微晶片科技 (MCHP) 2003 Q1 法說會逐字稿

Good day everyone.

Thanks for holding.

Welcome to this Microchip technology first quarter fiscal 2003 financial results conference call.

Just a reminder that today's call is being recorded.

At this time I'll turn the call over to Microchips' Chief Financial Officer, Gordon Parnell.

Please go ahead, sir.

Thanks, Kevin.

Good afternoon.

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

We wish to caution you that such statements are predictions and that actual events or results may differ materially.

We refer you to our press releases up to date, as well as our 10K for the fiscal year ended March 31st, 2002 filed with the SEC which identifies important risk factors which may effect Microchip's business.

In attendance is Steve Sanghi, Microchip's President and CEO and Dave Lambert, Vice President of Fab Operations.

I will start with a review of the June quarter financial results and then Steve will discuss the agreement that we have executed to acquire a semi-conductor manufacturing complex in Gresham, Oregon from Fujitsu Micro Electronics.

And then Steve will also comment on the June quarter and outline our guidance for the current September quarter.

All three of us will then be available to respond to any specific investor and analyst questions.

Net sales for the June quarter at $159.7 million were up 7.3% from the March quarter.

An increase by over 15% on a year-over-year basis.

The geographical mix for the quarter was basically in line with our expectations with Asia showing the strongest sequential growth at over 15% and the Americas growing sequentially by over 9%.

Business in Europe continued to be weaker than the other two regions resulting in a sequential revenue decline of approximately 4%.

The total revenue breakdown for the quarter resulted in sales for Asia representing 39% of sales, Americas 33% and Europe at 28%, obviously reflecting the sequential changes that I indicated earlier.

The revenue associated with the [INAUDIBLE] acquisition after June 5th are included in the overall reported information reflecting approximately half a percent overall with the majority of those revenues being recognized in the Asia region.

Net sales from microcontroller products represented approximately 79% of revenue in the quarter.

Analog products represented approximately 8% of revenue, and Manly products represented 13% of all revenue.

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

The diluted earnings per share for the June quarter before the end process reception development charge associated with the [INAUDIBLE] acquisition were 15 cents on 211.5 million average shares outstanding, an increase of approximately 17% from earnings per share in the immediately preceding quarter.

The charge for IPRD was $9.3 million.

Including the full effect of this charge earnings per share were approximately 10 cents.

Year-over-year earnings per share in the June quarter increased by over 39% prior to the effect of the in process R&D charge.

Gross margins in the June quarter were 52.5%, an increase of approximately 190 basis points from the prior quarter.

Gross margins in the June quarter were primarily impacted by the increasing waiver Fab loading levels, the product mix of microcontrollers and analog products, the flat pricing environment that we saw for memory products, manufacturing yields or fixed cost absorption as well as the competative conditions in the market.

Operating expenses were 27.2% of sales in the June quarter, compared to 28% in the previous quarter.

Research and development costs were $21.6 million representing 13.5% of sales.

SG&A expenses were 21.9% -- $21.9 million, representing 13.7% of sales.

Operating expenses overall showed an increase of $1.8 million from the March quarter, but leverage improvement from the growth in revenue resulted in the .7 improvement in overall operating expenses.

Microchip's total inventory position was $88.7 million which was essentially flat with the March quarter.

Inventory turns are 3.4 this compares to 3.3 in the prior quarter.

The date of inventory now represent 107 days down from a high of 127 days earlier in the prior fiscal quarter.

We continue to anticipate that the days of inventory levels will fall over the next several quarters.

At June 30th microchip's receivable days were 47. This was this line with receivable days at the end of the March quarter and compares with 50 days for receivable days a year ago.

Our total receivables were $82.9 million, which showed an increase of $2.5 million from the prior quarter, or approximately 3.1%.

The increase in receivables reflect the overall sales improvement in the quarter, the addition of receivables associated with the Paris Mart acquisition these were offset by the linearity of revenue in the June quarter compared to the March quarter.

Although receivables beyond 90 days continued to be extremely low with only 2.5% in the over 90 days category.

From a cash perspective as of June 30th, we had about $284 million on the balance sheet and there were no borrowings against this -- against any of the lines of credit.

During the quarter Microchip generated approximately $81 million in cash from operations which was offset by capital expenditures of approximately $25 million and $54 million associated with the aquisition of Paris Mart which we completed on June the 5th.

As I indicated, capital spending was $25 million for the quarter and depreciation for the same period was $28 million.

This is versus $27 million in the same quarter last fiscal year, and $26 million in the March quarter.

Microchip shipped approximately 10 and 1/2 thousand new application development systems to customers during the second quarter.

This created a new record for development systems in a single quarter.

We have now shipped approximately 237,000 systems representing one of the larger user bases of development tools in the industry to date.

I now ask Steve to give us some additional color on the June quarter, outline our guidance for the September quarter and discuss the significant business opportunities that we can achieve through the manufacturing facilities we have acquired in Gresham.

Steve?

Thank you, Gordon.

Good afternoon everybody.

As we reported in our press release today, Microchip has executed an agreement to acquire a semiconductor manufacturing complex in Gresham, Oregon from Fujitsu Microelectronics for $183.5 million in cash.

This Fab which will be named Fab 4 will be a significant addition to Microchip's waiver Fab capability, enhancing both the short and long-term capacity needs of the company.

The Fab is located on a 196 acre campus east of Portland, and has approximately 200,000 square feet of clean room space within the overall campus of approximately 826,000 square feet.

The facility comes equipped with approximately 350 process tools and 170 support tools, which facilitate the production ramp at the facility.

It will be our intention to produce 8" waifers in the facility utilizing our .5 and .35 micron processes. The facility is designed to support more than a billion dollars in annual sales when it reaches full capacity.

The close of the transaction is currently scheduled for October 2002, however the seller has an option to close the transaction as of the end of August 2002 if all the closing conditions have been met.

Assuming we close the transaction at the currently scheduled date, we expect to commence volume production in June, July of 2003 time frame.

Initial hiring will be about 60 people, which will ramp to over 100 during the pre production phase and we expect that the facility will employ approximately 360 people as the facility achieves high volume production over the next few years.

The Fab will also enhance our product road map as the facility is capable of operating at a process technology down to .13 microns.

We are planning to establish an advanced technology development program in this facility to develop a .25 micron process to support future product development at the company.

As we were examining our existing capacity requirements at Microchip and determining the anticipated demand for the products in fiscal 2003 and beyond, we needed to add significant capital in both our Tempe and Puallyup sites to expand our production in support of .5 and .35 micron technologies.

We initially started looking at used capital equipment at the facility.

Then we discovered the entire equipped Fab could be bought at a price where it will be more cost effective than buying the equipment and then moving it to Puyallup.

Our capital needs for just the next 18 months were well in excess of $180 million.

With the acquisition of the Gresham facility, we will be able to reduce our capital requirements over the next three to four years by approximately $400 million.

The impact on the longer term gross margins that the company will be extremely positive.

We have been guiding our long-term gross margins to be in the 53% to 55% range.

The Gresham facility and the equipment in it will allow the gross margin target to go up to 56% to 58%.

A significant enhancement to our already very strong business model.

We had $284 million cash on the balance sheet at the end of June quarter with $60 million free cash generation in the September quarter, and contemplating a close of the $183.5 million Fab at the currently scheduled date, which is at the end of October, we will have approximately $330 million cash on the balance sheet at the end of September.

Now, if the seller exercises the option to close as of the end of August, then a cash balance will be in the range of $150 million.

In either scenario, whether the facility closes at the end of August, or at the end of october, our cash balance at the end of December will be approximately $200 million.

After this purchase, our total capacity -- our total capital expenditure for the next three years is expected to be only about $70 million per year.

And this is a real benefit of this Fab. Very, very substantial savings in the capital expenditure.

You know, for the next three years, we will only be spending about $70 million per year.

Therefore, we expect to be generating a free cash flow of about $50 million per quarter or about $200 million per year for the next three years.

This makes this Fab purchase a very cost effective solution to the otherwise very capital intensive nature of this business.

The acquisition of the facility is subject to several closing conditions.

We have received excellent cooperation from city, state, and county officials to date and look to continue the dialogue to complete the establishment of a strategic investment program which will cap the property taxes at the site.

This is certainly one of the key closing conditions that we need to resolve working with the authorities in the Portland area.

The obvious question now is what happens to Fab 3 in Puyallup, Washington?

There are two parts to the assets in Puyallup.

First is about $80 million of equipment.

Microchip will relocate the equipment to our existing Fab 1 and Fab 2 facilities to provide additional capacity until Fab 4 is in production. Some equipment will also move to Fab 4 to provide a closer equipment match in Fab 4 to our existing process equipment.

So the entire $80 million equipment in Puyallup, Washington will be used up mainly in Fab 2, but some also in Fab 4.

The second part of the asset is Fab 3 building itself, Microchip intends to maintain this facility for future manufacturing requirements. We'll also be evaluating the facility to determine if an impairment charge will be required.

We will update our investors when additional data is available on this.

Now let me move to the second part of my comments today which are on the various product lines.

First microcontrollers.

Microcontroller business was a solid performer with over 6% sequential growth in reaching an all-time record in net sales last quarter. During the quarter, we also shipped a record 10,500 new development tools surpassing an old development system record of 9,500 systems per quarter.

This continues to point out that our microprocessors are getting very rapid design substance in our customers' applications.

We are also eagerly awaiting the release of Data Quest year 2001 market share numbers which are expected out any time now in the next couple of weeks.

These numbers we believe will show that while microcontroller market was down nearly 33% last year, Microchip's eight bit microcontroller business was down in single digits and hence we continue to gain substantial market share.

Second analog business.

Microchip's analog business was up 16% sequentially.

The other important point is that the cellular phone portion of this business was actually down.

Therefore all the growth came from expansion of our analog customer base and selling a lot of analog products to a large microcontroller customer base.

I must say that analog is still a relatively new business at Microchip and the growth is lumpy.

The investors should not analyze one quarter's growth.

Having said that, I could not be more pleased with the results that we are achieving in our analog business.

Let me give you a few more highlights about the analog business.

Motorola is now only 9% of our analog revenue, down from a high of 30% pre acquisition of Telecom Semi-conductor.

So Motorola is really down to only 9% of the analog business and less than really a percent of Microchip's overall business.

The cell phones are 21% of analog revenue down from a high of 54% pre acquisition.

51% of our analog sales are attached to our pick microcontrollers, designing opportunities -- excuse me.

Designing opportunities being tracked now exceed $500 million in the analog business.

70 to 80% of products released on a quarterly basis in the analog business are now totally proprietary products, and 80% of the analog revenue now is coming from Microchip's internal wafer Fabs.

So we have made substantial progress in really consolidating that manufacturing inside resulting into substantially lower cost and higher gross margins.

Turning to City Lease [INAUDIBLE] business. City Lease [INAUDIBLE] business was up 8% sequentially on higher units and flattish ASP.

We have seen lead times from certain competitors in certain geographies and on certain products move up.

Therefore there are scattered dislocations on the supply chain.

Microchip has been able to move certain prices up.

We do not expect the Esquare business to make dramatic moves in either direction short-term, up or down.

PowerSmart.

PowerSmart's net sales last quarter were about $2 million, but only about $900,000 of that happened on our clock since June 5th.

We expect PowerSmart sales to be between $2.5 to $3 million in the September quarter, the backlog is already about $1.5 million.

Finally, let me provide investors some guidance for September quarter.

Based on strength in microcontrollers, analog products and PowerSmart, the stability in City Lease Square and geographical strength in Americas and Asia, combined with the usual weakness in Europe during summer, we expect net sales in September quarter to be up approximately 7% sequentially.

The gross margin percentage is expected to stay in the same range as the June quarters.

Earnings per share in the September quarter is expected to be up 1 cent to 16 cents per share.

With that, Kevin, will you please poll for questions.

If you have a question today, please signal us by using your telephone pressing the star key followed by the digit one.

Again, that's star one on your phone to ask a question. And we'll pause very briefly.

First up in the roster will be Chris Danly at Merrill Lynch.

Hi, guys, congratulations on the acquisition.

A couple quick questions.

What's the turns percentage needed for the guidance this quarter?

Chris, it was last quarter it was 57%.

In the September quarter, the current quarter it will be approximately 50%.

Okay.

Great.

And then, can you guys spend a little bit of time on microcontroller inventory out there?

Sort of what you are seeing in the channel, and what you think we are going to be looking at going forward?

Channel inventory from our direct customers, we -- from what information we have hasn't changed much at all.

It continues to be very low.

There is no reason for our direct customers to be having higher levels of inventory of our proprietary products.

That's never really been a substantial issue for us as it has been for some other parts of the industry.

Our distribution channel inventory overall went up very modestly in the June quarter from 2.4 months to 2.6 months.

Obviously a portion of that would be microcontrollers.

But that's still in the very lower end of the range.

Over the last three years, inventories have been between 2.4 to 3.7 months.

Inventories are still very lean out there.

Okay, great.

I guess, are you expecting more relative strength from micros or analog in the September quarter?

Well, Chris, the microcontroller is a very substantial business.

79% of our business.

And, you know, with many thousands and thousands of customers, and it's much more stable.

It has behaved extremely well, even at the depths of the recession.

Analog is a relatively new business.

As I mentioned it tends to be lumpy.

If you look at last four or five quarters we have had huge growth, sometimes less, sometimes it is more.

We cannot as, predictably, give you growth numbers for analog.

We expect it to grow again very handsomely.

As for the microcontrollers are concerned, we think microcontrollers will drive a good of the growth.

Okay, great. Have lead times changed at all on the products?

No, the lead times are pretty low.

Two to four weeks on most products.

Great. Okay.

Thanks a lot, guys.

Moving on we will hear from Doug Lee, Banc of America Securities.

Hi, congratulations on a good quarter.

Couple of questions. I was trying to get a better appreciation for the Fab.

I hear that you are buying.

It looks like it's off the cuff, automatically submicron capable.

$183 million.

If you were to build a Fab, Steve or Gordon, if you were to build a Fab with a 200,000 square foot clean room submicron capable with the similar process tools in place, what would that cost you today?

We'll have Dave Lambert answer that.

He's our VP of Fab.

The way you would look at costing a factory like that.

We think it would be about $6,500 a square foot of clean room and there's 200,000.

This would be well over a billion dollars just just for the building.

And then, of course, in addition to the building there's a significant installed equipment set.

You have the value of the equipment, plus the lease hold improvements to install the equipment.

It's a pretty significant value.

The equipment, we believe it is close to $800 million.

If we had to buy knew.

If you had to buy new.

There's a very, very substantial value in it, somewhere in the 1.7 to $2 billion range.

Obviously for brand-new.

And for Microchips process requirements this is state-of-the-art equipment because it's on the leading edge of technology. This is absolutely ideal.

I got it.

That's terrific.

The lowest wafer cost out of any of Microchip's Fabs and it is such wafer cost driven acretion to which we are guiding that our gross margins will rise to 56% to 58% when we're there [INAUDIBLE].

Will there be a short-term impact as you start to retool and move things around?

How should we think about margins at least over the next couple of quarters?

The margins for the next couple of quarters should be flat.

After we close the purchase on the facility, the expenses we will have in this facility will be totally balanced by the acreation coming out of the unutilized capacity in Chandler and Tempe.

As the revenue is growing, our inventory is about 107 days today.

Over the next couple of quarters inventory comes down even more.

Then we start ramping the Chandler and Tempe facilities to provide the growth to the company.

That almost perfectly balances the expenses coming into the Oregon facility while the facility is not in production.

So we are guiding that our gross margin should be flat in this range for the September, December and March quarters.

You know, for the next three or four quarters.

As the facility comes to production starting June, July of 2003, then the gross margins start to become very good very rapidly as we start to absorb the facility and they get up to 56% to 58% in the subsequent six to eight quarters.

Great. I understand.

That's terrific.

On the operating margin side, I imagine do you have a long-term operating margin goal say out six, seven quarters?

Do we see where working margins go?

I think I have been thinking of a 30% range longer term.

Operating expense is in a longer term model in the 24% to 25% range.

So that would get you in the high end to about 33% operating margins.

30% to 33% would be the longer term guidance.

Okay. Terrific.

Then just very last thing then I'll let other people ask questions.

The cash. You spent some time talking about cash balances, et cetera.

We should infer from this discussion that at this time you have no plans to do any additional outside financing, this is totally self funding?

This is totally self funding.

We have $284 million cash right now.

We are planning to generate another $60 million this quarter.

We will be about $330 plus, we he can pay $180 out of that and still have significant cash.

Going forward the CAP-X requirements are only about $70 million a year which will generate in net positive cash flow of about $200 million plus every year for the next three or four years.

Got it. That's excellent. Thank you.

Great quarter, guys.

Thanks.

From William Blair and Company this is Jeff Rosenburg.

Hi, Jeff.

Hi, how are you?

First on the bookings front, it sounds like the book to bill was positive if you saw the turns percentage go down.

Could you give a little detail there on bookings growth sequentially or book to bill or however you want to articulate it?

Book to bill was 1. Close to 1 - 1.9.

It was strong really through the entire quarter.

And in all product areas.

So it really helped to improve the visibility, which is -- the visibility is, as we often say on these calls is often a little bit over played by the street.

It certainly enables our manufacturing organizations to work more efficiently and effectively knowing what's ahead of them.

And so that's -- that is very helpful in the environment that we have.

Great. Okay. Then just wanted to clarify one thing.

When you talk about the eventual capacity of the new Fab being a billion dollars is that with the equipment you are purchasing for the $184 million or is there additional expenses - expenditures that would need to get there in the long run?

The additional equipment we would need in that facility is really fairly small $20 to $30 million.

That may be a high number going forward. There is not a

Per year.

Per year.

There is not a perfect equipment match to what we have.

So -- because the process Fujitsu was running was a different process.

Matching to our process, there are few incremental pieces of equipment we will need every year, but they are very, very small.

For a company of our size, as we grow from here our capital expendature and Fab would be $175 million and up from there per year, ordinarily.

And purchase of this facility will bring the Fab capital expenditure down to $30 million per year or so for the next several years.

So it's fair to think of this as like $250 million or so in rough numbers to get a billion dollars in capacity?

Pretty close.

You are pretty close.

I think that's -- just -- probably $250 to $300, yeah, which is really a very sweetheart deal.

Ordinarily to get a billion dollars you require about a billion dollars of investment.

Great. Okay.

We are only talking about Fab here.

The conservative number we gave today is over the next three to four years it saves us $400 million of capital investment.

Okay.

And in terms of depreciation and how that'll phase in, historically you have -- I think you have been able to have the depreciation come on line as you actually start to produce with certain equipment, given the fact that you are buying so much equipment at once and the way the lines are laid out is there any color there in terms of understanding how the depreciation might come into play differently than it has for you historically?

The equipment will be brought on production as we need it.

And our current take is when we start the production in July of next year, we start with only about 17% of the equipment.

Then as we, you know, ramp every quarter, more and more equipment will trigger depreciation.

The equipment is worth about $800 million, probably we'll assign somewhere in the $80 million range to the equipment out of the $183.5 we are paying.

So it's really 10 cents to a dollar. So it's already impaired.

It's already at an impaired value sitting there.

So equipment which will be not being used right away will not require any other treatment.

It will simply be kept.

As we start production on it, we will start depreciation on it.

We'll actually go through a full appraisal on the site. That will be part of the closing.

Value will be assigned based on the appraisal or appraisals, there may be more than one, to line buildings and lease hold improvements and the tools.

So that will be a very lengthy process that we'll go through and from there the depreciation for each tool will be assigned and go forward from there as Steve described.

Is it hard to say at this point whether there might be some short-term dip in gross margin, or do you feel like the way you'll bring it on you can absorb it pretty much like you always have?

It will be absolutely should be no dip in gross margin.

The math we have done is when we bring the first 17% of equipment to production, that's really nothing.

You are talking about, you know, what, $14 million of equipment to bring to production?

With $14 million in production you couldn't do anything in a Fab.

So it's actually immediately accretive.

The gross margin goes up the first quarter you start production.

Okay.

Great.

Last question is, when you talked about where you expect to be in September '03 in terms of, you know the capacity you have today and where that would bring you to and you were discussing some of the modeling you have done in terms of gross margins and utilization as that is affected by the early expenses from the Fab.

Without getting into guidance beyond the current quarter, can you give us any flavor has to what sort of revenue growth you are expecting over the next year or so? I mean is it -- should we expect some seasonality as we make our way into the end of the year like you have often seen?

Or do you believe you can maintain a comparable sort of revenue growth rate we have seen in this quarter and last going forward?

A lot of questions in there.

Let me try to take a shot at it, Jeff.

As for the guidance is concerned, we are only giving one quarter guidance in revenue of 7% growth.

Now, if you go beyond that, you know, usually in the Microchip March quarter tends to be slightly slower.

But it's usually slightly slower because ordinarily there is inventory build at the end of the December quarter.

Then you go into some inventory depletion in the March quarter, the Chinese new year comes in and all that equasion drives the March quarter to be slightly softer.

So if you are growing 7%, March quarter usually is 5%.

We didn't see soft March quarter this year because inventories were so low out in the channel and we expect that the next March quarter may not be bad, either.

Maybe fine and maybe just lowered by one or two percentage points.

So you can do your own math on what the growth numbers could be.

The other equation in that is, you know, the whole industry is not doing as well.

And some day, somewhere this industry has to kick in high gear.

You know, we are doing these kind of growth numbers, look at Intel's numbers, while the industry is not doing well.

So I think it's quite possible that when the industry catches up and starts growing, Microchip does better, which we have done in the past cycles.

Having said all that, our current expectation is that by the end of the September next year, Fab 1 and Fab 2 facilities will be full.

We will have taken the Puyallup equipment, moved toTtempe, brought it to production, we are using it, ramping it and it has been used all up.

The Oregon facility starts production in July.

From there on, you know, majority of our growth really comes out of the Oregon facility.

And when you say full, I've typically thought of that as billion dollar run rate.

Is that still the right number?

8 to $900 million is probably in the range.

Changes with the product mix and the new technologies going through.

Changes with the product mix but I think the biggest change is the technology.

The Tempe facility in the mix we had two years ago probably could do 900 to a billion. You know, the mix has changed where we have more advanced technology requirement.

And to put that advanced technology in Tempe will cost many more dollars than would be the right thing to do today.

Because that equipment is all available in Oregon.

So it actually, Tempe will cap out while we still have some room, space to put equipment in.

And over time, over the next two or three years, we might fill that.

Tempe would be what you call 90% full?

It would be about 90% full, there would be some more space, but we do not need to make the investments because there's equipment available for very inexpensive cost in Oregon.

Okay.

That's helpful. Thank you.

Before we move on, I would like to remind everyone to press star one if you have any questions.

We'll move on to Hans Mostsman at Prudential Securities.

Thank you and congratulations on the proposed acquisition. Looks like you'll get it for a steal.

Question regarding this acquisition.

What brought you to actually look for a new Fab with that Fab in Washington?

Seems like you had a lot of capacity.

Can you give us more detail there?

Then I have some follows up. Thanks.

Well, you know, we had a model for what the Puyallup Fab, Fab 3 would look like. How much equipment would be needed.

We needed to bring up more advanced technology there. We were going to do it two years ago.

You know, that was going to require some construction of the CMP room and all that the capability we didn't have in Puyallup.

We had a model of what it would take to bring Puyallup essentially to production.

We went out looking for essentially equipment.

There is some used equipment available in the market, and we looked at this facility as well as, you know, many others and it rapidly became clear that, you know, we could buy the hole turnkey facility equipment, all installed, lease holds and everything without having to move the equipment and pay the price, installation in Puyallup and all that at this price, which is probably lower than I would have spent in the next 12 to 18 months, anyway.

I mean that's how it came about.

I see.

The -- on the product side of the equation, the lead times tightening or going out for some of your competitors, what is driving that on the serial E squared side of the product line?

Well, you know the information is very sketchy.

That's what I tried to say.

Certain products in certain geographies at certain customers.

It doesn't seem to be very broad based at all.

There's spotty problems here and there, you know, where competitor x isn't shipping this and competitor y isn't shipping this.

There's a problem in Asia.

This customer in Europe can't find this.

And I think it's just driven by, you know, people have shutdown, moth ball there capacities and as people come back, there are temporarily dislocations of either labor, equipment, or piece parts, or assembly or test.

I do not know.

Our lead times are always short.

So you probably have to ask the competitors why they are scattered dislocations.

Okay. One more.

In terms of end markets, consumer, wireless, industrial, government and the others, what -- which ones seem to be the better positioned here and drivers for your business growth in the second half calendar year? Thank you.

I think the best way I will describe that is the four business segments of consumer, automotive, industrial and the office automation look very strong to me.

And the weakness is in the telecommunications area which is about 17% or so of our business. Maybe slightly less than that now.

Since, you know, the cellular phones are substantially lower, so I believe that may have shrink to more like 15, 16% of our business.

85% of our business is quite strong.

You know, let's pick one.

Let's pick out a model.

We have shown to investors in presentations how many more different models of cars and applications we have design wins for model year three cars versus model year two cars.

There's a substantial increment in model year three cars versus model year two cars.

Model year three cars go to production essentially now and they are in show rooms by labor day.

So Microchip is seeing substantial new designs in automotive turn to production as we speak.

Secondly, with the return of the zero percent financing, the indication we are getting from our own manufacturers is they are increasing the number of cars they will build in this year by about a million and a half cars more than they were going to build in the U.S. alone.

So we are getting higher unit volume projections.

Street sometimes is worried about the margin squeeze of the automotive customers because of the zero percent financing.

That's not having any impact on us. We are selling proprietary products with higher units and more model year penetration.

So that piece looks very good.

Consumer portion of the business looks very good with thousands and thousands of design ins.

Industrial business has stood very well.

The other questions you may have is regarding the PC business, especially looking at, you know, Intel's release.

In PC business, investors should understand the difference between desktop sales versus the laptop sales.

The weakness is in the desktop sales.

The laptops actually are growing.

And Microchip parts are in laptops.

You know, we have chips in power management, microcontrollers on the break, we have chips in mouse and peripherals and printers, you know and the other elements of the laptop like fan control and all that.

And lots and lots of new designs as well as the laptops are growing.

So our PC section of our business looks quite good.

Quite good.

Great. Thanks a lot, Steve.

We have a question from Edward Hinyelguard at Shakier Investments.

Hi, Steve.

Just kind of a little bit more of an elaboration on the -- from a longer term standpoint about where you would expect to see growth.

I mean a Fab with a billion dollar production capability is significantly above your current capacity.

Over the next say three or four years, where do you -- do you expect to see the growth from just your existing product lines or do you have any intentions for, you know, new areas that might really be driving growth?

Well, let me say this.

I'm going to work like hell to fill that for you and for all the investors and for ourselves.

You know the Fab is accretive as soon as it begins production because the way it is being purchased and the way we are going to start, you know production and depreciation on it.

And as you use more and more of the Fab, it becomes more and more and more accretive.

In fiscal 2005 time frame it starts to look extremely good which is almost calendar year 2004 which is not too far away.

Can I fill it even faster? Your question is if I acquire some other companies or get some more revenue and all that?

That's hard to comment on.

We are constantly always looking for complimentary opportunities.

As you know, they are difficult to find and most of them are delutive and you get a lot of junk along with some good stuff.

We are very, very selective.

Our model over the years has been if you buy something, we buy along the technology lines something that's really strategic like the analog was and PowerSmart, which was a very small business, very few people that you can bring into our culture.

We are really not the kind of company that buys two, three, $4 billion worth of companies and don't expect that kind of acquisition.

Okay.

You haven't talked about you have your, I guess I think you indicated in 2003 you expect to begin introducing products in the combined --

Signal controller?

Right.

Okay.

DS6. Yes. That product is basically should be sampling later part of this year, and this Fab, Oregon Fab will start up on our current pick microcontrollers and some memory products, and then very quickly the digital signal process -- digital signal controller will go in that Fab, definitely.

Have you gotten or can you give anymore color on your outlook for that product line?

You haven't talked about it in about six months.

You know, it -- it's doing well.

It's still in design phase.

We are not shipping product yet, so kind of hard to comment on.

But it's total opportunity management system has grown well passed $100 million already.

You know, it's -- you know, customers think it's just the absolutely right product and they are chomping at the bit to get silicone in their hand and we are getting very, very close.

Okay.

Any, just one last question.

Any potential to, you know, use this capacity to say more dramatically expand your flash business?

Flash what? We don't -- flash memory?

Your EEPROM, your E square business?

No. No. We don't make any flash memory.

Let me say that so there's no confusion. We have no flash memory business.

We have a serial E square prom business which is currently 13% of our business.

That's the business we talked about. It's kind of stable today.

I don't expect major moves on that business in either direction.

It should stay about 13% of our business, over time should come down as the others grow harder.

But that business is not that elastic.

If you tried to make any meaningful changes in the business you create the price down through the earth.

Okay. Thanks.

Continuing we'll hear a question from Mark Edellstone from Morgan Stanley.

Good afternoon, guys. And congratulations on the quarter and what looks like a home run acquisition for the Fab as well.

Steve, you gave a number of data points on it.

I wonder if you have run the numbers and could share it with us on what the difference in wafer costs are going to look like once you've got that facility ramped up versus what the prices are like now at Tempe and Chandler?

Mark, I think everyone of my competitors will listen to that call 10 times to learn that.

Nobody will ever know that.

Sorry, but, again I can put that in the terms of, you know, what that capability would be.

You know, you can see that if you are doing, you know, billion two -- pick a number.

If you are doing billion two of revenue, about $400 million will be coming out of this and $800 million will be coming out of our existing Fab 1 and Fab 2 at that kind of level it will be producing approximately a third of our output.

Its wafer cost is so good at that time we will safely be making what, the high kind of gross margin --

Into the 56 to 58% gross margin.

Into the 56 to 58% margin. So that's the intact. Now, there is assembly in there, there's test in there, there's proprietary products.

There's all these lots and lots and lots of moving parts.

What portion of the business is serial E squared and I don't think I could break that model out for you that minutely.

But the waifer cost is very good.

If you have any assessment from the industry what portion of the waifer cost is usually capital depreciation, then you can really do some math and look at what the capital depreciation would be in this Fab, which is what it would have been at the full price.

Okay, great. That's helpful.

The other question, I guess, Steve when you look at dynamics of this, you are buying this at a price that your capital intensity is gonna change pretty dramatically here over the next several years.

Your cash balances are going to move up to levels that are unprecedented for the company.

You guys have historically run the company with pretty lean levels of cash.

What -- can you give us your observations on how you might look to run the business differently in the future?

And how you think the implications go for you?

Well, you know, basically what we are seeing going forward here, short-term is we are going to be producing $50 million plus of cash per quarter.

You know, as the company gets larger in a year or two years, that number goes higher.

So start with a $200 million per year of cash generation and up from there with a growth rate, you start to have very, very substantial cash balances two or three years from now.

And, you know, what we do with it, you know, as we get there, we will find an answer.

There are always opportunities of, you know, buying stock back, to, you know, strategic acquisitions, to, you know, leaving a higher cash balance than we historically left, to investing in a portion of the cash higher out on the a yield curve so you get more return for it.

I don't have that problem today.

I'll have $150 million of cash after this quarter. Then it builds up from there.

Once it goes passed 300, $400 million of cash, you know, then we'll worry about it. That's not the problem today.

Last question.

When you look back from the beginning of Microchip, you guys have done a great job of expanding your product line, and obviously analog is now an area that you are focusing on.

As you look at opportunities at this new very low cost manufacturing operation is going to give you, is it your expectation that there might be some other product areas that, before now, would not have made sense, but now make sense today with what should be an incredibly competitive cost structure?

Um -- the answer to that could be yes, but I think it will continue to be in our core and better controlled business, we could expand more into the 16 bit micro, the DS6 product line into some areas where it otherwise may not have made sense, could be opportunities in the 32 bit area, there could be opportunities, you know some opportunities in doing some asic portion of the business from our current product lines of 8 and 16.

It could provide some of those opportunities.

I think what I would not want to do, probably, is more memory business.

But we will continue to look for more opportunities in the proprietary business.

Great. Sounds like it gives you a lot of flexibility.

Congratulations, once again.

Thank you very much.

Next up is Tom Smith, Standard and Poors Equity Research.

Yes.

Hi, nice quarter.

Thank you.

I was curious about these developments that sounds like you are moving along a nice number of them, which presumably encourages new designers to start working with your products.

I wondered if, when you ship these things, are you handing them to them, or is this a proactive decision by the designer to ask and pay for these?

Is it something they need a new one of every so often?

Um -- these development systems are sold.

It's a portion of our business. Have we said what it is roughly?

What is it?

$2.5 million range.

2.5 to 3.

Yes.

It's about a 2.5 to $3 million quarter business.

We sell them.

Now, you know, could you take one and give to a large strategic customer, discounted or for free?

Sure, once in a while we may loan them or give it free.

But large majority of them are sold at a fair market value.

That portion of the business is very profitable. So it's not like we are losing money on the development tools.

They are sort of razors and then you want to sell a lot of blades, but we are not giving away razors, they are sold.

Okay. That sounds like they have a sincere interest in wanting them. How often would they need to buy an update?

You know, if you start from the beginning, you know 12 years ago, we have come up with about 3 to 4 very significant upgrades to the tools.

Probably about three upgrades to the tools where you come up with a brand-new system that next generation.

We recently came up with one in the last six months which works on a 32 bit platform.

It's state-of-the-art work system.

So when that happens, you know, many customers, especially the larger volume customers who can afford it like to upgrade their system because it's faster, gives them faster ability to really develop the tools.

Many, many old systems continue, but at some point in time, as you come up with the new products, you can no longer support the older systems on the new product.

At some point in time that compatibility becomes a nightmare to bring it up on four different platforms.

So for example a DS pick as we are bringing it up it will be brought up on our most advanced generation system.

Any customers who do not acquire that, you know, cannot do DS pick.

Many times it's a combination of bringing a new system up and new products up and it's a complex equation as the systems have to be upgraded for new products.

Right. Ok. Thank you. That's very helpful.

You're welcome.

And again, a reminder to press star one with your questions.

We'll move on to Tim Mahon with CSFB.

Thanks. Nice quarter guys.

Um - How long will it take to call a new Fab for automotive quals?

Um -- basically it will be about a quarter longer than you qual for everything else.

It takes about one quarter longer.

So we will be in production in this Fab in June, July next year and when we first bring it to production, we'll be shipping it to all of our other markets other than automotive which is, you know, 81% of our business is other markets.

It will take about a quarter longer to qual it.

Could be four months.

Okay.

Maybe one for Gordon. Head count at the end of the quarter?

About 3200 people.

3200.

And then just finally, I'm sorry if I missed this.

On the Esquared business, I read your prepared comments from the standpoint you said prices were flat, I think in kind of during the conference season you guys had been out saying there were some spot increases here and there.

Correct me if I'm wrong, but did something happen in the month of June that you saw prices start to trend back down a little bit that caused prices to be flat sequentially? Thank you.

I think during the quarter we were seeing prices could be flat to modestly up.

It all really depends on the sale through, in terms of specific densities and specific areas.

It's very difficult to track a business that is as volume sensitive as that is on a real time basis.

So didn't really see any substantial change in the June quarter that we would reflect in the September quarter.

We feel in the September quarter prices will stay pretty much in the range that we achieved in June.

With a -- you know, with a very minor change in mix from a 256 Esquared to a 2 k or 4 k esquared, you know, can substantially change the average ASP of the product line.

If the question will be, "Will the scattered ASP's went up with a given product?" the answers are yes.

When you took the entire business together from 1 k to 256 k and add them all up, seemed like there were enough mix changes that the business was almost flattish, but the gross margin was up.

Ok. That makes sense, Steve. Ok. That's kind of what I was looking for. Perfect. Thank you very much.

The gross margin was up.

Gentlemen, at this time there are no further questions in the roster. I'll turn things back over to you.

Thank you very much, Kevin.

We would like to thank everybody for the time this afternoon.

Steve, Dave and I would still be available later today if you have any follow-on questions.

I thank you very much for your time today.

That concludes today's conference call.

Thank you for joining us and have a good day.