive inventory.

And I'm sorry, Ron, I missed which of the categories actually grew in line with the Company?

Was that the POS or was that your own shipments into distribution?

Our shipments into distribution.

Then do you have the relative growth that was in POS for the quarter?

I don't have that specifically, other than the revenue, or our shipments in grew about the same levels as overall Semiconductor.

Joe Osha of Merrill Lynch.

If I can go back to this fall through issue for all of 2004.

Should I still be thinking -- obviously there is some quarter to quarter variability -- should I still be thinking about that gross margin fall through in the neighborhood of 70 percent?

And secondly, you did give me a R&D target.

Can I get you to talk to what we might try and manage SG&A to?

Thank you.

Yes, Joe.

On the fall through, as you have observed, it bounces around from quarter to quarter.

But you're probably not that far off from what you are projecting there when you take a look at what it would average during the course of a cycle as they go through the upturn for 2004.

On the SG&A, we expect to continue to manage that very tightly.

To the extent that we let it grow it all, it is just going to be on the marketing costs.

So that we continue to (indiscernible) to gain shares with our customers.

I would remind you that on the profit-sharing front, I mentioned there was a $60 million quarter over quarter accrual in profit-sharing, and that is roughly equally distributed between cost of revenue, R&D and SG&A.

So that alone during the course of the year will bump up our SG&A.

And the only elements of SG&A we intend to let increase would be towards marketing efforts.

Is there some percentage of revenue that you're managing towards?

You have talked about a target.

Is there something a little more specific I can work with?

Yes.

We're right now at about 11 percent, and we kind of like the number being low like that.

So I would not expect it to be exceeding that going forward.

John Lao of Bank of America.

Focusing in the broadband semiconductors, in terms of the wireless LAN migration, how fast do you think the migration will move from the 802.11B standards to the G and multiprotocol standards?

And I have a follow-up on that.

I don't know that I have a specific time line, but I would say certainly the most aggressive growth we're seeing in wireless LANs probably since the second-quarter timeframe is on the multi-mode, or on the G products.

And I think we expect that B will be around for awhile.

But clearly the market demand is strongest and going fastest for multimode G related product.

As a follow-up, in terms of the market, or additional catalyst, the consumer electronics market has been experiencing some problem using the G standards.

Do you think the multiprotocol will be the way for those home appliances, as you are focusing on your design efforts?

Oh, boy.

I don't know that I have a perspective on that, John.

I think as we're saying, the G is being rapidly adopted, as well as multimode.

I would suspect that multimode from the standpoint of A/G will be driven more by the enterprise as opposed to the residence or the consumer side of things.

But that is probably about as far as I would do.

William Pal (ph) with Friedman Billings Ramsey.

Mr. Pal, if you are on a speaker phone, can you please unmute your line?

We will move on to our next question, which is coming from Ambrish Srivistava of Harris Nesbitt.

A question either for Kevin or for Ron.

On pricing we heard a lot about commodity pricing.

I just wanted to recalibrate my thoughts as to what percent of your business do you consider commodity?

And then just to follow up on pricing, you talked about commodity and you talked about high-performance, what about some of the faster growing end markets, specifically DLP?

And what should we be looking at in terms of trends, whether Q over Q, or year-over-year?

Commodity I would describe as just a little under 10 percent of our Semiconductor revenue today.

And that includes standard logic products, some standard linear devices, and you probably can throw some of the display drivers in that category as well.

From the standpoint of DLP, to start with looking at the markets that business is focused on, if you look at, for example, the data projector market.

That market is growing at about a 35 percent cumulative annual growth rate.

And it is expected to continue at that rate for some period of time.

If you look at TI's performance last year, I believe in fourth quarter '02 we had about 20 percent of that market based on DLP.

By the end of the third quarter, I think is the latest date I saw, we were right around 30 percent, a little bit over 30 percent.

So it is a fast-growing market, and one in which TI's position is expanding.

That bodes well for 2004.

If you look at the other probably major margin segment and emerging market segment would be the television market.

And in that space, if you look at the number of greater than 40 inch televisions, last year in '03 it was about 5 million units.

It is expected to be about 6 million units in '05.

DLP marketshare late in the year was around 5 percent.

And that number is rapidly expanding.

If you go back to earlier in '03, we would have been in very, very low single digit percentages.

So again, we are participating in high-growth markets with DLP technology, and our penetration in both of those markets is expanding.

If you look at last year, revenue growth was right around 50 percent for DLP, '03 compared to '02.

So hopefully that will help you out, Ambrose.

Do you have a follow up question?

Okay, let's go to the next caller please.

Chris Stanley of J.P. Morgan.

Just a couple of quick clarifications.

I know you don't talk about utilization rates, but is it safe to say that is going to be the biggest incremental driver of gross margins this year?

Chris, as we use our analog capacity, that is certainly going to be a significant contributor to our gross margin expansion.

And in addition, as we commented on commodity prices, as it begins to continue its move upwards, that will also be a large contributor.

Can you delineate which would be greater?

No, Chris, I don't think that I have enough insight to be able to really give you a meaningful answer to that.

That's fine.

Can you give us an idea of when you think utilization rates should peak, or at least into the mid-90s, where you need to really start adding capacity on the analog side?

No, we're not really forecasting beyond the current quarter right now, Chris.

We're pretty comfortable with the capacity in place lasting for quite awhile.

Since those two questions went nowhere, can I get one more?

We will give you one more, Chris.

Try again.

On the lead time side, obviously they're going out.

Can you give us an idea of when you think you will be able to stabilize lead times, and maybe even bring them back down a little bit?

Lead times are -- actually on most of our proprietary products, I would describe as reasonably stable now.

In the commodity area we have probably a range of 4 to 16 weeks with those that are most extended based upon particular package types.

We are putting in place additional assembly test capacity to bring that down.

And our goal would be to have it down, say, everything in the 4 to 12 week range here very shortly.

I don't have a specific week to tell you where we plan to have that done, but we're trying to get it done here very quickly.

Would that be this quarter or Q2?

That would probably be this quarter or Q2, right.

Clark Westmont of Smith Barney.

As far as R&D and your $2 billion budget for this year, could you rank priorities in there?

And also let's say if there is upside to sales growth this year compared to your current expectations, would you be raising your R&D budget?

Aside from compensation issues, would you be raising your R&D budget as sales -- if sales would come in higher than expected?

Clark, let me take the last question first on the R&D.

Right now our projection is not to change what we've got in there.

We believe it is a manageable number for where we see the Company going.

So it is difficult to imagine that we would be making a significant change to that R&D plan right now.

As far as where we are spending the $2 billion budget, it will be distributed obviously into our wireless market, being our biggest source of revenue from an end equipment standpoint.

We will continue to invest very actively in our high-performance analog, in our broadband, and also bring up our advanced manufacturing lithographies in our 90-nanometer and early 65 nanometer.

So it will be fairly well spread across an assortment of activities.

Actually it is similar to what we have been doing the last couple of years.

And a follow-up on your Capex plan could you -- so DMOS 6 and I imagine and some others -- could you just give us time line or ramp rate for DMOS 6 and any other identifiable parts of the Capex plan?

Well part of that capital will, of course, go into DMOS 6.

That is where we're putting the 90 and 65 later in the year.

So by the end of the year we will probably have DMOS 6 -- the clean room roughly half utilized by the time we get to the end of year with our current Capex plan.

Is that 15,000 wafers?

About 14.

And and our plan, Clark, on 90-nanometer is we plan to have production out of our Kilby wafer fab.

In first quarter we will be qualifying DMOS 6 in, call it, first half, and anticipate full production on that line that Kevin referenced as we then move into the second half.

Ben Lynch of Deutsche Bank.

In the past year you have sort of commented on how much chipsets was of your wireless rev in a given quarter.

Could you give us a rough feel for what it was in the third and fourth quarters please?

We don't break it out by quarter, but what we do say is for the year it was about 20 percent.

It probably would have been a little more weighted in fourth quarter than third quarter just because third quarter was still depressed as customers were depleting handset inventory there in Asia.

But we had a nice rebound again in the fourth quarter.

Are that sort of rebound, which you had in the fourth quarter, are all the other factors in place which drove this delta between the actual Q4 and normal seasonality?

Are they all still in place for the first quarter?

I'm not sure exactly what factors that you're referring to.

Well, you know the mix.

Basically you're saying it is more advanced products and stuff.

Right.

I think what drove TI's revenue growth was a pretty robust market overall in terms of units of handsets being sold, the weighting of those handset sales toward advanced featured handsets, and TI's high penetration inside of those handsets.

And if you look at carrying forward into first quarter, probably the big variable that would potentially change, naturally, would be units as you come off of a holiday strong holiday period in the fourth quarter into one that is probably less robust in first quarter.

But certainly those other two factors, consumer appetite for advanced featured phones I don't expect is going to slow down at all.

And certainly our penetration is only getting better.

So, Ron, if I just try to strip out the chipset effect, I could think maybe the delta in Q1 versus normal seasonality is not too different to what it was in Q4?

I will let you make that analysis.

I don't have any specific guidance for you in terms of what our wireless revenue will do in the first quarter relative to normal seasonality.

Again, we don't want to get more specific on our guidance other than at the semiconductor level.

I have a quick follow-up for Ron or Kevin again.

What much was foundries of your production in Q4 and calendar '03?

And how much do you think it will go to full year '04?

Just sort of rough order of magnitude?

Ben, the foundry was at about a 45 percent level of our advanced lithographies on the digital side.

You may recall that we source all of our analog internally.

And we probably see that kind of outsourcing percentage fairly close to that same level as we go through '04, maybe a little bit higher at its peak.

And again, advanced being specifically 130 nanometers.

So 45 percent of our 130 nanometer demand was outsourced.

Eric Gomberg of Thomas Weisel Partners.

Must my questions have been answered.

Just building on the last question though, you talked about the foundry outsourcing hitting a peak.

Do you expect that as 90-nanometer ramps at, say, twelve months from now your mix from internal versus external will be different than today?

By mix, I am presuming, Eric, that you mean the mix --?

Foundry versus in-house when it starts coming down again.

It may start coming down, but in fact, as we introduce new products onto 90-nanometer, it should start replacing some of the products that are being expired on the 130.

So it is difficult to say that it will necessarily come down.

It will be up-and-down as we go through the quarters.

But on average, it will probably stay close to that range we just talked about.

Certainly, we would expect that as we move products from 130 over to 90, and our demand for 130 nanometer goes down, that decrease will come primarily from foundries.

So as a percentage of our 130 demand, foundries will go down.

But again, they're going to ramp right up with us on the 90, so it is difficult to say how that nets out.

Probably pretty stable in total.

Did you have a follow-up question, Eric?

Yes.

Just a question.

In your strong fourth quarter bookings you had, was there discernible difference in any end market as particularly strong or weak?

The strength was across the board.

It was a very robust market virtually across every product line that we have available and sell to our customers in the semiconductor market.

Bookings -- the distribution of bookings was not dissimilar to our overall revenue mix probably in the fourth quarter.

Manish Goel (ph) of Neuberger.

I want to understand your earlier comment regarding the margins that you will have higher margins at the peak revenue if you were to exceed or match your revenue from the last peak?

I'm looking at my numbers and it seems like at about $3 billion you'll have over 25 percent operating margins.

Now to get to that number I'm not sure how you do that with 70 percent incremental margin if your R&D is about $100 million higher per quarter.

Could you explain that to me?

Manish, what I was trying to explain there is that during the last peak when we reached the 3 billion level, we peaked out at about 22 percent margin.

In fact, we hit that level at slightly lower revenues and peaked at that margin level.

What we expect on this next portion of the cycle is that we will not peak at 22 percent.

When our revenues get past our prior peak, we do not expect that to put a lid on our profitability.

We expect our profitability to continue to expand to get beyond what it was at the prior peak as the revenues continue to go above the prior peak revenues.

So when you address -- when you say peak, you mean higher than 22 percent?

Is that the right way to think about it?

Yes, that's correct.

I have one more question.

When I look at your guidance for next quarter's sequentially higher revenues, which end markets do you expect to grow sequentially in the first quarter?

If I look at cell phones and PCs, the too larger markets it seems like on a unit basis it is likely to decline sequentially, and which end markets are you expecting to grow that will offset those declines?

Manish, we're not providing any further color below the semiconductor level when it comes to our forecast.

I would just reiterate that we have seen robust orders across all of our products.

Some growing faster than others.

And that order pattern that we saw coming out of fourth quarter is continuing in first quarter.

And certainly, our shipments in first quarter don't necessarily correlate completely with the end shipments of our customers.

Certainly our shipments in will lead their shipments.

So part of what they are expecting in second-quarter will be reflected in our first quarter shipments.

Hans Mosesmann of Schwab SoundView.

Thank you.

No questions.

They've been answered.

Michael Crandall of Maverick Capital.

Okay.

It doesn't sound like he is there.

Let's go to the next caller please.

Quinn Bolton of Oppenheimer.

I just wanted to get a clarification on the gross margin change in the fourth quarter.

If I just back out the $48 million restructuring charge from Q3, it looks to me that you got about a 50 -- low 50 percent incremental gross margin fall through here in the fourth quarter.

I was wondering if you could review again what some of the changes were Q3 to Q4 that led to that slightly lower than 70 percent incremental gross margin, once you back out that restructuring charge?

Quinn, I will be happy to go through that again.

One of the main elements that you can take a look at as an example is depreciation.

It went up $14 million sequentially.

And that would be about 4 points of that fall through that you would have expected.

In addition, as I mentioned earlier, these things -- the fall through will fluctuate quite a bit from quarter to quarter based upon the timing of an assortment of things, various expenses, employee taxes, other types of spending that we may have going on.

And in addition, as we introduce new products that need to be qualified and ramped those numbers of products being introduced to the line from quarter to quarter will differ.

And so the costs associated with ramping those will differ from quarter to quarter.

So you have an assortment of things that occur.

A good example in the fourth quarter was depreciation.

So, Quinn, just in general that is why we encourage -- a lot of the things that Kevin is describing are seasonal type of considerations.

So for example, even though our depreciation in '03 was down from the '02 level, you saw trends like third to fourth quarter depreciation going up.

To get the true depreciation turn, if you look at it fourth quarter to fourth quarter, is probably a more meaningful number of the overall trend.

William Conroy of Sanders Morris.

A couple of questions.

First, Ron, can you give us a couple more details on the high-performance analog business?

And specifically can you give us any sense of how much of the growth is coming out of new products, however you care to define that, versus just increases in general for that particular sector?

Sure.

Again, from an end markets perspective, or product line perspective, it is pretty broad-based.

Some of the areas that I noted in the opening comments, maybe were weighted more so than others, but it is very broad-based.

From a new product standpoint, we believe new products are a big factor in our success.

For example, in fourth quarter if you look at high-performance analog revenue, about half of it came from products that have been introduced since the beginning of year 2000.

So what that tells us is that we're meeting customer needs for products that we're rolling out.

They like what they see from us, and they're running with them.

And it is turning into revenue growth here pretty quickly.

So certainly new products are a big factor in that.

Great.

Can you give us any sense of orders thus far in the quarter, or business in general thus far in the current quarter?

I don't have a specific order number to give you, but I would say orders are coming along nicely, and are completely consistent with the guidance range that we have provided here, as you might expect.

Clark Hughes of Fulcrum Global Partners.

Can you remind me just very quickly what you said about the high-performance analog growing quarter to quarter and year on year?

Let's see.

I don't know that we said, but I can certainly provide you that information.

The sequential growth -- actually it was pretty -- the numbers we gave for catalog DSP and high-performance were pretty evenly distributed between the high-performance analog and the DSP part.

But the high-performance analog sequential growth was 17 percent.

And it had 37 percent growth from the year ago quarter.

Usually at this time you kind of give the end markets split for the various end markets, if you've got that for how 2003 ended up between communications, computer and consumer and whatnot?

Can you give us those?

And can you outline whether or not DLP grew faster or slower than the overall Semiconductor Group?

Clark, I've got that for you.

Communications is about 45 percent of Semiconductor.

Computer, which is mostly peripherals, is about 30 percent.

Consumer and industrial reach about 10 percent.

And automotive is about 5.

And, I'm sorry, what was your question about DLP specifically, Clark?

The question on DLP was whether or not the DLP revenues grew faster or slower than the overall corporate number?

For the year 2003, DLP grew about 50 percent.

So it grew faster and increased somewhat as a percentage of revenue, although I would still describe it as about 5 percent of our semiconductor revenue.

Thanks, Clark.

Next caller please.

We will be able to take probably two more caller's questions, and then we're going to have to cut off for time

David Wu of Web Bush Morgan.

Twp quick things please.

Number one is, I was looking at capital spending.

The delta from 800 million, which you have sustained in '02 and '03, and going to 1 billion 1, where did that extra 300 million going to be spent on?

And the other thing about lead times.

Ron, you mentioned about they haven't gone or stabilized.

I remember the last time you talked about lead times and proprietary products, they were the four to six week range.

Should we think about lead times still in the four to six weeks range?

David, on the capital expenditure, as we look into 2004, the large share of that $1.1 billion will be spent on our 90 nanometer and 65 nanometer advanced lithographies.

And another large portion of that will be spent on our assembly test sites as we expand capacity there.

As we have been gaining market share, our volumes are going up, and we're keeping our capacity in front of those volume requirements.

So that is where you'll see that 1.1 billion being spread in 2004.

From a lead time perspective, I guess it is going to vary depending upon product line.

And for example, if it is a standard product area like high-performance analog, where we carry inventory versus if we're building custom products.

What I would say to try to give you a little give a little more granularity, if you look at areas like high-performance analog, catalog DSP, so proprietary standard products, I would say lead times generally range four to six weeks.

For custom products, where it is going to represent more manufacturing cycle time, it would be in the eight to fourteen week range.

And then as I said, commodities are four to sixteen with our goal to bring them down -- to have them all under the 12 week range.

Tim Luke of Lehman Brothers.

My question really relates to the CDMA opportunity.

And obviously you are suggesting that it is on playing and you have some active customers.

Can you give us a sense of when you might be in a position to confirm or announce some customers in that arena?

And then more broadly, as you move into WCDMA could you give us a sense of the positioning there and how you see the opportunity unfolding?

Just first of all as far as our standard product, where we mentioned we have active customer engagements, as you have perceived, we're not disclosing the specific customers and that is, you might say, at their request.

And, frankly, I don't know that they will be interested in having TI or FT promote those engagements in terms of more specifics, until they're ready to rollout phones.

So I would not look for specifics, at least on the existing engagements, until those customers are ready to rollout phones.

On WCDMA, you're going to see kind of different directions of our engagement.

We're shipping revenue today primarily in the 3G WCDMA space in our application processor engagements.

We have what I would describe as custom OEM engagements that will ramp, and be part of our revenue growth probably even in 2004.

And then similarly we have standard UMTS products that we are in the process of sampling and expect to be in production in the second half '04.

So the whole 3G space is one where I would say that the revenues, specifically in the application processor area, is significant even in '03, but will continue to grow in '04 and probably even more in '05.

Any sense of the range of the business as a percentage of wireless that one might expect it to be in either '04 or '05 in either CDMA or WCDMA UMTS?

In terms of the CDMA modem, I would say -- let me star with '03, I would describe it as less than 10 percent of our revenue.

And again most of that would be what Nokia has in production in their hand -- their CDMA-based handsets today.

I don't have a specific number for '04 or '05 other than we expect that our penetration into that market certainly will be increasing, both because of Nokia's success, as well as our emergence of the standard project product line.

So unfortunately I can't give you more granularity than that in the outlook.

And with that, we will wrap up.

Before we end the call, let me remind you that a replay is available on our website.

Thank you and good evening.

Thank you.

And thank you, callers.

This does conclude today's conference.

You may disconnect your lines at this time, and have a pleasant day.