微晶片科技 (MCHP) 2006 Q3 法說會逐字稿

Good day, everyone.

Welcome to this Microchip Technology third quarter and fiscal year 2006 financial results conference.

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

At this time I would like to turn the conference over to Gordon Parnell, Chief Financial Officer for Microchip.

Please go ahead, sir.

Thanks, Sarah, and good afternoon, everyone.

During the course of this conference call we will be making projections and other forward-look statements regarding future events for 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 release of today, as well as our 10K for the fiscal year ended March 31st, 2005, and our AK current reports that we have filed with the SEC that identify important risk factors that may impact Microchip's business and results of operations.

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

I will comment on our third quarter fiscal 2006 performance, reviewing geographic data and discussing balance sheet and cash information, and Steve will then give his comments on the results, outline our guidance for the March quarter, and update other pertinent matters regarding our business.

We'll then both be available to respond to investor -- specific investor and analyst questions as required.

So, net sales for the December quarter were at record levels of 234.9 million, up approximately 3.3% from net sales of $227.3 million in the immediately preceding quarter, and up approximately 14.4% from net sales of $205.4 million in the prior year's third quarter.

Our GAAP net income for the December quarter was at record levels of $70.7 million or $0.33 per diluted share, an increase of 7.7% from GAAP net income of 65.7 million or $0.31 per diluted share from the immediately preceding quarter and an increase of 33.1% from GAAP net income of $53.1 million or $0.25 per share from the prior year's third quarter.

Geographically, sales in Asia and Americas grew by approximately 7.6% and 1.7% respectively, with Europe being down 1.8%.

Asia achieved revenues of approximately $108.6 million for the quarter, representing about 46% of our total revenues.

Customer designs from both the Americas and Europe continue to be a growing portion of our Asian revenues.

We established record gross margins for the December quarter of 59.7%, reflecting the proprietary product positioning and rich value-added products that Microchip has in the marketplace, world-class manufacturing yields and efficiencies, and richer product mix.

Our operating expenses were 23.7% of sales in the December quarter compared to operating expenses of 24.3% in the previous period.

Research and development costs were $23.4 million, representing 10% of sales.

Sales and general administrative expenses were $32.3 million, representing 13.7% of sales.

The tax rate for the December quarter was in line with our guidance of 24%.

The dividend declared today of $0.19 was an increase of approximately 19% sequentially, and an increase of 171% over the same quarter in fiscal 2005.

Microchip's total inventory position at the end of December was approximately 114.6 million, an increase of approximately 4.4 million from the prior quarter.

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

Distribution inventory is continuing at very moderate levels in all geographies, and as of the end of December, distributors held about two months of inventory.

Given the current industry conditions and the level of inventories at distribution, we are comfortable with the overall level of inventories which support our business today.

At December 31st, Microchip's receivable days were 47 compared with 48 days at the end of September.

Our overall trade receivable balance decreased approximately 4.8 million, or 4%, versus our revenue growth of 3.3.

At the end of December Microchip's cash and short-term investment position was $992.4 million, with $45.4 million of debt on the balance sheet.

During the quarter, Microchip generated cash flow from the business of $126 million prior to the dividend payment of $34 million and $3 million in stock buyback activity.

Capital spending was approximately $17.6 million for the December quarter, and depreciation expense for the same period was $26.8 million versus $30.2 million for the same quarter last fiscal year and $27.6 million in the September quarter.

For the full year, we anticipate capital expenditures of $60 million and depreciation of $110 million.

Two other items I'd like to touch on today.

We have indicated in our 10Q filings the ongoing discussions at the Company related to our considering repatriation of earnings under the American Jobs Creation Act.

Under the Act, Microchip can repatriate up to $500 million in earnings from her foreign subsidiaries in the United States.

While the analysis is not complete at this time and Board action has not been confirmed, I want to remind investors of the tax impact should the Company decide to repatriate earnings at a later date.

The repatriation would result in a substantial reduction from the approximate $180 million tax expense that would be charged without the Act if the Company decides to repatriate the maximum permitted amount.

Under the Act, we anticipate the additional tax expense would be up to approximately $34 million.

The tax expense would be based on the actual amount of cash the Company may elect to repatriate and would be charged in the appropriate fiscal and financial quarter.

Should management decide to do a repatriation, the decision will be effective once the Board ratifies the results of management's analysis and the dividend is declared.

The second area I would like to touch on relates to the adoption of option expensing, FAS 123R.

Microchip will adopt FAS 123R for fiscal years beginning April 1st, 2006.

Based on investor feedback, Microchip's board decided to use restricted share units, RSUs, as the equity vehicle to incentivize employees.

RSUs will replace stock options as the primary longer term compensation vehicle at Microchip.

Based on our analysis of FAS 123R implementation, we anticipate an earnings impact of 7 to 8% during fiscal 2007.

We will report earnings on a GAAP basis as well as excluding the effects of the stock compensation costs to help our investors and analysts to understand the effects of the adoption of 123R.

With that I'll ask Steve to discuss the performance of our business, our guidance for March, and other matters as necessary.

Steve?

Thank you, Gordon, and good afternoon, everyone.

I would like to apologize to our investors on the east coast for the timing of this call.

It was a crowded calendar today with many companies announcing; therefore, we had to choose this time.

I would also like to wish all of our investors and analysts a very happy and prosperous new year.

I hope that we help you make money with a 25% appreciation including dividends in 2005 and 131% appreciation over the last five years.

And we will continue to deliver you incremental value going forward.

Today I would like to reflect on the results of the December quarter, then comment on the product lines, and finally discuss the guidance for the March 2006 quarter.

So let's begin.

I'm extremely pleased with our continuing excellent execution and delivering yet another record quarter in every respect.

We exceeded our guidance in net sales, in gross margin percentage, and in earnings per share.

Net sales were $234.9 million, which was an all-time record.

Gross margin of 59.7 was also a record and was up 50 basis points sequentially.

Our operating profit was another record and exceeded 36% for the first time in our history.

Earnings per share was $0.33, exceeding our guidance by $0.01 and also set another all-time record.

We also produced a record free cash flow of $126 million before the dividend payment of $34 million last quarter.

We are pleased to be increasing the dividend paid to our shareholders by about 19%, to a record quarterly dividend of $0.19 per share.

Our dividend rate was already at the highest level in the semiconductor industry, and with this increase we continue to further increase the value delivered to our shareholders.

Now I shall talk about the product lines.

First, Microcontrollers.

Microcontroller business grew 3.5% sequentially in an otherwise seasonally weak quarter for Microchip.

Microcontroller business was up 12% over the year-ago quarter.

Our Flash Microcontroller business was very strong and grew by a robust 8% sequentially and 38% over the year-ago quarter.

Flash microcontrollers now represent over 55% of our microcontroller business.

We are continuing to experience strong design activity with our 8 bit Flash microcontrollers and we are seeing a significant number of new applications emerging, which are utilizing our products.

We are very well positioned to continue to gain market share in this segment.

I want to thank our shareholders for ignoring the [SIA] data, number of cars built, as well as data points from certain competitors.

The skeptics have been consistently wrong and have done you a disservice.

This is the first time when I did not see the stock get whipsawed because an uninformed someone had something to say.

I ask you to continue to ignore such data in the future as applied to Microchip's microcontroller business.

Our 16-bit microcontroller business was up 19% sequentially in December quarter, albeit from a small base.

At the end of last quarter, we had 233 customers in production versus 199 in September quarter and 165 in the June quarter.

Based on that metric, the customer penetration is proceeding as expected.

We shipped 748 new development tools last quarter for a cumulative total of 11,387 development tools, supporting our 16-bit products.

Based on the momentum with customers we see, we expect 16-bit microcontrollers to be up approximately 15 to 25% sequentially in the March quarter.

Putting a feather in our cap, our PIC24 16-bit microcontroller architecture won two key awards last quarter.

One award was the best microcontroller of 2005 by EETN Magazine in their 2005 product review and outlook supplement, and in the second award, PIC24 was named co-winner of one of the 2005 products of the year by Electronic Products Magazine.

Serial E-square memory products net sales were down 4% sequentially, but up 16% over a year-ago quarter.

Pricing remained firm, with our average selling price in the quarter actually up 1% sequentially.

Analog products were the sequential growth winners for the quarter.

Analog products net sales were up 18% sequentially and up 40% over the year-ago quarter.

We believe that we outgrew nearly all of our analog competitors and are gaining market share in analog.

The design momentum on analog products has been very strong and we see continuing growth ahead.

Our backlog for analog business going into this quarter is quite strong and we expect a growth of about 4% to 6% in analog, which is much better than typical seasonal pattern for March quarter.

Now I shall discuss our guidance for the March 2006 quarter.

As we look at the March quarter, we took several factors into account.

Our March quarter is usually a seasonally weak quarter for Microchip, but we achieved record bookings in the December quarter with a book-to-bill ratio of 1.04 and the bookings so far in January are again at a record pace.

Taking all these factors into account and after checking expectations from our direct as well as distributors channels, we expect net sales in the March quarter to be about $242 million, which is a sequential growth of about 3%.

It is much better than seasonal pattern for March, yet it accommodates the expected shutdown during lunar new year in most of Asia.

Gross margins are expected to be about 59.75 to 60% this quarter.

I remind investors that we increased our long-term gross margin guidance to 62% during last quarter.

Operating profit in March quarter is expected to be about 36 to 36.5% and earnings per share are expected to be about $0.34.

We expect to build approximately $110 million of net cash flow before payment of $40 million of dividends just announced today, and we look forward to sharing this cash with investors with another healthy increase in dividends in the next quarter.

I also wanted to briefly talk about the lead times.

We reported during December that lead times on some of our products were starting to stretch.

We said that on some of our products, the lead time had gone from four weeks to about six to eight weeks.

Now, during the quarter, we have been able to balance the mix and bring on additional assembly and test capacity online.

The lead time on products has not become shorter because of strong demand, but the lead time is not getting any longer either, therefore, we have achieved balance and we will continue to watch the situation very carefully.

So let me summarize a few key points.

Our net sales are expected to be about $242 million, up 3% sequentially; gross margin to be about 59.75 to 60%; operating expenses to be about 23.5%, to 24%; operating profit to be about 36 to 36.5%; and earnings per share are expected to be about $0.34.

With that, Sarah, would you please poll for questions?

Thank you. [OPERATOR INSTRUCTIONS] We'll take our first question of the day from Michael Masdea with Credit Suisse.

Thanks a lot.

Your commentary on lead times, Steve, is pretty interesting.

Is there a risk that you run that some of the other pieces of the bill materials lead time, which seem to be stretching, like we heard from Fairchild and some others.

Does that put any pressure on your business or any risk of over-ordering that you see out there?

I didn't hear Fairchild.

What do they say?

They said that lead times continue to stretch and that it's getting tighter, basically, and the back-end utilization is kind of getting tighter, et cetera.

It's consistent with what [AHN] has said, and some others.

Just that we have this environment of expedited orders and stretching lead times, et cetera.

Lean inventories in the supply chain may be too lean.

Does that kind of pressure on the overall potential bill of materials put any pressure on your business, or risk over order, even though your lead times are pretty stable?

Well, in some cases we feel what they're feeling, in other cases we are doing a little bit better.

We are expediting a lot of product, channel inventory is very low, and customers are expediting a lot of product and expedites are at an all-time high.

So that isn't really all true.

But for the last several quarters, we have been saying that we feel comfortable with our inventory, our total inventory is about 111 days, plus distribution is getting about two months of inventory.

And most of Microchip's inventory is at the die inventory level, so we have a healthy die stock, and we can very quickly put them in the required package and test them according to customer specification.

We also have a very, very large assembly and test plant in Thailand.

I don't know which company's purely subcontract and which companies have their own.

So having its own, having our own, we have tremendous ability to be flexible and work overtime and work Sundays and so we are really doing all that and been able to add some additional capacity as well as acquire some additional capacity at the subcontractors.

So I think we've done an overall pretty good job of managing the lead times so it doesn't become so long that it is not helpful to our customers.

With you doing that, is it a little surprising to you that the Memory business didn't do better in the quarter?

I missed it if you say what really drove that.

Many times it's a matter of choice.

I think some of the Memory and analog packages are very similar.

Some of the Memory packages are similar to microcontrollers in the low end, and largely we want to ship the highest margin product for a given package available.

I also said memory pricing was actually up a percentage point.

So, our Memory strategy is not one of market share.

It's really one of managing T&L profitably.

In the microcontroller and analog, in both of those areas, we're gaining market share.

So I think we put our money in resources, where, really, we should have put.

Thanks a lot.

Makes sense.

we'll take our next question from Adam Parker, Sanford Bernstein.

Hi.

Two things.

Gordon, can you at all break down the impact of the options by COGS, R&D, and SG&A, for fiscal '07?

Manufacturing COGS would be about 20%.

The balance would be in operating expenses, Adam.

So, 50/50 between R&D and SG&A?

Probably in that range, yes.

About 30 to 40% in R&D and the balance and SG&A.

Okay.

All right.

The other thing was just about analog.

Steve, can you talk at all about some of the successes there -- when you're winning now, why do you think you're winning?

What's your competition not offering?

Basically, why do you feel you're gaining momentum that seems sustainable?

I think it comes on specific products and winning strategies.

We have really chosen to really not expand on that.

Everybody listens to these calls and we really don't want to signal to any competitor where we are winning and why we are winning in either of our product segments.

I think that's the best position we can take.

Okay.

So same as your 8-bit strategy.

And our 16-bit strategy.

There's really -- I think -- over time, I feel investors and analysts gain not too much from it.

It may make you feel better, but, you know, competition gains the most.

So there's no reason really why we should signal to any of our analog competitors in which areas we're gaining, for them to go analyze those products.

If somebody has good market research, they know it already.

If they don't, then we don't want to tell them.

Okay.

Sorry.

One last thing.

Can you provide a base at all for all the sequential growth numbers you gave in 16-bit?

I mean, 19% of, you know --

We can't.

Providing that on analog and overall microcontrollers and memory.

Those numbers, we have them available.

We filed them in our Qs, but we're not going to break out 8-bit and 16-bit microcontroller.

None of our competitors do and we're not going to do it, either.

Great.

Thanks, guys.

Good job.

Thanks.

Moving on to Chris Danely, JP Morgan.

Thanks, guys.

Before I get to my question, I'm sorry, can you just run through the geographic breakout again?

Certainly.

Yes.

Geographically, Asia was up 7.6%.

Americas was up 1.7.

And Europe was down 1.8.

Is he asking in percentage terms what they are, a percentage of revenue?

I can figure it out.

I have the revenue from last quarter.

Okay.

And then, just a follow-on on Parker's question.

Since you won't give us the percent of revenue from 16-bit, can you give us a timeline on when it becomes material, say, in the 5 to 10% range?

No, I really can't do that either, because that would be answering the same question in a different way.

All righty.

Strike two.

How about, can you give us your expectations for the different product types for growth in the March quarter?

Like, how much you expect from micros versus analog versus E-square.

I provided some of that.

I said on 16-bit microcontroller we expect it to grow 15% to 20% -- 15% to 25% sequentially in March quarter.

And on analog, I said we expect it to grow from about 4% to 6% sequentially in the analog.

And we didn't break down the memory and the rest of the microcontrollers.

How much do you expect the overall microcontroller business to grow in Q1, or excuse me, in the March quarter?

Strike three.

Okay.

Thanks, guys.

I'm out.

From Thomas Weisel Partners, Eric Gomberg.

Congratulations on the strong quarter.

Thank you.

Could you maybe discuss a little bit more what goes into the decision on going forward with the repatriation and what we'd expect with cash, if it brought into the U.S.?

Well, if you look at it, a very large portion of our expenses are in U.S.

Our fabs are in U.S., most of our R&D is in the U.S.

In Europe, we largely have sales infrastructure, and in Asia, we do have manufacturing.

But a lot of the huge amount of expenses and all that development really happens here in the U.S.

So, we have a very large amount of cash parked overseas.

And many, many companies are already doing what we're doing.

If you look at Fairchild and Xylinx and a lot of other companies, and if you bring the cash to the U.S., it brings you more flexibility in future over time to really do whatever you wanted to do with it.

This is a one-time opportunity which is not available after the fiscal year ends out, where you can bring it in with one fifth off or one sixth off the tax payment compared to the regular tax rate.

So this is really kind of a no-brainer in a way.

Okay.

Just wondering, maybe you'd give more color on what drove such strong bookings in what typically is a seasonally weaker period.

Kind of revisiting the question of things being tight.

Are people simply concerned they can't get enough product right now or is there more to it than that?

Well, you know, all through the December quarter as I went to several conferences, what we basically said was, if you go back 90 days ago, the world was looking at the environment extremely negatively with the effects of hurricane and very high oil prices and so on and so forth, and with all the interest rate increases and people were just even calling for a recession.

And all of a sudden, things improved, and in the November, December time frame, they improved quite a bit.

It looks like the end of the interest rate was coming, the oil prices came down, the gas prices came down.

The effect of Hurricane Katrina essentially, in the rest of the U.S. the business activity was strong and so it was the rest of the world.

And everybody was preparing fairly conservatively, so some of the lead times starting to go out, and when that happens you always get booking activity because people want to reserve their spot in the line.

And now, as you look out to -- in the middle of January, seems like some of the oil prices and others have gone back up, but the environment continues to be fairly positive.

And we're seeing continuing strong bookings and it's really no longer driven by lead time changes because lead time is sort of constant, but it's not constant at two to three weeks that we used to be.

It's a little bit longer.

But we're seeing a reasonably strong business environment and we are hearing the same thing from others who have been announcing their numbers.

Okay.

And just a quick housekeeping question.

Gordon, maybe you could help, in terms of interest income you had a nice jump this quarter, cash balance obviously keeps going up, interest rates have moved higher.

What ought we expect in coming quarters?

Well, we had a stronger cash generation in the December quarter.

Receivables went down quite nicely in a period where obviously revenues grew.

So that meant we were liquidating receivables quite nicely.

I think if you look at the overall effect of interest rate on that growth and then apply it to what we're guiding you to this quarter, I think you get close to the same sort of metrics.

So after the dividend, we expect cash balances to be going up about $70 million, which should translate into another healthy increase in interest income.

Okay.

Thank you.

Next we'll hear from Mark Edelstone, Morgan Stanley.

Good afternoon, guys.

Nice --

Hi, Mark.

Go ahead, Mark.

Can you hear me?

Yes, you're coming feeble, but we can hear you.

Can you hear me, Steve?

Yes.

Go ahead.

I'm sorry.

Question on turns.

What were they in the quarter?

And then, when you look at the business, given the strength that you're seeing in analog, I'm just curious, as you look back over the past year or so if its turns profiled that business different than what you see in microcontrollers [inaudible] I would guess that the turns are going to be lower there.

But, so what were there overall turns of the Company in the quarter, then, just a profile of the analog business.

Mark, you made a call, three quarters ago we stopped talking about turn percentages and turns.

It was always confusing.

We have always maintained that turns really do not correlate to our business, whether you get the business early on, as scheduled business, prior to the start of the quarter, whether the customer gives you the order on the 20th of December or the 5th of January, it does not really make any difference if the order is being shipped in February.

So we don't really talk about turns any more.

Second part of your question was --

How does analog look in that environment?

You know, this would -- analog turns are really no different than the rest of the business.

I think when you look at the various product lines, they look very, very similar internally.

So, there is really nothing I can break it out there.

Is there just any additional insight that you can share here on the recent trend you've seen in analog in terms of your geographies or end markets that are driving that growth?

Well, I know the geography, I know the end market.

But again, if I start to talk about that I really have to get to the application level and I don't want to do that.

Because it will really signal which products -- if you look at -- in analog, we are in five different market segments.

We are in power management, we are in terminal management, we're in mixed signal.

You know, we're in Op Amps, which is called linear products and we're in interface products.

So we're in five different product segments and some of the segments we are seeing very, very strong strength.

We have great leadership position.

We have best-in-class products.

We have won some product awards, and we're getting strong design activity.

We don't want to break out any more than that, really, while we are winning.

Okay.

Thanks.

Most of this upside actually came from U.S. designs.

And some of them get shipped in the U.S. and some of them get shipped in Asia, but upside is from U.S. designs.

Moving on to Cody Acree with Stifel Nicolaus.

Thanks.

Steve, just maybe following up there on the analog question, though, do you -- are you seeing more strength attached to your microcontrollers as a system or are they stand-alone analog sales?

Well, what I have described in the past is that we had a three-prong analog strategy.

Phase 1, Phase 2, and Phase 3.

Our Phase 1 was where we wanted to attach the analog products to our microcontroller products because our name recognition was with microcontrollers.

That's where we could most easily win.

And after the acquisition of TelCom five years ago, our first phase was really to maximize the attachment of analog with our microcontroller products.

Two or three years ago, we implemented Phase 2, which was we wanted to win the analog products with any microcontrollers.

So if a microcontroller happens to be one of our competitors', we still had very good analog products in the signal channel that microcontroller, and we wanted to compete with, it better be our product rather than TI's or [inaudible -- highly accented language] or somebody else.

And so that was fairly successful because that showed us that these products could stand on their own two feet and compete successfully, even in a socket where we didn't have analog.

I'm sorry, where we didn't have the microcontroller win.

And then the third phase of that, which was implemented maybe about a year ago, is to [inaudible] analog anywhere, whether it's going in a DSP socket, whether it's going in a microprocessor or SPGA or essentially anywhere and everywhere, in direct competition with all of our larger analog competitors because we now believe, with 500-plus products in analog, we have the breadth and the strength and the best-in-class products in many areas where they could stand on their own two feet and compete with anybody.

So right now, all three phases are applicable.

We essentially go anywhere analog could win.

And we are no longer restricted by taking analog around a microcontroller.

So therefore, we're winning everywhere.

Steve, thanks for the detail on that.

On your guidance, somewhat above a seasonal norm, you mentioned -- I think it was mentioned earlier that some of the others in the industry are still tight.

Are you simply saying that the industry is above seasonal norms at this point?

Do you think you're benefitting from some market share gains here that maybe are helping to add to that?

I don't really know how to break it out because I don't really know what everybody else is feeling in terms of their growth and why their growth is coming or whether their growth is being restricted by tightness or whatever.

So I do not really know.

What we are seeing is March quarter usually is a very, very weak quarter for Microchip and we are seeing not as weak quarter as we talk to our customers in Asia.

And March quarter is, by the way, weak primarily because of Asia.

Primarily because of the Chinese New Year, the Lunar New Year.

There are no other factors, really.

That's really why it is weak.

It is 46% of our business today.

And when that 46% of the business takes a couple of weeks off in the middle of February, there is just almost no way to recover from that, historically.

As we look at it now and talk to our customers, most customers are talking about very abbreviated shutdown schedule for the Chinese New Year.

Not as long as it has been many times in the past because a lot of the end customers are around the world, in U.S. and Europe and they can no longer afford to shut down their factories for two weeks because of product demand.

That's really what we are seeing and we are essentially working hard to make all the product available with reasonable lead times to take advantage of that.

In our business, you can't really win very short-term market share from somebody unless you won that design six months, nine months, a year ago.

So, if just somebody else can't supply the product in the month of February and I happen to have the product available, I'm not likely to win market share.

I'm only likely to win if I won that market share nine months ago in that design.

Great.

Thanks, Steve.

Next we'll hear from Chris Caso from Friedman, Billings, Ramsey.

Hi, guys.

Thank you.

Nice quarter.

Thank you.

Just, if you guys could talk a little bit.

When you set your longer term margin targets you were talking about some tradeoffs that you were evaluating between perhaps being a little more aggressive in the migration of Flash versus bringing the margins higher.

Is that one of the factors that is moving you a little better than seasonal now?

Or maybe you could talk about if you're taking that a more aggressive posture when we might see the effects of that on the top line?

I don't think the effect of all that could be so short-term, because it's really a gain on the proprietary product has to go to the design phase where you are willing to bid more aggressively on a design because you have healthy margin and you have is a cushion beyond that.

So really what you're seeing in the December quarter and March quarter really has nothing to do with that.

That effect is probably a year out.

If you look at the last quarter and analyze it, you will see a 75% incremental gross margin, which is a straight calculation, an incremental gross margin divided by the incremental sales, quarter over quarter.

So while our Company GAAP gross margin was 59.7%, the incremental gross margin was 75%.

So that's where it tells you that as sales grow, there is still margin leverage in growing the margin overall.

And I guess, would you -- looking forward over the next couple of quarters, would you assume that you would remain at that sort of 75 incremental gross margin level, next couple of quarters?

That one is hard to pin down, because too many moving parts and expedites and nonexpedites and packages and mask types in a large-volume business.

I haven't really monitored the graph as well of incremental gross margin over time.

And depends on when you run into a capacity and then you add incremental capital, and that has to depreciate, which lowers the incremental gross margin from that kind of numbers.

So that one, incremental gross margin is not that perfectly linear, you know, as the gross margin seems to be.

So I can't answer that it's going stay at that level.

Okay.

Just one more and I'll go away.

Regarding the depreciation guidance you guys provided, it sounds like it is going down on a year-over-year basis.

And if you could just clarify, the depreciation guidance you gave, that was for fiscal '07?

No.

The 110 was for the current fiscal year.

For the current fiscal year, okay.

Yes.

Do we have a number for next year yet?

No, I don't have a number for next year off the top of my head.

We should be lower, right?

Yes, I think --

Because some of the depreciation rolling off the fiscal year seven, depreciation should be lower than fiscal year six.

And it can also depend on growth rates and recapitalization and particularly on assembly and tests.

So there's some moving parts there that can affect it, too.

Okay, great.

Thank you.

We'll take our next question from Tore Svanberg with Piper Jaffray.

Yes, good afternoon.

Can you share with us where you are on utilization rates at your fabs at this point?

Well, the strategy at Microchip, we always run at nearly 100% utilization.

We don't really build a large factory and then start depreciating it and then really try to find products to fill it.

So a Fab 1 and -- I'm sorry, a Fab 2 is running at nearly 100% utilization.

And our Fab 4, for the equipment that is installed and turned on and being depreciated, the utilization is nearly 100%.

Having said that, there's a large amount of additional equipment available in Fab 4, which we acquired with the Fab when we bought it from Fujitsu, which can be turned on with fairly short notice and tuned up and ready to go.

So if I look at the utilization, looking at that equipment as a factor, and we have said before that we have capacity in Fab 4 to take the overall Microchip revenue to about $1.4 billion, compared to where we are based on last quarter.

So you can figure out the utilization.

Great.

Very helpful.

And then also, on inventories, your internal is 111 distributor, two months.

Would you classify that as a regular level or do you view that as low, and if so, could it go lower?

Well, the distributor inventory is basically as low as it has ever gotten.

I think the lowest may have been 1.9 or something for the month or so.

But, I mean, two months of inventory is pretty low.

Historically it has been closer to 2.7 to 3 months.

However, the change in this inventory and distribution from closer to three months down to two months does not seem temporary.

It seems like it's more permanent type, where distributors through cycles of learning and supply chain solutions have really worked out a way to deal with lower assets, trying to improve their own return on assets.

So as distributor is keeping a lower amount of inventory from historical levels, it basically requires Microchip now to keep slightly higher inventory from the historical levels so the lead times don't go out and we're still able to serve our customers.

If you look at Microchip inventory, the lowest it ever has been during the 2000 boom was about 74 days and the highest we ever were was I think about 135 days.

So at 100, 110 days right now we feel the inventory is basically right.

We're not trying to take it higher.

We're not trying to take it lower.

If distributors were to stock up a little bit more and our inventory were to get down a little bit closer to 100, probably be a better mix, but I don't think that's going to happen, short-term.

Good.

And just finally on your analog business, you mentioned the five product categories.

Data converter wasn't one of them.

Is that because that's on their [MIC singular] interface or are you just not doing data converters?

No, we are in the data converter business.

People have different names for these titles, but yes, we make A to Ds, D to As, we make Op Amps.

Yes.

We make all those products.

Great.

Congratulations on all the record metrics.

Thank you.

[OPERATOR INSTRUCTIONS] Next we'll hear from Shawn Slayton at SG Cowen.

Hi, guys.

Nice results.

Maybe you'd rather, than talk about the business segment, Steve, can we talk a little bit about the end markets?

Can you share with us what end markets are showing, perhaps relative strength, to fuel the momentum here in January?

Well, you know, as we have always answered this question, with 46,000 customers world wide, and so many of them buying through distribution, top ten customers making up less than 10% of our business, we don't really have any end market commentary.

We can't distinguish -- we don't have huge customers that are 10, 20, 30% of our business.

We can't distinguish strength of one market versus the other.

An average customer at Microchip buys 20, $30,000 a quarter.

Over that many customers.

So we do not have any commentary.

We just will reiterate the same numbers we give.

The consumers is about 34% of our business.

Automotive is approximately 18% of our business.

Office automation and industrial at about 16 to 17% each, and communication is about 13.

These numbers over several years haven't moved by more than a percentage or so.

Okay.

Let me move on to the Flash MCU then.

Can you explain this?

What's the dynamic related to migrating existing customers to Flash for maybe OTP-type products?

And is that a strategy that's benefitting your business?

Thanks.

This strategy is absolutely benefitting your business.

Approximately 62% of our business by our estimates in 8-bit microcontrollers is still nonprogrammable, which is a ROM-based microcontroller.

ROM-based microcontrollers have to be programmed in the factory where the customer gives the factory a code, then you burn the mask with that code and then you apply the mask and the fab and you produce the product for the customer.

Completely custom.

If the code changes you can't use that product.

If a customer's end product isn't successful and the customer ordered 100,000 parts and he's not able to sell his end product, only able to sell 30 or 40, then all those raw microcontrollers are wasted.

So there's been a long-term trend which has driven the ROM memory to EEPROMs and Flash, [ASIC] products to FPGAs, the ROM-based microcontrollers to field programmable microcontrollers.

So that's one long trend that has been going on for 10, 15, 20 years.

On the top of that, Microchip was originally the leader in field programmable microcontrollers by using the EEPROM technology, which was one time programmable.

And in 1996 or something we made the first Flash microcontroller, maybe even slightly prior to that, in the industry.

And now 55% of our business is Flash microcontrollers, and much of that has come from converting the ROM-based microcontrollers of the competitors to our Flash and some of that has happened by converting our own one-time programmable EEPROM-based microcontrollers to Flash.

The benefit of Flash is it can be programmed multiple times, so if you make an error, you can recover from it.

It can also be programmed in the [bold.] It can be reprogrammed on a modem.

It can be reprogrammed remotely, so you have much more benefits like any Flash memory would have.

It's electrically visible and programmable.

How does that -- how does a microchip salesman incentivize a customer to migrate to Flash?

That I wouldn't want to share with you, for competitive reasons.

Okay.

Thanks much.

Yes.

You're welcome.

We'll take our next question from Randy Braeman with Eaton Vance.

Steve, could you talk a little bit about the dsPIC, independently, or is that folded into the 16-bit stuff?

Well, the revenue we have today is all dsPICs.

The 16-bit microcontroller, which is a PIC24 architecture, was announced to the world I think in October, November time frame.

And so it's really largely in the designing stage and there may be some small development tool revenue, but it's product revenue is zero.

So all the revenue we have today is dsPIC.

However, going forward as the PIC24 starts to shift, PIC24 actually is a larger market than the dsPIC would be, much more cost effective, and it's a pure microcontroller.

So eventually it will get intermingled and we will break those two out.

Okay.

Thank you.

Also, just a comment on the 6 o'clock time.

That's fine.

Okay.

We appreciate it.

Thank you.

We appreciate it.

All right.

Thank you.

We'll take our next question from Sumit Dhanda with Banc of America Securities.

Hey guys, nicely done.

Couple of quick questions.

First, Steve, it seems like last quarter and then again, based on your outlook for this quarter, the 16-bit business really seems to be picking up some momentum.

Can we -- I know you're not giving out the base from which the revenue is growing, but can we think about roughly similar sequential growth numbers on a more sustained basis over the next few quarters for this business?

Well, you know, the customer base is in hundreds rather than tens of thousands.

So it's harder to really give much longer term guidance where you can model that going forward.

If you look at it in September quarter, actually, the growth was zero.

And the December quarter, it was 19% and we're guiding 15 to 25 going forward.

So it's really -- it can be fairly volatile in the early stages because of the small revenue base.

So it's really much more difficult to tell you -- tor really model it for an extended period of time.

Okay.

Question on your Flash microcontroller business.

I mean, is there anything special you're doing?

I understand the rationale for migration in your customer base, but what makes it difficult for your customers to replicate the strategy that you have in terms of incentivizing your customers to migrate to Flash business?

Customers or competitors?

I mean, how come the competitors can't replicate your strategy?

Is there something special that you're doing that you could share with us, or not?

No because then they will know it, too.

I think you should probably ask them that question.

One final question for you, Gordon.

The deferred income line was down fairly substantially.

Is that reflective of a reduction in inventory of distributors, or is there something else, or am I reading too much into it?

On the deferred income on the balance sheet?

Yes, deferred income.

I think it was down modestly.

Still about the same number a month of inventory overall.

So it's down, you know, about $4 million overall.

So really, just some aspects there.

There tends to be in Europe, as you saw in the results, their business is more driven from their distribution complement.

And so that's reflective in inventory held in all geographies, but that that's certainly one that had an impact there.

Okay.

Thank you very much.

From Merrill Lynch, Joseph Osha.

Hi, folks.

Congratulations.

Steve, a longer term question.

As I have looked at your market share in the 8-bit business, it kind of has picked up pretty regularly a percentage point or so each year.

As I look at 16-bit, should I sort of think about that in the same way, kind of steam ahead and take a point or so each year, or is there some maybe nearer-term discontinuity because it's a newer business for you?

Well, if we can do what we did on 8-bit, that would be our dream.

That is really what we want to accomplish.

From -- starting from where we are right now, whether it goes up that rate or goes slower or goes faster, it's kind of very, very hard to say.

It's really pretty hard to say.

I don't think I could model this with a certain percentage growth in market share per year over an extended period of time from where we are today.

And we're really on the first or second hole here.

Okay, that's fine.

Second question: I don't know if this is for you or for Gordon.

As we look at the dividend -- and by the way, I hope your competitors are listening to the call on this -- how should we think about the payout ratio or how you set that number over time as you continue to generate operating cash flow?

Well, we have internally talked about payout ratios but we're not comfortable with the concept of payout ratio yet because we feel we have a way to go in terms of including the dividend.

Once we start to [inaudible] reach a number equal to that payout ratio internally, then we can talk about staying at that number and increasing it maybe less frequently, while we've been doing it every quarter and really trying to peg it to a payout ratio.

Right now, Board is very comfortable to continue to aggressively share the cash with the shareholders because we still have, you know, current quarter will go well over a billion dollars in cash.

Indeed, I think it's great.

And then the last question, I guess for you, Gordon, can you just update us on what your philosophy is in terms of where operating expenses should come in relative to revenues, your target over the next couple years?

If I can ask you to include the impact of options compensation in that number.

Well, of course that's going to obviously change the targets in terms of the longer term goals.

You know, we've said that getting into the range of 24 to 25 points, excluding the effects of option expensing, is where we see our longer term goals.

As we continue to invest in technical resources and our sales and marketing organizations, as we continue to have field application engineers that are the appropriate advocates for our product in microcontrollers and analog, as well as having the ability to continue to have design activities and process technologies to support the products as they enter the market.

Those are sort of the levels that we would expect.

And as I said, of the increase of 7% to 8% in terms of the overall effect of 123R, approximately 80% of that we reside in the operating expenses, with 20% being in the COGs area.

All right.

Thank you very much.

You're welcome.

From William Blair and Company, Jeff Rosenberg.

Hello, Jeff.

Hi.

How are you?

I wanted to follow up on that last question about operating expenses.

I think going back a couple quarters you had expected to be spending more on operating expenses, whether it was R&D or more marketing resources to drive demand.

It doesn't seem like that's come through.

Has there been some offsetting cost reductions or reasons you haven't added the resources you originally expected to?

Well, you know, several moving parts there.

The industry is pretty tight, the employment picture is pretty full.

So, we have a lot of jobs open, we're trying to hire people and it's taking longer to get people on board today than it would have taken six months ago.

We're hiring people around the world.

We have a large number of jobs open.

And we're getting increased employment.

We have people, but really, it's taking longer to hire people than we thought six months ago.

Okay.

And then another follow-up on a question asked earlier about your strategy to perhaps be more aggressive in terms of driving cost reductions to generate demand on the gross margin line.

You said it's not having a short-term impact and I know there's a lot of moving pieces, longer term.

But just isolating that, have you been doing that more, and do you feel like that's been driving design wins, and should we think about a slower trajectory in gross margin improvement, other things being equal over the next year or so, than we've seen in recent quarters?

Well, we basically have given you the gross margin trajectory.

We said a quarter ago that it takes about two to three years to get to 62%, and when we guided it, we had about 250 basis points to go and there were ten quarters, so trajectory, on a pure math, is about 25 basis points per quarter.

And then the numbers always deviate because there are too many moving parts.

So we have already modeled that trajectory at which we can see gross margin can improve after accounting for what we're doing on pricing and all that, and yes, we are winning more designs.

Yes, it is helping.

In answering your prior question, I wanted to add one more point on the expense side of the equation, that if you just go back about four months ago, I earlier mentioned in the call that the whole industry was very negative on the prospects of the industry, driven by all these negative things that were happening, the oil, the gold, Hurricane Katrina, so on and so forth.

So you know, our revenue estimates internally were much more conservative.

We were prepared if a recession were to come in or whatever.

So we didn't want to have expenses get ahead to the current growth levels we were achieving.

So we have actually done better in revenue.

We have just guided you with 3% sequential growth, which is significantly higher.

You look at your own estimate, it's not you, just one person, but but consensus, I think it's -- our revenue guidance is about 6, 7, $8 million higher than what the March quarter consensus guidance is out there.

So as it a result, when you apply -- you divide by a larger denominator of revenue, the operating expenses as a percentage of revenue look lower than they would have if the revenue wasn't as high.

Okay.

That helps.

Thanks.

We'll take our next question from Simona Jankowski with Goldman Sachs.

Yes, hi.

Thank you very much.

Steve, if I understand it correctly, it sounds like your inventory distributors was down a little bit.

I just wanted to find out what was behind that.

Was there resales being higher or your shipments being, or a combination of the two?

Overall inventories at the end of the day, Simona, were largely in line at about two months of inventory.

There's a modest change in the balance sheet, but that's not really material in the overall picture.

Cumulatively, world-wide basis, they were about at the same range at about two months.

So your shipments were pretty well tracking their resales, it sounds like.

Yes.

I think, Simona, I seem to recall for a few quarters you have tried to correlate distribution inventories to projected strength or softness in the business.

And I think it will help if you give up that correlation.

Because distribution is trying to improve the assets, the return that inventory, and that phenomenon of lower distribution inventory has been happening for several years now.

I wouldn't be surprised if distribution inventory over the next couple years finds its way from two months to 1.8 months, or 1.9 or 1.7.

And we are keeping a higher level of inventory and distribution often is expediting microchip because they didn't have the right product on the shelf when they needed it and sometimes they would charge an expedite charge, sometimes not.

That kind of goes, it's a negotiation.

But really, trying to correlate the amount of distribution inventory on really what's happening in terms of resale.

I believe that correlation is not good.

It would help if you really not think that way.

At least for our business.

Actually, in this case I was trying to get more kind at kind of the end demand picture.

But I do find your commentary interesting kind of on the structural changes for distribution, and that's very helpful.

Then just quickly, on the modeling side, Gordon, are you able to provide us with the after-tax ESO expense number for the quarter?

After the option expensing?

Yes, for the option expensing.

We don't have any option expensing, currently.

We'll only implement that in April.

Oh, I know.

Just the number that you would normally disclose in the 10Q, based on Black Shoals assumptions. et cetera.

Is that available yet or should we just look for the Q for that?

You'll have to wait for the Q for that.

Okay.

Got it.

Thank you very much.

We'll take our next question from Tom Thornhill of UBS.

Question on your forward gross margin goals.

Steve, do you expect the achievement of these goals to largely be efficiency improvements in pricing or mix improvements in each of the individual sectors, one-time programmable Flash, analog, or is there some margin differences between the sectors, i.e. analog higher, perhaps, that the mix shift in the business would help facilitate achieving these margin goals?

I think, you know, there are a lot of moving parts and we can't take -- like, we have about 225, 230 basis points improvement left in margin to our longer term guidance of 62%.

It's very hard to break it out by several elements.

But I could tell you several elements.

One big portion of that is the increasing utilization in Fab 4, where we have large amount of incremental capacity, and as with the revenue growth, that capacity goes to work.

The incremental costs are much lower.

So incremental gross margin is much higher.

That's a big piece.

The second piece is the equipment depreciation rolling off.

A lot of the equipment we added five to seven years ago in our factories is rolling off equipment depreciation so there's no more depreciation on it going forward.

And we'll continue to use that equipment because, you know, Microchip is not a [inaudible] company.

We continue to use all the processes, all of the equipment, for an extended period of time.

We're still using equipment that went into production ten years ago.

So there's an element that improves gross margin because of that.

The third element is what you pointed towards where, you know, product mixes, analog has high margin, flash is higher margin, shrinks, and lots of these moving parts.

And last quarter, if you look at it, Memory business was down and all the other businesses wer up, so it's a continuously [inaudible] mix, by various elements, more analog, more analog microcontrollers, more stand-alone analog, more mixed signal, more flash microcontrollers, more this, more that.

So there are all these elements and it's very hard to take a 230 basis points kind of number and break it out much more than that into pieces.

Thank you very much.

That helps.

You're welcome.

Next we'll hear from Chris Danely, JP Morgan.

Thanks, guys.

Just two quick follow-ups.

Gordon, I think you mentioned that you thought that fiscal '07 depreciation would be down.

Does that mean you expect fiscal '07 Cap Ex to be down?

No.

The depreciation that I quoted, the 110, was for the current year.

I don't have an '07 number yet to share.

Do you guys know which direction it would be versus '07, up or down?

It would be flat to down.

Okay, great.

And then a second one on the lead times, it sounds like you've done a good job of ramping capacity a little bit and getting those under control.

Is your plan to maintain those extended lead times or would you look to bring them down a little bit?

Well, you know, lead time is basically a function of demand versus capacity.

I would -- we work hard every day to bring those lead times back down because that will be better service to our customers.

However, demand continues to be strong, so we really -- we can't.

As long as they are stable, I think it's okay.

But over time, we will bring those down, yes.

Exactly.

So maybe you can get a handle on that in the June quarter or something like that?

Well, the June quarter -- after March, June and September quarters are the two traditionally stronger quarters for Microchip.

So we're skating into actually two stronger quarters after this quarter.

So I don't really know if the lead times come down during our stronger quarters .

Got it.

Okay.

That is all the time we have to take questions.

Mr. Parnell, I will turn the conference back over to you for any additional or closing comments.

Thanks, Sarah.

Appreciate everyone's bearing with us this evening.

I know it's a very long earnings season for everyone.

So, we will be available here for any additional questions and we look forward to seeing you at conferences or in your offices during the period.

Thanks very much.

Again, that concludes today's Microchip Technology conference.

We thank you all for joining us.