德州儀器 (TXN) 2003 Q1 法說會逐字稿

Please stand by for realtime transcript The Texas Instruments conference call will begin soon.

Good afternoon, ladies and gentlemen, and welcome to your Texas Instruments first quarter 2003 earnings conference call.

At this time, all lines have been placed on a listen-only mode, and the floor will be open for your questions following the presentation.

It's my pleasure to turn the floor over to your host, Mr. Ron Slaymaker.

Sir, you may begin

Good afternoon, thank you for joining our first quarter earnings conference call, with me today is Bill Aylesworth, TIs Chief Financial Officer.

In a moment, Bill will provide his perspective on TIs results for the quarter.

This call will last one hour.

For any 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 it can be accessed through TIs website.

A replay will be available through the web.

Before I turn it over to Bill, let me remind you this call will include forward looking statements that include risk factors that could cause TIs results to differ materially from managements current expectations.

We encourage you to review the Safe Harbor statement contained in the earnings release published today, as well as TIs most recent SCC filings for a complete description.

Beginning with the first quarter 2003 results, we are no longer including pro forma information as a supplement to the company's GAAP financial reports.

Accordingly, today's discussion and the earnings report issued today describes the company's result and expectations in accordance with GAAP.

After Bill reviews the quarter, we will open the lines for your questions.

Bill.

Thank you, Ron, and good afternoon, everyone.

First quarter revenue exceeded our expectations reflecting better-than-expected demand for TIs semiconductor products.

TI revenue increased 2% sequentially and grew 20% from the year-ago quarter.

Semiconductor revenue in the first quarter was up 2% sequentially and up 23% from the year ago quarter.

Sensors in controls revenue was up 5% sequentially and up 11% compared to the year ago quarter.

Educational and productivity solution or EMPS revenue declined 2% sequentially and declined 10% from the year ago quarter due to excess channel inventory that resulted from weaker-than-expected retail sales.

In semiconductor, DSP revenue in the first quarter grew 6% sequentially and was up 31% year over year.

DSP revenue benefited from a continued richer mix of 2.5 G products in wireless.

Although the dollar impact was less, DSP revenue also had solid sequential growth outside of the wireless market, including TIs catalog DSP products, as well as in markets such as automotive, DSL, and DVD.

TIs leadership market position in DSP across a diverse mix of vertical and horizontal markets should enable us to continue to grow our DSP revenue faster than the market.

In 2002, TIs DSP growth of 30% was more than double the 14% growth in the DSP market.

In fact, our DSP growth exceeded our competitor's combined growth by almost 10X.

Turning to analog, revenue declined 4% sequentially but grew 15% over the year ago period.

The sequential decline was mostly the result of OEMs specific declines in the wireless market.

TIs analog revenue for the printer market grew over 15% sequentially, as this market began to recover from excess inventory that resulted in a sharp sequential decline in our printer revenue in the fourth quarter.

TI also gained share in analog in 2002.

According to a recent report released by Dataquest, TI grew its analog revenue faster in all but one of the top 10 analog suppliers in 2002.

In other product areas, digital light processing or DLP revenue grew over 25% sequentially and more than doubled from the year ago quarter.

We also had sequential growth in micro controllers, RISC micro processors and basic products as well as royalties.

Standard logic revenue declined sequentially due to a weak pricing environment.

Compared with the year ago quarter, growth exceeded 20% in RISC micro processors, standard logic and royalties, with lower growth in basic and micro controllers.

Our wireless revenue was better-than-expected in the first quarter.

Essentially even with the fourth quarter and growing 39% compared with the year-ago quarter.

The transition to 2.5G products continues.

With these products comprising more than 55% of TIs wireless revenue in the first quarter, revenue on these products is up more than 150% from the year ago quarter.

With ASPs that are more than double those of 2G digital basebands, the higher 2.5G mix allowed our higher DSP revenue to grow sequentially, despite shipping seasonally fewer overall limits.

On the analog side of wireless, there is not a specific analog silicon content increase with 2.5G products and both units and revenue declined sequentially.

Our revenue and wireless chip sets expanded sequentially on higher unit shipments.

Chip sets are sold to a diversified set of customers, primarily original design manufacturers or ODMs.

Chip set sails result in higher content of TI, DSP, and analog products for handset, including the digital and analog base bands, power management, RF, and protocol stacks software.

Compared with the year ago quarter, TIs chip set revenue more than doubled in the first quarter.

Revenue for chips sold to wireless infrastructure customers also grew in the quarter.

Do you to 3G deployments in Japan and Europe as well as 2G deployments in China.

TIs gross profit was $862 million in the first quarter, or $39.3% of revenue. 3.5 percentage points higher than the fourth quarter, and 5.9 percentage points higher than the year ago quarter.

Semiconductor gross margin of 40% in the first quarter, increased sequentially due to a higher factory utilization levels and lower depreciation expense.

Average semiconductor factory utilization increased to over 75% in the first quarter, compared to 67% in the fourth quarter.

Higher volume in TIs 130 nanometer process along with increases in the manufacturing yields on this process, contributed to the sequential gross margin improvement.

Censors and controls gross profit margin was 35.7%, and EMPS gross profit margin was 49.5% in the first quarter.

Total operating expenses of $709 million in the first quarter were about even with the fourth quarter level with R&D down $4 million sequentially and SG&A up $12 million.

Operating profit was $153 million or 7% of revenue in the first quarter. 3.9 percentage points higher than the fourth quarter.

First quarter operating margin in semiconductor of 7.9%.

Censors and controls 24%, and EMPS, 19% of revenue.

For TI, other income and expense or &E, including interest income, investment gains or losses and other items produced income of $14 million in the first quarter.

Interest expense paid on loans was $13 million in the first quarter, down a million dollars from the prior quarter, due to the company's lower-debt level after a redemption of $250 million in convertible notes during the first quarter.

This redemption will lower interest expense by about $3 million in each future quarter.

Interest expense will also decline when an additional $127 million of debt matures in the second quarter.

The company's effective tax rate was 24% in the quarter.

Net income was $117 million in the first quarter or seven cents per share.

The quarter's results included $28 million of amortization of acquisition-related costs and a $10 million charge associated with the redemption of the convertible notes.

Total cash of $4145 million, was even with the fourth quarter and increased by $690 million from the year-ago quarter.

Cash flow from operations was $196 million, compared with $744 million in the prior quarter, and $296 million in the year ago quarter when we received a tax refund.

Capital expenditures of $132 million in the first quarter, were down from $236 million in the first quarter, and up from $120 million in the year ago quarter.

Accounts receivable of $1365 million dollars increased sequentially by $148 million due to stronger end-of-quarter shipments in the first quarter compared with the fourth quarter.

Compared with the year ago, accounts receivable were up consistent with the increase in revenue.

Day sales outstanding were 56 days at the end of the first quarter compared with 51 days in the prior and 57 days in the year ago quarter.

Inventory in the first quarter increased to $882 million, up by $92 million sequentially and $114 million compared with the year ago quarter, primarily due to semiconductor work and process increases in anticipation of higher second-quarter shipments.

As a result, days of inventory increased to 60 days at the end of the first first quarter from 52 days at the end of the prior quarter and 57 days in the year-ago quarter.

TIs orders in the first quarter were $2295 million, up 10% sequentially, and 20% from the year ago period.

Semiconductor orders of $1914 million, increased 9% sequentially and 24% from the year ago period.

The book-to-bill ratio for the semiconductor business was 1.03, compared with .96 in the fourth quarter.

Turning to our expectations for the second quarter, revenue should grow about 7% sequentially.

Semiconductor revenue should grow about 4%, sensors and controls should be even, and EMPS revenue should about double with the seasonal retail stocking of educational calculators for the back-to-school period.

Operating margins should be about 7% of revenue and earnings per share should be about eight cents plus or minus a few cents.

Gross margin and operating expenses as a percentage of revenue are expected to be about even with the first quarter levels.

Operating margin is not expanding with revenue growth in the second quarter due to charges associated with restructuring actions in both TIs sensors and controls and semiconductor businesses.

Sensors and controls is moving certain production lines from Attleboro, Massachusetts, to other TI sites that are geographically closer to its customers and their markets and are more cost efficient.

The semiconductor action will effect jobs and manufacturing operations in the U.S. and the international locations, as those operations continue to become more productive with fewer people.

Ultimately, about 1250 jobs will be affected in the two businesses.

In the second quarter, the company will incur a restructuring charge of about $40 million.

Impacting EPS by about two cents.

About $60 million of additional charges associated with this restructuring will be distributed over the quarters in which the jobs are eliminated through the end of 2004.

This action in total will result in annualized savings of about $80 million, $45 million in semiconductor and $35 million in sensors and controls.

Amortization of acquisition-related costs is expected to be about $26 million in the second quarter, or one cent per share.

Other income and expenses is expected to be about $25 million of income in the second quarter, and interest expense on loans should decline to about $11 million.

For the year 2003, we continue to expect R&D to be about $1.7 billion.

Capital expenditures to be about $800 million, and depreciation to be about $1.4 billion.

In summary, we believe the company is showing results from continuing our investments through the downturn.

One example is our market share gains for 2002 in both DSP and analog.

Another is our higher margins, reflecting the leverage cost structure that characterizes TIs in-house manufacturing operations.

And we continue to make the difficult decisions that will structurally enhance our profitability for many years ahead.

With that, let me turn it back to Ron.

Thank you, Bill.

At this time, I would like the operator to open up the lines for your questions.

Due to today's time constraints and in order to provide as many of you as possible an opportunity to ask your questions, please limit yourself to a single question.

After our response, I would provide you an opportunity for an additional follow-up.

Operator?

Thank you, sir.

The floor is now open for questions.

If you have a question, please press the numbers 1 followed by 4 on your touch tone telephone.

We do ask that if on a speaker phone, please utilize your handsets or provide optimum sound quality.

Our first question is from Joe Osha of Merrill Lynch.

Sir, your line is live.

Boy, I feel like I won the Academy Award here.

Can you please talk, Bill, for a little bit, about what, say, let's look at the fourth quarter of this year, if you want to look at the year in aggregate, leaving these one-time charges aside, what kind of gross end operating margin you think you can manage this company to?

Joe, I would first like to point out that at the Academy Awards, our DLT projectors were very prominently in display so you're a winner.

I'm glad I said it then.

We feel that these charges and on the the savings that will come to us in the manufacturing area just give additional credibility to our objective of getting our gross margin above the level maintained we attained at the last peak.

At the last peak that was about 49%, this time by the time revenues get to $3 billion or so in the quarter, it will be certainly a couple of three points above that, and we continued to have the objective of managing our R&D tightly and our SG&A tightly as well.

Our objective -- object whiff revenue growth is going to be to continue to take advantage of our leverage cost structure and keep our operating expenses under tight control.

I can't predict when that $3 billion quarter will come, but the actions that we have taken, we think just improve the possibilities of our higher gross margins at that time.

Ron, may I follow up on that?

Yes.

In no way, Bill, am I asking you to bless this number.

Let's say we get to a $2 1/2 billion run rate by the fourth quarter of this year.

Can you give me insight to into what the margin might look at that revenue run rate?

Joe, pretty much proportional.

We would expect that as our semiconductor revenues grow, about 70% of that revenue growth will turn into gross margin for our other businesses which had slower growth.

We don't have as much cost leverage.

With most of the growth expected to come from semiconductor.

I think that would be a good model to use.

If you look at the first quarter transition, where depreciation went down year over year, the transition was much more than that.

Our gross margin went up something like 1.25 times the revenue delta, fourth to first.

Overall, about a 70% gross margin incremental gross margin is the right guide for the semiconductor business.

Thank you, Joe.

Operator, next caller, please.

Thank you, our next question is from Adam Parker of Sanford Bernstein.

Yeah, hi, I just want to make sure I heard you right.

Did you say you finished at 75% over utilization this quarter?

Yes, Adam, our utilization in the first quarter, really, on average for the first quarter was over 75%.

Do you expect that to rise again in Q2 from the 75% level?

Slightly, yes.

Should be up slightly.

And is that in combination with -- what is your expectation of inventories sort of rising or falling in the quarter?

Generally flat, maybe up some more, by the time we get a view of the third quarter, a little through the quarter.

So, probably pretty flattish on inventories in the second quarter compared to the first, Adam.

Okay, great.

And sorry, one last thing.

Are any big OEMs in any of your verticals, you know, extending the lead times here at this point and in the quarter.

No.

I don't see any examples [INAUDIBLE] pushing lead times out, I guess you're question is.

Yeah, I'm saying that more optimistic that their conditions are good enough to merit ordering farther in advance from you.

Oh, okay.

I guess from our point of view that would be pulling in lead times or trying to build more, more backlog.

Not really anything that we can say specifically.

Certainly as we look at our inventories in general and our customers inventory, they appear to be in pretty good shape.

And so, we think in general, our customers are ordering to their consumption patterns and we don't see any particular areas where their trying to build in an extra pad any place.

Don't see any particular evidence of that.

Okay.

Nor are they giving us extended visibility, Adam.

We're at an environment where lead times are pretty much short across the board, especially in standard product areas, high-performance analog catalog DSEs, lead time is four to six weeks now.

Right.

Commodity products, probably in the two-week range.

And then the custom products are more manufacturing driven by 8-10 week type range and customers are generally placing backlog consistent with those lead times.

Nobody is in a position where they're --

Think visibility better than it was a couple of quarters ago?

I would say it's consistent with what it was over the last few quarters.

That visibility is pretty on near-term.

Okay, it hasn't improved in terms of people saying we feel more confident about the back half and, therefore, we will start ordering appropriately.

If you're talking back half of the year.

Yeah.

As opposed to the back half of second quarter, that would be correct.

Okay, thanks, guys, appreciate it.

Next caller, please.

Our next question is from Jim Jonjohan of CIBC World Markets.

Hi, guys.

Just one quickie, and that is, was there any currencies benefit in the quarter?

No, Jim.

Nothing material, we're pretty well balanced in currencies as it reflects on our balance sheet.

We get, in some cases, a little more revenue in some cases, a little more revenue in places where we bill in local currencies other than the dollar, but that's offset by the fact that our expenses go up in some places in our regional marketing, for example, with higher, with the weaker dollar.

The overall effect was not material for us.

On the smaller division, on the broadband side, how is that business?

Was it up sequentially and was it profitable?

Broadband for us was down about 2% sequentially, but well up from a year ago, up about 47% from a year ago, we're clearly still investing in broadband because of the opportunities that we see there over a long period of time, but we're particularly encouraged on a quarter-to-quarter, on a sequential basis, DSL was up for us, cable was up slightly.

Wireless LAN was seasonally down offset somewhat by the first [INAUDIBLE] base band as well, a year-over-year basis, all of the parts of our broadband business were up strongly to have the revenue up 47% from the first quarter of 2002.

Right, in broadband last quarter, you gave us a Y5 number, a growth number.

Is there any numbers you could give us on Y5?

Only in the first quarter, we were seasonally down as we would have expected, but that we would expect continuous growth to pick up again after that seasonally for the rest of the year.

Okay.

Thanks, guys.

Thanks, Jim.

Next caller, please.

Our next question is from Chris Danley of J.P.

Morgan

Hey, guys, thanks.

Nice quarter.

Just real quick, a clarification on the inventories and then my question.

Did you say you expect your own inventories to increase quarter over quarter?

Chris, I would say flat to up slightly, our own inventories.

And it seems like the inventories will be the highest since early '01.

Did you do you feel comfortable with that and what would be a long-term inventory plan for the year?

We feel comfortable in the environment we're in that Ron was describing of our customers really demanding our standard products and commodity products very short lead times, which is really requiring us and it pays for us to make the investment in work in process in the bank of semifinished products so we can package the product to their specifications when the order comes in.

There has been no indications of that environment changing in the near-term, and, frankly, it's a good marketing advantage to us to make sure we have that inventory.

We monitor it carefully, take all the required reserves against it, but we think that that is a good marketing advantage for us.

In our custom areas, as our revenue -- what revenue growth first to second, we began building some of that work in process in the first quarter that may continue to some degree in the second quarter and, again, we feel confident about building that invenator to custom orders.

So does that mean you think pricing is going to start to flatten out here if you're building inventory?

Well, I'm not sure that's exactly the correlation.

It's very rare we have issues ending up with product that's below market in terms what have happens in some of the commodity memory chip areas.

The fact is in our most commodity products, pricing is continued to be weak, but there is some indication the rate of decline, pricing has slowed somewhat in the standard logic area, for example, and in, for example, in broadband, there is certainly continues to be pressure on pricing but we're building more and more integrated chips there, allowing us to effectively approach that market.

Thanks, Chris.

Next caller, please.

Our next question is from Ben Lynch of Deutsche Bank.

Yeah, hi, Bill.

Just trying to understand the utilization question again and, therefore, the fact to an impact on gross margins.

Semi's revenues were up 2% in Q1, you mentioned that yields, in general, were up, so you're basically getting more units out of your existing capacity, so it feels like a lot of the utilization increase would have come primarily from this inventory.

The inventory thing.

Why do you feel it's okay to have such a big Q-in-Q increase in inventories?

We feel it's okay because, Ben, our revenues, we believe, in semiconductor will increase 4% first to second quarter and because we need the additional dye stock in our standard products.

Also, if you look at the end of the first quarter, our total inventories as a company, days of inventory compared to a year ago, they were up three days.

That's from 57 to 60 days.

That's not anything of particular significance to us.

Are customers putting pressure on you to just improve even more the availability of parts?

Generally speaking yes.

That is the lead times that Ron described, four to six weeks for standard products, as little as two weeks for commodity products.

We think we're very competitive in that area, but that's as short as lead times have been in that area and in the, in custom-product areas, we want to make sure that we have at least in places where we can start wafers and finish to the customer's satisfaction.

We're comfortable about the ability to finish that.

Great, thank you very much.

Thank you, Ben.

I would also add that TIs inventory days are at 60, compared to most of the semiconductor industry that 60 days of inventory is still a very aggressive level of inventory management, so, I think our relative inventory level is still pretty aggressive.

Next caller, please.

And, in fact, that also reminds me, in days of the inventory anyway, since that's a numerator/nominator with our lower depreciation in the first quarter, the cost of revenue goes down, that tends to make that ratio not quite as comparable on an apples-to-apples basis in terms of the sequential days of inventory and even the year-over-year days of inventor with lower depreciation in the base.

Good point.

Next caller, please.

The next call is from Michael Mestia of Credit Suisse First Boston.

Thanks, guys.

Gross margin guidance for [INAUDIBLE] is that more of a function that utilization is looking flat and you're not secure sure about Q3 at this point?

It's more a factor, Michael, of the charge we're taking since we're reporting here on a GAAP basis that a significant amount of the $40 million charge roughly 2/3 of it, I think, will impact cost or revenues in the second quarter.

So that's a point, plus a margin all by itself.

Michael, I would suggest, that on the semiconductor revenue you can are start by assuming a 70% incremental gross margin, kind what have we generally guided over time, but then account for, as Bill indicated, a restructuring charge on top of that which flattens out the gross margin level on a sequential basis.

Did you have a follow-up question?

Yes, real quick.

In the wireless handset market you have been moving more into the ODM and standard market.

You have a feel between the basic base, wireless handset chips and the standard base?

Yes, Michael.

In the, our ships have revenue now has been running, last year, it ran about 20% of our wireless revenues and that probably increased a little more in the first quarter, with mid-single digit growth, quarter-to-quarter growth on chip set.

So, our chip set revenue has been growing from that point of view as a percent of our total wireless revenues.

So no standard product outside of the chip sets?

Is that fair?

I think that's fair.

As we would, that's right.

Even our chip set revenue is really built, I'm not sure you call it standard product exactly.

Every one of our ODM customers has, you know, different designs, different configurations, so none of it is exactly generic, but chip stat is, you know, packaged more for those particular customers whereas in our OEM customers will frequently shipping only a digital base band, maybe only analog products.

Thanks.

But Michael, there is -- we're not religious about this.

If a customer wants to buy a -- part of a chip set or a complete chip set or one chip, other chip set, we will on we will make the sale, so, a chip set in some cases can include a digital base being analog, power base, power management, RIF, TI software or include any subcomponent of that, and that is in contrast, I think, the point you're making to what we generally sell to the big OEMs will be custom products to their specifications.

Next caller, please.

Our next question is coming from Andrew Ruth of Goldman Sachs.

A couple of quick questions on DSP analog.

Great quarter on DSP.

Can you split out the sequential growth by handsets versus the non-handsets?

I think, Bill, you mentioned you were gaining share in analog as well, but I think analog, you mentioned it was down 4% sequentially, which is a little bit softer than most of the analog companies we would follow.

You can let us know what you're looking at to look at market shares on the analog side?

Yes, Andrew, for DSP, the other areas that grew an additional to wireless for us, wireless DSP did grow sequentially.

Our catalog DSP grew 5% sequentially, something like 40% year over year of growth.

In addition to, that we have some DSP growth in auto motive and in DSL as well, quarter-to-quarter.

In analog, our market share comments were specifically 2002 on a year-over-year basis where we clearly gain share, we don't have enough data yet in, for the first quarter to be definitive, and even though our analog revenue declined about 4% sequentially in analog, we think that's basically in the ballpark what have the market is going to do in the first quarter as well, and for example, our catalog analog area, high-performance analog was down slightly in the first quarter sequentially most of the analog decline was driven by some specific analog base band customers and wireless where we had declines fourth quarter to first quarter.

Okay, what proportion is high-performance right now?

High-performance analog for us is about 30% of our total analog revenues.

Okay.

Andrew, let me make it very clear.

In wireless, both DSP and analog are units declined sequentially and that should be no surprise, given what handsets seasonality is.

In the case of DSP, our revenue increase, because of the higher mix of 2.5 G with higher ASPs, don't get that same factor that applies on the analog side.

Fair enough.

Thank you.

Did you have a follow up?

Perfectly.

Okay, thank you, Andrew.

Next caller, please.

Thank you, our next question is from David Wu of Wedbush Morgan.

Yes, good afternoon.

You can amplify -- in the third quarter what kind of products do you see sequential growth, that 4% number, and on the wireless handset, when do you think you will get to 75% 2.5G?

David, for the second quarter, which I think you're asking about, we really see by market the ability to have sequential semiconductor revenue growth of a few points in wireless because the trends both for units in the second quarter and the continuing trend of 2.5G will help us there in PC and peripheral markets, conductivity, printers and in digital consumer products again, like DVD as well.

We would also expect continued sequential growth in broadband in the second quarter, particularly in DSL and resuming sequential growth in a wireless LAN as well.

I don't have a particular time period for hitting 75% in 2.5G, we don't plan to try to forecast that by quarter by quarter, but the overall trend there is quite clear of continuing benefits that we'll get, and the way we're running now in the first quarter with about 55% of our -- of revenues in wireless coming from 2.5G, that is less than half of that in units, so, you know, 20, 25% of our wireless units are going into 2.5G in the first quarter, so we think there is a pretty long runway of continued growth there.

And the thing we don't know, offsetting that, is we would continue to expect, for example, chip set growth in 2G in certain markets as well.

Thank you, David.

Next caller, please.

Our next question is coming from Eric Domberg of Thomas Weisel Partners.

Good afternoon and nice quarter.

Given your receivables growth, it looks like business improved steadily as progressed.

I want to confirm that you didn't see any disruption with SARS or Iraq, that in itself, is that accurate?

Eric, that is accurate.

Our first quarter was predictably quite backend loaded, which is in contrast to the fourth quarter which tends to be more linear by month.

March was a significantly larger part than a third of the quarter, exactly what we would have expected from traditional seasonality.

And in effect, in the face of or despite uncertainties caused by the war and by SARS.

Yeah, I know, some of your competitors seem to be blaming.

Have those trends kept up thus far in April?

Can you just also give us a sense of what your relative semigrowth expectations are for DSP versus analog for the second quarter.

Yes, Eric, in April so far, both our orders and our revenue, I would say, are tracking our expectation.

That doesn't mean they're at the level of March, because they would tend to drop back down.

The first month in the quarter for us tends to look generally different from the third month in the quarter, but April is tracking to our expectations at this point.

For the second quarter with the breadth of market growth opportunities that we would see, including, we think the ability of our high-performance an log to resume growth in the second quarter, I would think that DSP and analog growth would each be comparable to the overall semiconductor revenue growth that we're expecting of about 4%.

Thank you, Eric.

Thank you.

Next caller.

Thank you, our next question is coming from Nimal Velipuran of BRKW.

A couple of questions if I may, Bill, first of all, given with some other semiconductor companies have given some sort of a negative, you know, preannouncement, you seemed to be doing exceptionally well in almost all of the markets and, all product categories except a few I have mentioned.

You know, I just want to understand where exactly is the market demand coming from, and how much is due to DLP for the first quarter, as well as the second quarter, as well as if you look at your OEM customers, their fourth quarter inventory has been one of the lowest in the last three years, is there any way it's possible that some of them might be building inventory?

You can give us color on this questions?

Yes, Nimal, in terms of where the demand is coming from, while as a component supplier, we're always more limited than that.

Clearly demand is coming from wireless in the switch to 2.5G and then beyond, demand is coming from various broadband markets that we have described where we were up sequentially in DSL and cable modem, for example, even fourth to first, and we think that seasonally demand will continue beyond that, pick up and wireless LAN as well, and in digital consumer markets.

Clearly demand is not coming from corporate customers, generally for equipment markets, although I think it's appropriate to point out that in the first quarter, as I think I had noted, our wireless infrastructure revenues grew sequentially, as we're seeing more demand there.

We're seeing 3G wireless infrastructure demand growing in the 3G in Europe and Japan, and UMTS in Europe, and in 2G, wireless infrastructure build out in China.

So those are all places where we think demand is growing, and will continue to grow.

There are potentially or are identifiable, you know, pockets of some inventories, for example, in the first quarter in wireless, we saw some we think -- we would characterize as isolated pockets of OEMs specific inventories that got really resolved during the quarter, and in wireless, we think there is some model specific low-end inventories in China that continue to be an issue in our printer customers, for example, still have some low-end inventories in excess that we think won't be resolved until during the second quarter here.

Outside of that, though, we think our customers' inventories are in pretty good shape, inparticular infrastructure inventories have been resolved.

We're also, it's pretty clear to us and our distributor's inventories are in good shape, so I cannot point to any OEMs that are -- our customers of ours building inventories beyond what they think they need.

Yeah, Nimal, I would add in terms of some of the pure analysis, especially when I look at the non DSP analog areas, you know, we saw the clients in standard logic, which is commodity, and there is no doubt that pricing environment is weak, the units even probably were down a little bit sequentially.

So, for other competitors that are playing as a higher percentage of their revenue with commodities, they're probably seeing a more disproportionate impact than what TI would with the proprietary line.

Just one more question.

I guess the question that everyone was trying to ask in this conference call is that, does this indicate that we're probably in the first inning of next upcycle or so?

Second upcycle.

Can you categorize what you have seen so far as part of that?

Oh, it's hard to describe, Nimal.

I would say what the overall cycle is.

We're in the point in semiconductor where in the first quarter we're 23% above the revenues a year ago.

We're well off the trough and so we're clearly in the next upcycle.

We frankly think we have been for a year or so.

I think the issue with a weak market economies, with weak markets and economies in various parts of the world with the issue of the war and so on, we have been in a pause of some end equipments and it's very hard to tell when some much that really resumes in a big way again, so, you know, 4% sequential growth for us in the second quarter for semiconductor pretty much hit on the average, I would think, what have we would normally expect, and 2% sequential growth in the first quarter was slightly better than average.

Thank you, Nimal.

Operator, next caller, please.

Our next question is from Cody Acrey of Legg Mason.

Thanks, and congratulations on a good quarter.

Maybe you can talk about your DMA 6 fab where you're sitting today in production and where you see that ramping to the end of the year.

Yes, Cody, we have maintained now our 130 nanometer production in DMA 6 and the 10000 wafers per month that we obtained at the end of last year on schedule, and we plan to keep that that way and that capacity is 100% utilized at this point as well.

In the rest of DMA 6, we're installing equipment for 90 nanometer production.

We've already sampled 90 nanometer digital based band.

It's 100% functional and one of our customers and in DMA 6, we think by early 2004, we'll have -- we'll be in volume production in 90 nanometer, will actually be in production in our KILBE 200 millimeter fab and 90 nanometer in the fourth quarter of this year, so we think we're at or ahead of anybody in the industry in 90 nanometer, as well as operating well in the 130 nanometer as well, the yields continue to increase and improve in 130 nanometer in the first quarter in our DMA 6 by the end of the first quarter, our 130 nanometer yields were equivalent the first time to the same yields of 130 nanometer and our 200 millimeter KILBE fab, which means there is no longer a disadvantage there for the 300 millimeter in terms of the larger wafers.

We think it will be about the middle of this year before our 130 nanometer process and our 300 millimeter wafers with cost savings, versus the [INAUDIBLE] in terms of yield, but we're exactly on track to get there, we believe.

Great, and one follow up.

Sure.

Your position with Nokia, specifically, as they're making a bigger push here in CDMA again.

How much of that did you see in the first quarter, how much of that would you expect to continue to help you in the second, and maybe you can talk about your CDMA strategy a bit.

Cody, I don't know that we have that much visibility on what our customers like Nokia do with our programmable DSPs.

They provide quite a bit of their intellectual property, regardless of the interface standard.

For CDMA, Nokia clearly uses TIs programmable digital base brand chip, and they provide that intellectual property, so, I don't have a lot of visibility beyond that except that they're using our base band chip in both markets.

We ourselves have not, at this time, introduced a chip set using TI intellectual property for CDMA, that is on our road map to do so, potentially at the 1X EVDV intersection point, which looks like -- which would look like it would be a possibility for us, but until we don't we don't have our own chip sets in that market, but with our programmable platform, any of our customers, in fact, can produce product to that area, interface standard if they desire to.

Okay, Cody, thank you.

Operator, next caller, please.

Our next question is from John Barton of Wachovia Securities.

Yes, good afternoon.

I wonder if you could comment on, you know, earlier this year, Intel came up aggressively with much fanfare into respect to an entry into one of your core markets of wireless handsets, I was curious what you see in the marketplace with respect to design wins, feedback from customers, et cetera?

John, in Manitoba, I can't say we have seen anything specifically in the marketplace or with customers.

We think that that Manitoba chip is the base band chip, the chip that we have been shipping to customers as much as two years already, and we have not seen any indications in the marketplace to date that the edition of imbedded flash provides any particular advantage to customers, or at least we haven't seen it in design wins that I can point to by them.

If anything, you know, the ability to provide other kinds of memory configuration, the commodity flash itself, stacking asynchronous DRam, for example, rather than embedding flash, would appear to be a more cost-effective solution.

So, cannot point to any direction competition to date from them.

I guess I would just add when Bill refers to the cost-efficiency of our solution, generally we have set our GPR at digital base solution roughly at $10, device, I believe Intel's positioning Manitoba as a $35 type of device where we frankly don't see much of the market opportunity for the digital base band at that price point with its performance level.

Do you have a follow-up question, John?

Yeah, I do.

Thank you.

Just go back to your previous answer, previous caller.

You talked about DMA 6 utilization, et cetera, just give an update as to how the relationship are going with the foundaries, the [INAUDIBLE] of the advanced technologies, et cetera

Yes, John, they're proceeding well, in the first quarter, I would senate probably about 30% of our total 130 nanometer capacity was coming from external foundaries that would include wafers that we purchased from TSMC and UMC and as well as some of the back-end of the line flow that we get from SMIC in China.

So, I think that says in addition to the, you know, 10,000 wafers a month in DMA 6 and additional 130 nanometer capacity, that we're getting from our KILBE fab, all of that is about 70% of our total, 30% coming from the foundaries.

As we have described in the past as we get to a, what is probably a peak over the next several quarters of 130 nanometer production before that starts for the client again, and many of the products transitioning to 90 nanometer, our use of external foundary in 130 nanometer at the peak could be 40 to 50% of our total sales to customers, so, we're part way up that slope, exactly as we had intended to be.

And that strategy is part of how we are structurally changing our levels of capitol -- capital expenditure and ability to drive deappreciation down to lower levels in the future years.

We're doing things differently on the operational front going ahead, and I would also add all of those foundary relationship are using, um, are using, are building parts designed using TIs process technology, meaning our engineers developed digital base bands or whatever, using TIs process, and then it's an operational decision whether that particular device is built in an TI fab or at one of the several external foundaries.

Thank you, John.

Next caller, please.

Our next question is coming from Mahneesh Goyal of Neuberger & Berman.

Hi, when you say 30% of your 130 nanometer capacity is coming from [INAUDIBLE] products carry similar type of gross margins that you generate internally?

Not necessarily, Mahneesh, because we know we're paying the foundaries a margin on the product.

We can never tell exactly what product costs are, you know, internally versus externally, but they're pretty comparable, we think so that the offset to what Ron described on the fact that we're avoiding capital expenditure is that we assume we're paying those suppliers in margin for their product.

If I can add that the way the math works, according to our financial analysis is at the peak, the foundaries, you will reduce our gross margins a little bit versus if we were strictly manufacturing internally; however, on average, our gross margin will lift because of the lowered depreciation cost.

Sorry, go ahead.

I mean I would just struggling with your guidance, the gross margin guidance of over 50%, because you approaching 75% utilization this quarter likely to go to 80% if you achieve 4% type of revenue growth and with $800 million of Cap Ex, I'm not sure how much more capacity you will have internally, and if external capacity don't make so much gross margins, how do you get to 50 plus percent gross average margins?

For one thing, the advanced logic areas, like 130 nanometer, we're still not at fully-learned yields, so the yields on each technology note will continue to increase over time.

Secondly, that, you're exactly correct on your average utilization; however, that is very different in the mix of our various products.

It is 100% utilized, 130-nanometer, it's quite a bit less out likewise than the average in our 200 millimeter analog fabs, for example, where we think we have a multiple quarters of filling up the fabs with new product, all which have will have a cost advantage over the previous product that was laid out at 150 millimeter wafers and, therefore, is a pretty long road map of margin increase as well as those fabs are more fully utilized.

If I may ask you very separate question, when you look at your business by device types, which businesses do you expect to grow faster than the average in the semiconductor business during second quarter?

Oh, I don't know that we is can have that granularity at this point.

What about by end-market types?

Well, by end-markets, you know, the markets that we think are going to grow for us are clearly wireless, which will be a combination of DSP and analog.

Some of the PC [INAUDIBLE] markets, primarily analog.

Digital consumer markets, a combination, and broadband, which is a combination of DSP and analog.

But, when we look at our forecast by-product line by end market, there really aren't major outwires, things generally will be running at about the level that we indicated for semiconductor overall with not any real significant outliners there.

Operator, next caller.

Thank you, Mahneesh.

Our next question is coming from Thomas Dornhill of UBS.

I'm not sure there's many questions left.

You've talked about capacity utilization, could you touch on capacity utilization in your analog sector and how you expect that to play out.

On the restructuring you're doing with regard to semiconductors, could you elaborate a little on what you're doing there?

Yes, Tom, on the capacity utilization in analog, it's, you know, significantly below the 75% average because of the fact that we built substantial capacity in 200 millimeter analog by bringing three fabs to that level a year and a half or so ago and it's only gradually that we're filling that capacity, so we have got a pretty long runway of multiple quarters of utilizing and continuing to fill out that 200 millimeter analog capacity.

And that's essentially where all of our analog capacity is now is at 200 millimeter.

On the restructuring, Tom, we don't have a lot more granularity on that right now just because we're rolling out, because the announcement in the release here was literally the first announcement, including to our employees, so there will be multiple sites involved, U.S. and international until those sites are notified and all, we won't be talking specifically about them, but really, but none of these actions is due to any weakness in demand.

It's all due to driving towards further efficiencies, toward more efficient staffing of the various fabs, and the fact that with greater productivity such as what we have been describing, we have to have a lot of discipline on the staffing of the fabs, so it's reduction in manufacturing, which will in total be about 450 jobs over multiple quarters.

Thank you, Tom.

Operator, next caller, please.

Our next question is from Clark Wispons of Smith Barney.

A lot of questions have been answered.

I'm wondering if you can give us -- is there a utilization hurdle that once you hit, your lead time starts trench stretching out.

If so, you can give us some type of time frame when you think that might happen.

Then I have a follow-up.

Clark, I guess in some, you know, theoretical way, you get 90-plus percent, even 95% and you want to make sure that they have other back up for making sure your customers are served, but in our case in the area where we have the highest utilization today in advanced logic, again, we have the benefit there of continually increasing yields, so by measuring utilization strictly on a wafer-start basis, it's not a complete issue of capacity because our capacity is continuing to increase effectively in 130 nanometer from continued yield improvements, which we think will continue for, you know, multiple quarters and as we have indicated, we have both the ability and the commitment to have additional foundary capacity there as well and in analog, we're quite a long ways from being at anything like those levels of where we would start to have any constraint there.

I guess anticipating, though, you know, maybe your follow-on or another question is that it's certainly not out of the question to us, and it's anticipated, though over a period of time we're going to continue to add, you know, advanced logic capacity by the fourth quarter of this year.

We will have production 200 millimeter, 90 nanometer capacity qualified and on line, so for example, that will continue to add to our overall advanced logic capacity of what we assemble the appropriate time.

Clark You have a follow-in on?

Yes, I did.

On a separate subject.

With your inventories rising and maybe up a little more or flat a little bit more in Q2, why not ask the distributors to carry some of that inventory load or is it direct customers that these are targeted for, just help me out on that.

Oh, it's a mix, Clark, and what distributors carry in inventory is up to them.

They buy the product, they pay for it, and they're not going to stock additional product that they think they might be at undue risk for.

In many cases, especially in our commodity logic, you know, where we're clearly gaining share in some of those markets, our distributors, you know, want to carry that additional inventory.

There is nothing we're going to force them to do there at all.

For example, in the first quarter it's pretty clear our revenues to distributors' decline slightly, and their resales increased slightly.

So, our inventory apt distributors, what distributors own of our products actually went down somewhat in the first quarter

Okay, thank you, Clark.

Although I realize there are more people with questions in order to prevent dueling conference calls and I realize some of you need to go to other ones, we're going to end it here.

So, before we end the call, let me remind you the replay is available on our website, and with that, thank you and good evening.

Thank you all for your participation.

That does conclude your teleconference.

You may disconnect your lines at this time.

Have a great evening.