微晶片科技 (MCHP) 2003 Q3 法說會逐字稿

Good day, everyone.

Welcome to Microchip Technology Q3 fiscal 2003 financial results conference call.

As a reminder, today's call is being recorded.

At this time I'd like to turn the call over to Microchip Chief Financial Officer, Gordon Parnell.

Thanks Eric and good morning, everyone.

During the course of this conference call, we will be making projections and other forward-looking statements regarding the future events or future financial performance of the company.

We wish you caution you, that actual events may differ materially.

We refer you to our press release of yesterday as well as our 10-K for the fiscal year ended March 31st, 2002.

Our 10Q for the quarter ended September 30th, 2002, and your 8-K reports.

That's all been filed with the SEC and identifies important risk factors.

In attendance with me today is Steve Sanghi, President and CEO.

I will start with a review of the December quarter results.

Steve will then comment on the quarter and outline our guidance for the March period.

We will then be available to respond to specific investor and analyst questions.

Net sales for quarter at $171 million were up approximately 1 percent from the September quarter and increased by almost 21 percent on a year-over-year basis.

Geographically, Asia grew by approximately 4 percent while both the Americas and Europe declined by 1-to-2 percent.

Total revenue breakdown for the quarter resulted in sales for Asia representing 41.8 percent.

Americas at 33.5 percent.

Europe at 24.7 percent, reflecting the sequential changes in the quarter as I've just out lined.

Net sales for micro-controller products represented approximately 80 percent of revenue in the quarter.

Analog products represent about 8 percent of revenue and memory products 12 percent of revenue.

Steve will address the performance of each of these product areas during his comments in a few moments.

Diluted earnings per share for the December quarter were $37 million or 18 cents per diluted share.

An increase of approximately 5 percent from $35.2 million or 17 cents per diluted share in the immediately preceding quarter.

Year-over-year earnings per share in the December quarter increased by approximately 56 percent from a net income of 23.6 million or 11 cents per diluted share.

Gross margins in the December quarter was 54.9 percent an increase of approximately 100 basis points from the September quarter.

This margin established a new record level surpassing the gross margin in the second quarter of fiscal 2001.

Gross margins in December quarter were primarily impacted by the product mix of microcontroller's and analog product, manufacturing yields and fixed cost absorptions.

Factory utilization levels were 86 percent in the December quarter.

Start-up costs for our Gresham, Oregon facility were about $3 million in the December quarter.

Excluding these costs, our product gross par begins for the quarter were 56.7 percent, evidenced in the underlying strength of our business model.

Operating expenses were 26.2 percent of sales in the December quarter, compared to 26.8 percent in the previous quarter.

Research and development costs were $22.3 million, representing 13.1 percent of sales.

Sales and general administrative expenses were $22.4 million, representing 13.1 percent of sales.

Operating expenses overall showed a decrease of .7 million from the September quarter.

The leverage improvement from the growth in revenue and lower expenses resulted in the .6 percent change that I outlined.

Microchip's total inventory position at December 31st was $94 million, an increase of approximately 1.8 million from the prior quarter level.

Inventory turns are 3.3 with days of inventory representing 111 days.

At December 31st, our receivables were 46 days compared to 48 days a year ago and also 46 days at the end of the September quarter.

Our total receivables were $93.4 million, showing an increase of $3.5 million from the prior quarter or approximately 3.8 percent.

The increase in receivables reflect a non-linearity of shipments in the December quarter.

Overall receivables beyond 90 days continue to be at extremely low levels with only 1.2 percent in this particular category.

Distribution inventories were at very low levels in all geographies.

We expect that to be the case for the foreseeable future.

With this being true for Asia also.

That in line with all the other geographies.

As of December 31st, Microchip's cash position was $198.2 million and there were no borrowings against our lines of credit.

During the quarter, Microchip generated a approximately $75 million in cash from operations offset by capital expenditures of $21 million.

Free cashflow from the business after capital costs working capital costs and treasury activities resulted in adding $57 million to treasury.

Capital spending was approximately 21 million for the quarter.

Depreciation expense for the December quarter was 28 million versus 28 million and in the same quarter of last year and also the same level in the September quarter.

Microchip shipped approximately 6,500 new application development systems to customers during the quarter and we have shipped over 251,000 systems overall.

Excuse me.

With that, I'd ask Steve to give some additional color ton December quarter and then discuss our guidance for March.

Thank you, Gordon.

And good morning, everyone.

I would first like to reflect on the financial results that we have announced yesterday.

And then I shall give you update on sales force start-up and then give guidance for the March quarter.

First, I'm proud of our solid execution during this most difficult business environment.

While a majority of the industry and our competitors are still posting loses, Microchip announced operating profits of 28.75 percent.

Earnings per share were up 5 percent sequentially and up 56 percent over a year ago.

Our gross margins improved further to a record 54.9 percent an improvement of 100 basis points sequentially.

This is the same record gross margin that we achieved back in September of 2002 at the peak of the last business cycle.

Achieving record gross margins, while the industry still is in the thrust of the business cycle demonstrates the strength of our business model.

We had previously told investors that we will break out the Fab 4 expenses for startup.

Those expenses last quarter were $3 million, excluding the FAS 4 startup expenses, the product gross margin was even higher at 56.7 percent, indicating what can be achieved in a few short quarters as Fab 4 goes to production.

Next -- next, I would like to briefly comment on the sales from each product line.

Our flagship microcontroller product line was up 3 percent sequentially and up 24 percent year-over-year.

While industry results for calendar year 2002 are not known yet, our best guess is that the industry was flat plus or minus a few points.

Therefore, with the 24 percent year-over-year growth, we are confident that we gained a substantially additional market share.

Our new product announcement in the areas of mixed signals flash microcontrollers, new 8 and 14 pin flash microcontrollers and our low power NanoWatt microcontrollers have seen excellent reception in the marketplace.

Last quarter our Flash microcontroller product sales were up 24 percent sequentially and up 52 percent over a year ago quarter.

So Flash microcontrollers are driving most of our growth.

We believe that we continue to be positioned very well to accelerate the growth rate further when the market sentiment improves.

In the analog segment of our business, we were up 3 percent sequentially and up 34 percent over a year ago quarter.

There continues to be strong acceptance of our analog product portfolio and there are now over $700 million in our opportunities management (inaudible).

We believe that a growth of 34 percent over a year ago quarter is the best performance in the analog industry.

Our strategy of attaching analog products to microcontroller products is working.

Excluding the cellular phone part of the business, which is 27 percent of the analog business, the rest of our analog business has 73 percent attached rate to our microcontrollers.

Let me repeat for clarity, excluding the cellular phone portion of the analog business, which is 27 percent of the analog business, the rest of the analog business now has 73 percent attach rate to our microcontrollers.

We are continuing to gain market share in the analog products.

Analog products also achieved the highest gross margin of any product line in Microchip last quarter and contributed to Microchip's achieving the record gross margin.

The memory portion of our business was down sequentially by 15 percent after Intel's announcement of flash memory price increases, there were indications in the market about price increases on serial EEPROMs.

We did not see any evidence of such price increases.

There is too much industry capacity for price increases to stick.

Our strategy on serial EEPROMs is number one to focus on the proprietary segment of the serial EEPROMs business that is attached to our microcontrollers.

And, number three, to cherry-pick the rest of the business for higher gross margin.

Following this strategy, our serial EEPROMs business continues to be very profitable.

Fab 4 update.

A startup team has been put together with all critical positions filled from Microchip-trained executives and engineers from our Arizona fab.

These personnel are familiar with our processes and business methodology, thus, ensuring the timely success of Fab 4.

With five months passed since the acquisition of Fab 4, no negative surprises have been found.

January 17, just a few days ago, was a very critical milestone.

On that day, we started running silicon in Fab 4.

Now, it will take about six months to complete process installation and qualifies before Fab 4 can start production.

There are currently no show stoppers in Fab 4.

We made a press release on January -- 2003, we have retained Ben Schinger (ph) as the exclusive listing agent for selling our Fab 3 facility in Washington.

Ben Schinger (ph) is a highly experienced agent in selling special class facilities, such as Fab 3.

There are approximately $1.1 million first quarter expenses in Fab 3, which are included in our costs of good sold.

Once Fab 3 is sold, these expenses will go away, hence, improving gross margins by additional 60 basis points.

Now, the guidance.

March is seasonally our weakest quarter of the year.

Asia is our largest sales geography and much of Asia has holidays related to Chinese new year.

Therefore, our guidance at this point in time is, for our business to be sequentially flat, to up single digits, I will caution investors, do not analyze this number off as seasonably the weakest quarter of the year.

The history in the past years proves it out.

We expect earnings per share to be approximately 18 cents, which would equate to a year-over-year earnings per share growth of approximately 40 percent.

With that, Eric, would you please call for questions?

Thank you, sir.

The question and answer session will be conducted electronically.

To be placed in the queue for a question, please suppress star-1 on your touch-tone telephone.

We'll take as many questions as time permits and in the order that you signal.

Again, that is star-1.

To signal for a question.

And we'll first hear from Chris Donnelly of J. P. Morgan.

Hey, guys, thanks.

You know, Steve, back in October, you -- guidance based on what your customers were telling you.

And it sounds like December kind of softened up.

Can you just take us through, you know, what happened in the quarter and why you guys came in at the lower end of the range and what your customers are telling you right now?

Well, you know, December quarters started with a dock lockout.

Then the Thanksgiving was about six days late, you know, pushing our visibility into the later part of the quarter.

A lot of the holiday related sales numbers that you have heard from the industry were extremely weak.

And at Microchip, a lot of times, an indication of the product s that are selling, you know, right after the Thanksgiving and those products still need to be built and brought back on the shelves for additional sales towards the holiday quarter.

So the entire activity got pushed out later in the quarter.

And, you know, we basically made the numbers on micro and analog products.

The real mess largely was in the serial EEPROMs, where there was some pricing-related indication in the market.

What my opinion, everybody thought the serial EEPROMs market would be a lot better than it turned out.

And now, we, essentially, focus on bundling the business with microcontrollers.

We focus on the proprietary products.

And we cherry-pick the rest of the business.

And we kept the e-square business quite profitable.

Now, we're totally focused on the top line of the serial EEPROMs business.

It is 12 percent of the business today.

So it doesn't hurt us much year-over-year.

It can really do anything to a quarter, but really, over a year, it's a pretty small portion of the business.

And we expect it to go even lower as the microcontrollers and analog products continue to grow.

And that's really where we like to be.

Then what's the turns percentage needed for guidance in the March quarter?

The March quarter is in the same range we had last quarter, about 61 percent.

And how much of that is done so far in the quarter?

Can you tell us how the quarter started off?

The bookings have been fine so far and then with 23 days to go, I don't think we have a lot of color on it, but the bookings are fine.

So far it is fine.

I think you really can't tell much more meaningful information until we get much later in the quarter.

It just sounds like December really fell off the cliff.

I'm just wondering if things have picked up since then?

December fell off the cliff really going into the holidays.

January shows not much this time.

Thanks, guys.

next, we'll hear from Mark Edlestone of Morgan Stanley.

Good morning, guys.

Two questions, if I could.

One just relates to inventories and the distribution channel.

Can you give us kinds of an overview of how that looks on a regional basis and then I had a follow-up question on the analog?

Mark, overall inventories are like 2.71 switches, which is flat with prior quarters.

Regionally, they're all relatively close to the same levels.

We don't see tremendous changes.

And just as a reminder for investors, the range of inventories that we've seen in the distribution channel is from 2.4 months to 3.5 months.

So we are very much in the low ends of the range that distribution has seen on their shelves.

We don't expect to see much change in that in the near future.

Thanks, Gordon.

Then a question for you, Steven, on the analog business.

The data you gave there on the attach rates is pretty impressive.

Could you just walk us through how you think the business unfolds over the next couple of years?

Clearly, you have been able to get good penetration of your microcontrollers as kind of a lead horse there.

How do you capitalize on that strong attach rate going forward?

Is that only going to be trying to get incremental parts to attach or going out and just building the product portfolio to take advantage of that as well?

Well, really, you know, all those dimensions, we have a fairly large analog team now.

The one we had ourselves and the one we acquired from TeleCom (ph) two years ago prior to us acquiring TeleCom, they had acquired a design company in which is now a part of -- in Switzerland, which is now a part of Microchip.

So we had essentially three major design teams on an a lock.

One in Mountainview, California, one in Lausanne, Switzerland, one in Phoenix, Arizona.

And they are producing a large number of new products and a lot of those product s that we are producing, you know, are key criteria is that these products have to be in the signal chain of an embedded control design and they will attach to a microcontroller.

And we continue to do a very good job in that.

We have several hundred million dollars in the opportunities management funnel right now that we are trying to harvest to really bring them down the funnel and fall down to the orders.

The, you know, the only negative is really the marketplace.

You know, you look at the growth of the other analog companies and some have announced numbers and some have not.

And we see our numbers are really, you know, in the range.

Year-over-year growth has been better than any other analog company, but in a very, very tough to market environment, everything gets difficult.

So as this market environment improves, we believe we are extremely we go positioned on analog and are continuing to really grow that portfolio further to attach even further.

You know, finally, not all channels, all geographies, all regions, are perfectly working on analog yet.

And that's just introducing any new product line and the places where we have analog specialists, the places where we don't, and the places that need help.

So there are, you know, there are still a lot of different channels.

When I say channels, you know, the word for us is divided in about 28, 29 sales offices around the world.

And only about 70-75 percent are really engaged, producing these analog results.

The rest are still being worked or the analog specialists are being hired.

So there is still that multiplier left to go.

Great.

Thanks a lot, Steve.

Congratulation on the great margins.

Thank you.

next we'll hear from Jeff Rosenberg of William Blair.

Good morning.

Steve, can you give us an update on the digital signal controller, the timing of the introduction and when revenue contribution there might begin?

The update on digital signal controller is that the parties are out.

It's functional and we are Beta assembling right now.

We started doing that, I think, in December time frame, late-December.

It was sort of small news.

It's now going to the press release, but this was, you know, this is an update, so we should start very broad sampling of the DS-pick (ph) product in the next three or four months.

As a couple of bugs are being fixed, but the part is functional enough that it is sampling to our customers and we'll start shipping product for revenue the later part of this calendar year.

Great.

On Gresham, the question I had was, is the timing of the ramp to production there a function of where you're revenue gets do utilization elsewhere, that sort of things, or are there process technologies that you will be bringing up there that are necessary so that it's kinds of independent of how revenue progresses over the next several quarters?

There are two factors determining the schedule for Fab 4 production startup.

One is the overall capacity, which is what the revenue is and what the utilization elsewhere is, like you mentioned; with you the second one, equally important or more important, is the capacity need for the advanced technology.

And it seems to me that the technology need for advanced capacity will come first before the rest of it.

You know, but they're really both neck-and-neck.

You know, basically, you know, as we have done this quarter -- last quarter was 1 percent growth and this quarter was flattish to low single digits.

At that kind of growth, the growth will pick up in the June/September quarter because tows are the stronger quarters him we'll get to the points where the utilization of existing factories will, you know, get into over 90 percent.

At that point in time I think it gets too risky to not have the new fab startup.

If you look at the advanced capacity, which is building our advanced flash microcontrollers, that will even get tighter.

And that looks tighter now.

We will need the fab in the next several months.

Okay.

Then my last question is, over the course of, I don't know, going back several quarters at least, you've accumulated lot of design wins and a good momentum there.

It's just as you look at the progression of those market wins coming to market, have you seen when you look into your pipeline there a lot of designs that didn't ever make it to market?

What's the outlook for people bringing new design wins to market over the beginning of this calendar year?

Well, you know, we have spent a lot of time looking at the design funnel and going back a year ago what we added in the design funnel a year ago.

Why has it not going gone to production and why is it not driving the growth in you know, any kind of math you apply on the overall funnel would really say that the growth rate should be significantly higher.

When we do that analysis, the biggest thing that comes out of that, really is a large number of designs that really have gone nowhere.

They haven't been lost to a competitor.

They haven't gone to production.

They are still in the funnel and they are still on hold, basically, at the customer, waiting for a better launch environment for the customer to launch those new products.

And in some cases, the design either has been canceled or the engineer has been laid off or the division has been sold.

You know, all those economic-kind of issues in the global economy.

So what really that has done is, it has changed the dynamics of the funnel and has changed the math, you know, where it has a little bit less predictable than the once we had really got in use to -- gotten used to over the last several years in really working with the funnel.

So you know, such the metrics variables, they are sort of being reworked.

I think we still maybe 24 percent year-over-year growth, which is incredible in this marketplace, but the quarter growth, sequential growth, has to come little predictable.

Do you see that as a current -- currently that there is no change there or what is your feel for whether or not that situation, you know, turns in your favor as you move forward from, a you know, the seasonably weak period?

I think as you move away from the seasonably weak period, you know, the thing that comes favorable.

If you look at last year in a fairly miserable economy, miserable environment, we had a 17.3 percent growth in the June quarter last year and other similar in the December quarter.

These are not exact numbers.

So those are much stronger quarters.

We are confident there will be significant growth ahead.

And a lot of these designs are going to production.

Every quarter, we are taking a lot of additional designs to production.

They are just taking longer than the original map was and a less portion of them going to production than the original math was.

But if you apply the original math, the growth explodes.

We are certainly not expecting that, but we are expecting pretty significant healthy growth next year.

Okay.

Thanks.

Next, we'll hear from Douglas Lee, Bank of America Securities.

Hi, guys.

Great job over the last 12 months.

The first question is, with respect to the gross margin, I understand it's impacted by the utilization.

You are clearly guiding sort of flat in the March quarter.

Previously, the way we have been thinking at object the model is to sort of hold margins flat up until December when Fab 4 kicks in.

Is it safe to assume the new 55 percent range is sort of a new floor, all else being equal?

The answer to that is, yes.

You should expect the margins to be in the 55 percent range of Fab 4 production, which should happen late in the -- this should happen in the September quarter and then the margins should start rising.

when the margins start to rise, say, in the December quarter, all else being equal, mix and what not, does the gross margin step up to the gross product margin level or is there some depreciation costs that should offset that?

It doesn't -- go ahead, Gordon.

I mean, it wouldn't usually step up.

Obviously, we don't romp with a facility to the point where it absorbs all of the costs, but, you are right in the sense that there are other product-related costs and depreciation costs which also come on board.

So, again, relate you back to the modeling that Steve and I have showed the street, getting towards 58 percent margin range by the end of '05 and essentially tracking towards that from the start of -- through that period.

Great, that's helpful, Gordon, thank you.

Just lastly, so a lot of companies sort of have some weird seasonality with the Chinese New Year kind of coming late in January this year, how does that affect your visibility or is that not a factor?

Thank you.

That we have no cycles of learning on what happens with Chinese New Year when it comes early or late?

As a result, I don't think we know, which way, to apply it.

fair enough.

Thanks.

Next we'll hear from Chris Caso of SoundView Technology.

Hi, good morning.

I just wonder if you could talk a little about what your production plans are going to be going forward in the March quarter and maybe into June as well in terms of wafer starts and what you'd see in terms of the trend being in your own internal inventory?

The, you know, the fab utilization in that existing two Arizona fabs is basically inching up slightly.

You know, as the revenue grows a little bit, we take it up a little bit.

The inventory thing is short of flat shipment.

The next step is when Fab 4 comes into production in the September quarter.

So do you expect that inventory levels are going to remain flattish as you ramp Fab 4?

Yes, because that comes up in the fairly high growth seasonality -- seasonally high growth quarters up for the company, yes.

Okay.

I was wondering if you could break out a little bit, talking about the rising gross margins for this quarter and going forward, what are the components of that?

You know, this quarter, product mix, obviously, the proprietary product have done extremely well in terms of the component of the business.

We've talked about the additional absorption from utilization factors that we have in the business.

So, you know, as we go forward, we expect continue to see the proprietary products do well and continue to be a driving element of that and, as Steve said, we expect to see modest changes in the absorption rates, particularly as we get into the June and September time frame.

One last question I guess on the flash based micros.

As those are growing I guess faster than the overall market.

Are the gross margins on your flash-base products equal or higher than company average?

How does that change the effects of the gross margin?

Well, that has been a very positive influence over the last several quarters, because, you know, flash gross margins are quite good.

They're the best among the microcontrollers.

So the mix within microcontrollers, richening -- the mix in the overall company is richening with the more analog, more flash microcontrollers, more overall proprietary products as the portion of the company's business; and when the memory is shrinking, the portion of the memory that's shrinking is a pure commodity piece.

The proprietary piece is doing well, the vendor piece is doing well.

So, you know, the product mix is a major element of the substantial gross margin improvement.

A very small piece is coming out of the improvement in utilizations and also when you are running the factories flattish as we are running today, you have very high efficiency, you have a very well balance -- you have outstanding record years, you because you are not spending the energy into a lot of other trash, taking it up or taking it down or hiring or firing people an all those changes that you go through both in a good and a bad environment.

With the environment, -- we have in two areas on the fab is just, you know, very flattish, slightly up, 1 or 2 percent per quarter.

In that environment, the efficiency and yields are absolutely outstanding.

They are the record.

Okay.

Thank you.

Next, we'll move to Ray Ron (ph) of Schafer (ph) Investments.

Yes.

I was wondering if you could comment on the year-over-year growth rate in the analog that is not -- being used in cellular phones.

You made a point of talking about the attach rates for non-cellular analog.

If I understand the question, it is, if you take the cellular portion away, what would the growth rate be for the rest of the products?

I think that's your question?

Yes.

We haven't calculated it that way, but I can give you some data so you the triangulate it to that.

Two years ago in March quarter of 2001, two years ago, the cellular phones were 57 percent of the business in analog.

And now they are 27 percent of the business.

So they really have gone to half.

You know.

While the business has grown substantially.

It was up 34 percent last year.

I don't know what it was up a year ago.

So I think you can get some parameters from there.

But we haven't really calculated it that way and I can't do it online here.

Okay.

Thank you.

Next we'll move to Ambrush Razhaghava (ph) of GKM (ph)

Hi, thanks.

Most of my questions have been answered, but, Steve, I was wondering if you could just educate us a little bit on the SPIC offering that you guys have and the one that your largest competitor has also introduced?

And I'm thinking more in terms of what end markets and applications that are you targeting versus after what they are going after, realizing the mark is very large and the opportunities that offers.

Who is they?

Motorola.

I don't know what products of Motorola are you talking about?

You know, they're a large company.

They got lots of product.

yes, this is, I don't remember the exact product name off the top of my head.

When I look at the product release which came out a couple months ago, it sounds very similar to what you guys are talking about.

They, essentially, again, Motorola brings out their leadership in the DSP, leveraging that to go after the applications which would sort of straddle between high end 8-bit and low end 16.

The -- I can't talk about Motorola.

But I can talk about ourselves.

The product line we have is a product line which is, we believe, the most unique in the industry in which we have combined our registration set of a digital signal processor.

The registration instruction set of a high end microcontroller into a single integrated architecture, which is the highest performance, 16-bit microcontroller.

And gives you DSP for free in the same chip.

We do not know there is any other product ton market.

And I haven't seen the product you are talking about.

We are getting very, very good acceptance of that product in the marketplace.

We have well over $100 million in the opportunity management funnel already.

And we shipped our silicon samples just a few months ago.

We have a number of customers already committed to go to production in the later part of this year.

There is a very, very large opportunity emerging. 90 percent of the customers that we are seeing on those products are Microchip's existing 8-bit microcontroller customers who have been struggling whether we should use a microcontroller or a DSP and as the applications are migrating, they have to either use one of each.

One micro, one DSP, or use a DSP and lose the advantages of micro, or use the micro and lose the advantages of DSP or use this chip and get the best in the world.

And they are choosing to use this chip.

Okay.

That's goods.

And just based on your past experience with new product introductions and the stage where you are at right now, when do you expect to begin to see traction in terms of meaningful revenues?

Are we looking at the second half of CY 04?

Traction means different things to investors.

What does traction mean to you?

To me, it means north of $10 million run rate per quarter.

North of 10 million run rate is a couple years away.

Thanks.

Next we'll hear from Tim Mahon of Credit Suisse First Boston.

Thank you.

Gordon, can you comment on microcontroller and analog ASPs sequentially and year year-over-year?

No significant changes there in terms of year-over-year or dynamics in the current quarter.

The prices -- environment for microcontrollers and the proprietary analog is still relatively benign.

You know, once you're in the circuit, you basically are there for the life of the application.

There are obviously anticipated with our major customers, price moves year over year, they're still in the price range we talked about historically, five to 7 percent for a mature product.

When you look at the overall product, Tim, and you add in the emergence of flash products and the additional analog integration we have on our products the overall product line ASP, ASP is falling at a lower rate than that.

So those dynamics haven't changed.

Great.

Steve, just, you know, looking at the automotive market as we move into the summer time frame and we start hearing about '04 models, can you give us some kind of color, what's your position in relative to '03 models may be?

My position in 04 relative to 03 models is quite good.

I really can't quantify for you today, but in the next conference will be showing some slides which will show the proprietary Microchip products in the automotive and around the world.

We have shown those lines for model year '03 before.

Model year '04 goes to production in the July/August time frame.

People start buying those products in the April time frame.

The subcontractors of the automotive company start buying those products the next quarter.

And there is a significant increase in penetration in that model year, which will be showing at the next con -- conference we go to in February.

Increase in penetration for Microchip?

Yes.

Just finally, ail throw this at I because you talked about it at our December, you mentioned that you believed that street estimates for fiscal year '04, you believed were significantly light and I'm wondering, given what you are seeing today, if you still believe that to be the case?

Yes, I still believe that to be the case.

I think, you know, the model Has showed, I think we are likely to achieve that model.

Maybe with a slightly lower revenue that I showed.

But similar kind of earnings per share because of the higher gross margin.

What I would take out is probably some serial EEPROMs revenue out of that mix, which is not going to be very large.

Again, I really caution, you know, people really getting into the habit of annualizing a single weak or quarter.

Investors have done it, just way too many teams for Microchip.

All these numbers change again in June and September when the quarters are stronger and we come down in the December and March quarters, this happens every year.

Look at the year-year -- over-year growth in a terrible economy with 24 percent in micro, 34 percent on analog, the two major segments of the business.

So we choose to be very confident about producing pretty significant growth next year.

Great, thank you very much, Steve.

Our next question will come from Gail Alexander (ph) of Oxford (ph) Associates.

Congratulations.

When you look at '04, don't we get a situation where you can probably get to 31, 32 percent operating profit margins?

Which year, sir?

'04.

That's the model we have shown at the various conferences.

oh, I hasn't seen that model.

we have shown a 58 percent gross margin and getting to approximately a 25 percent type of operating expenses, which gets to about 33 percent operating costs, yes, slightly higher than you think.

Oh, I thank you.

And last question, could you give us a an idea of startup costs on Fab 4 for the September and October -- and December quarter of this year?

I'm sorry, say that again.

Fab 4 startup costs for the September quarter and December quarter this year?

We do not have a prospective cost and they're being analyzed every day based on where we need to run and how hard we need to ramp, so we don't have that model for you.

We have committed in the past that we'll give it to you going backward, so the last quarter was about $3 million.

I thank you very and congratulations.

Next we'll hear from John (ph) Lopez (ph) of Pilgram (ph) Baxter (ph).

hi, guys.

Who questions, if I could.

The first one is, just in terms of Q4, can you walk through again December, the holidays, obviously, we knew were coming, but, yet, you sort of -- it seems they surprised you.

So what was surprising about the holidays as sort of an adjunct to that, why won't you see something similar when the Chinese holiday rolls off at the end of January?

Well, the difference of the holidays is the Christmas holidays come at the end of the quarter.

The Chinese New Year is in the middle of the quarter so usually, every year, we see a tremendous strength when the people come back from the Chinese New Year, all the factories that need to buy products.

It's not the same.

If the holiday are in the middle of the quarter, people buy strong before the holiday, then they shut down and buy very strong after that.

You get my point there?

I do.

So according to that you should actually see some strength in the February period in Asia?

Well, you should see some strength, but you lose the holiday period in which there is no purchases in Asia.

Ahead of the holiday?

Or during the holiday.

During the holiday.

Okay.

During the holiday when the factories are closed, they're not using product.

Okay.

So weakness in the sort of latter parts of January that should be followed by a resumption of buying in the early-to-mid-part of February?

I think my belief in Chinese New Year starts on January 31st, so it's January 31st to a couple of weeks from then to the middle of February.

So that's the time in every company's differ, some are off a week.

Some are off two weeks.

Some have a hangover, some don't.

There is a lot of moving parts there..

But that time frame, you know, late part of January to middle of February by two, two and a half (inaudible) you really see a fairly skewed results out of the Asian part of the business.

Okay.

And so your visibility into -- would you say your visibility is roughly the same going into Q1 as it was in Q4 given the turns percentages, would it be the same?

Yes, that's what we said, our turns are about the same this quarter as they were last quarter.

But your visibility, everyone though the turn numbers are the same, the quality of the turns would be the same, is that accurate?

Yes

The second question, I'm sorry, do you have estimates internally for what the 8-bit microcontroller market is supposed to do growth-wise in '03 and '04?

No, we don't.

We have no way to get at those numbers and even if we have them, it doesn't make any difference our strategy in microcontrollers is to continue to produce hard-hitting new products, you know, like these flash microcontrollers, analog technology, the new 8-bit and 16-bit microcontrollers and go position them around the world and win as many designs as we can.

When the market is zero, we grew 24 percent.

When the market was down 33 percent in 2001, you know, we were down only 5 percent.

So we don't really care about the market.

Our goal is to continued to grow every year.

And we have done that in nine of the last -- 11 out of the last 12 years.

Okay.

Great, thanks, guys.

Great quarter.

Just a reminder, it is star-1 to be placed in the queue for a question.

We'll next hear from Ben Lynch of Deutsche Bank.

My first two questions were resolved.

Even if it's just (inaudible), could you maybe characterize are there any changes in the purchasing environment currently in the automotive and consumer segment ?

What was the question?

Are there any changes, even though it's subtle.

In the automotive and consumer markets?

You know, I think none of these markets, you know, one market is homogenous, you heard the housing starts which are, you know, highest in 16 years or something like that.

Around you know, house is loaded with Microchips and microcontrollers and every new house, you know, drives as lot of sales and the automotive sales are strong.

You probably have heard recent reports that some of those car bits will be going up.

You know, the PC market continues so-so.

So every market is different and Microchip got a lot of moving parts.

And in that, the geographic split, put the Chinese New Year in there, you know, you get too many different parts of the matrix.

I can't really give you when we return answer on it.

So there are segments of the business which are extremely strong.

There are other segments which are not.

Okay.

I thought I would try a difficult question given the diversity of your end-market mix.

A quick question.

How much is the flash microcontroller of total microcontroller economy?

That, we do not disclose.

Okay.

I mean, you were talking about strong growth, do you have a low base or something meaningful?

The base is actually quite large.

The base is quite large.

It's not a small growth.

It's one of the largest product lines at Microchip.

The base is very large.

But we are also converting a lot of other non-flash, OTP microcontroller business, very aggressively to flash, because they offer better value.

They are lower prices.

They don't have advanced technology.

Lower costs, I'm association not lower prices.

Most get a higher price at a lower cost.

Substantially high margins.

We have been driving that business very, very strong. they offer very large base.

But believe me, our competitors would love to know.

And part of that is cannibalization?

Yeah, we are doing self imposed cannibalization, which is really -- and in the process, we are driving our gross margins, we are driving our capacity.

We get more value per wafer out of those versus the ones we are converging.

We get higher margins, in some cases we get higher ASP and at lower cost.

You see that in the results with a relatively flattish 3 percent growth in microcontrollers lasted quarter we drove 100 basis points improvement in gross margins.

Some came out of analog, too.

But analog was only 8 percent of the business.

Okay.

A couple of questions then before you Gordon.

Would you update us we should be working with as a tax rate assumption for '04, give us a sort of maybe a qualitative feel for the serial EEPROMs gross margin relative to the corporate average?

The tax rate for this quarter and for '04 will still be in the 25.5 percent rage.

That would stay pretty much in place.

The second question on serial EEPROMs?

Yes.

Where gross margins are there, you know, I guess you might want to be (inaudible), but just broadly relative to the corporate average.

They will be south of the overall corporate average, but I think what is important is really the operating profit on that, on that line.

As Steve said, we do tend to cherry-pick where we want to play, so we are much more thinking of the overall requirement in terms of making that a profitable line versus top line growth and driving it from that perspective.

We never lost money on serial EEPROMs.

And even in an environment in December where the revenue was obviously -- didn't meet our expectation, was down.

It's still handsomely profitable at an operating profit basis.

Great.

Thank you very much

Next, we will hear from Joe Osha (ph) of Merrill Lynch.

A quick question for a clarification on gross margins.

The outlook for if March quarter of 55 percent gross margin, are we assuming a similar level of startup cost for Fab 4?

The startup costs for Fab 4 are higher in the March quarter than they were in the December quarter.

But those are being balanced by continuing richening of the product mix on the dynamics we talked about within the microcontroller market, you know, moving to a shrink technology, more flash, all that kind of stuff, higher mix of analog products, improving gross margins, also, so all those elements we have talked about, slightly higher utilization.

Should be able to cover for the additional startup expenses of Fab 4.

Great.

Then a housecleaning item.

I may have missed it, what was the book-to-bill for the December quarter please?

For December quarter it was approximately parity.

Great.

My last question is, can you give us an update on rfPIC both on a product and revenue level?

Thank you.

What was the last question?

rfPIC.

rfPIC.

an update.

From a product and revenue basis?

The revenue on rfPIC is still relatively small.

We have a handful of products, I believe we have five or six products.

Most of those products really have been introduced in the last year.

So the rfPIC is really not out of the funnel yet, in a way.

Because most of the products have been in the funnel for less than a year.

There is some small revenue in rfPIC, but it does not neat test of being very meaningful yet.

do you expect it to reach material revenues by December '03?

December '03, probably, yes.

Maybe the following year.

Thanks very much.

Next we'll hear from Pete Steelman (ph) of Davis Hamilton.

Hi.

For those of you who haven't seen your formal presentation, can you just real quickly run through those factors that you talked about kind of drive your profitability to the levels you are thinking about for fiscal '04?

Certainly.

If you look at the gross margin last quarter, the product growth margin was 56.7 percent.

The company GAAP gross margin was 54.9 and that difference was a $3 million expenses in fab startup.

So as fab 4, you know, goes to production and then ramps and becomes accretive you know which is really a couple of quarters after fab 4 starts up, you will have the gross margins that will approach the product gross margins that we are achieving today.

Then you add to that, the continued richening of product mix, the same phenomena which has driven gross margin from 49.5 percent just about a year ago to 55 percent last quarter can take us the rest of the way from 56.7-to-58.

Which is only another 100 basis points.

We have been improving that every quarter almost.

So you could really see a path to 58 percent gross margin easily.

I can see it higher than that.

Let's leave it at 58 right now.

The second element is the operating expenses, last quarter there were 26.2, the quarter before they were 26.8.

And our model is about 25.

So we have to make 120 basis points operating expense leverage, which is not very difficult at all, you know, as you are having some growth quarters, you can't really add expenses fast enough at the bottom.

So 58 percent growth margin minus the operates expenses is 33 percent operating profit.

Thanks.

Next, we'll hear a follow up from Chris Donnelly of J. P. Morgan.

Thanks, guys.

A couple house keeping questions.

Steve, I'm trying to understand the expected linearity this quarter.

Is it -- do you think it will be sort of front-end loaded or back-ends loaded with the Chinese new year?

Back-end loaded.

Okay.

And then, you talked a little bit about the start-up costs increasing.

Can we expect those to approximately double?

Can you just give us the sense of the range there?

I can't.

Another question was asked earlier.

I don't have that right now, but I think what I did say is that the startup -- increase in startup expenses in fab 4 should balance with the improving gross margin in the mix change, the improving utilization and all those factors to be able to drive about a flattish 55 percent gross margin.

Gordon, I guess depreciation is about 28 million a quarter.

Can you give us a sense of, is it going to be 28 million again this quarter and then also what you think it's going to be for '04?

Is it going to go up or down?

It will be pretty much in the 28 million ranges for the current quarter, Chris.

You know, when we look into the next fiscal year, we obviously have some level of depreciation coming on into play for Gresham.

But given the relatively low cost of acquisition, the split there into buildings, leasehold improvement and equipment, we still don't have that valuation completed, but we know that that's going to add modestly to our depreciation when Gresham starts to come to fruition.

Balancing that off, there is obviously some equipment in our older fabs that has reached the ends of the depreciation life.

So I'd expect the depreciation to increase but quite modestly in '204.

Great.

Just to -- in on the serial EEPROM.

Steve, what the decrease mainly due to pricing or did you guys get some push-outs, cancellations?

Did you not want to participate in lower pricing?

And if you could give us an expectation of your pricing going forward?

Well, the decrease in serial EEPROM was about half-half because of pricing and unit.

Units were chosen by Microchip and the pricing was the marketplace.

Then your expectations going forward for pricing and when do you think the serial EEPROM market bottoms.

I would say flattish.

I'm going to set the place where it's been kind of sitting at the bottom in a memory market, 5-to-7 percent decrease in a quarter is considered ordinary.

You know, when the prices go down, they go down 20, 25 percent in a quarter.

I don't expect anything to happen because everybody is losing money, except us.

Because of the strategy we are deploying.

So I expect things to change much in that.

Last question just on inventory.

Your net inventory have crept up a little bit and days have crept up to 110.

I'm just wondering what your comfort range here is on inventory and inventory days?

We really have an excellent business model in terms of the current environment.

Obviously, we've talked about visibility and returns and to be able to effectively address customers needs that come to us, we have the majority of our inventory in di-stores.

Then we move those through to finished goods very rapidly from our back-ends productions in Thailand.

We also have a strategic inventory of finished goods on more high-running products.

Given the long life of proprietary product we have in microcontrollers and an a lock, we're not uncomfortable as to where we are in inventories.

The mix is really excellent at this point in time.

That does not give us any concern from an operating perspective.

To add more to that, Chris, our products have a very long life cycle.

We are still selling products in high volume that we first introduced in early '90.

That's the really wonderful micro-model and the analog products tend to be very similar, where they have very long life cycles.

So, you know, Microchip doesn't have the record of inventory write-offs the, you know, hundreds of millions of dollars inventory write-offs you have seen from other semiconductor companies and competitors.

Our inventory is doing very well.

It gives us strategic advantage to have very, very short lead times in a low visibility environment.

So we really are not uncomfortable with the inventory.

To go slightly higher, go slightly lower, you know, we are sitting in the probably eighth or 9th inning of the recession.

I don't know what everybody's expectation is.

Certainly the next leg of the semiconductor market is up not down.

At that stage, we're not concerned about inventory being high.

We think we are gaining market share because of that.

We are serving our customers very well.

We are not writing off any inventory.

Inventory is in very good mix.

That's how we like it.

I just wanted to make sure you were still in the comfort zone.

Thanks, guys

That's all the time we have for questions.

Guess, I'm turn it over to you, Mr. Parnell, for closing comments.

We appreciate the comments on the conference call this morning.

Steve and I will be in our offices in if there is any follow-up questions.

Thank you.

That will conclude today's conference.

We thank everyone for your participation.