微晶片科技 (MCHP) 2002 Q4 法說會逐字稿

Please stand by. Good day everyone, and welcome to this Microchip Technology conference call. As a reminder, today's call is being recorded. At this time, I would like to turn the call over to Microchip's Chief Financial Officer, Mr. Gordon Parnell. Please go ahead sir.

Thank you very much, and good afternoon everyone. During the course of this conference call, we will be making projections and other forward-looking statements regarding the 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 release of today, as well to our 10Q for the quarter ended December 30th, 2001, and the 10K for the fiscal year ended March 31st, 2001, filed with the SEC, which identify important risk factors, which may effect Microchip's business. In attendance with me today is Steve Sanghi Microchip's President and CEO. I will start with a review of the March quarter and fiscal year financial results, and then Steve will comment on the quarter and outline our guidance for the June and September quarters. We will both then be available to respond to specific investor and analyst questions. Net sales for the March quarter at 148.8 million were up 4.9% from the December quarter. All geographies grew sequentially in the March quarter. Europe showed the strongest growth at approximately 7.8%, while Asia grew by approximately 4.1%, and Americas grew 3.2%. The total revenue breakdown for the quarter ended in sales for Asia representing 36%, Americas 33%, and Europe at 31%, essentially unchanged from the prior quarter. Net sales for microcontroller products represented approximately 80% of revenue in the quarter, while analog products represented 7% of revenue, and our memory products represented approximately 13% of revenue. Sequentially, sales of proprietary microcontroller products were up approximately 6%, and analog products were up approximately 10%. The growth in microcontroller products can be attributed to strong design win performance and continuing marketshare gains, and the growth in analog product sales represent an improvement in product penetration and diversification, as opposed to our performance in the wireless segment. Memory product sales declined 3% sequentially, which included a price erosion of approximately 6%. Net sales for the fiscal year were 571.3 million, a decrease of approximately 20% from the prior year net sales of 715.7 million. Geographically, Asia represented 35% of sales, Americas represented 34% of sales, and Europe 31% of sales. Also for the full year, microcontrollers represented 79% of sales, analog products were 7%, and memories were 14% of sales. Earnings per share, diluted earnings per share for the quarter were ¢19 on a 140.2 million average shares outstanding, an increase of approximately 11.4 from earnings per share in the immediately preceding quarter. Diluted earnings per share for our fiscal 2002 was ¢68, a decrease of approximately 40% from earnings per share of $1.14 in the prior fiscal year. Gross margin for the March quarter was 50.6%, an increase of approximately 50 basis points from the December quarter. Gross margins in the March quarter were primarily impacted by the improving product mix of microcontrollers and analog products, manufacturing yields, fixed cost absorption, wafer fab loading levels, as well as our competitive conditions we are experiencing. Operating expenses were 28% of sales in the March quarter compared to 29.1% in the previous quarter. Research and development costs were 20.5 million representing 13.8% of sales. Sales and general administrative expenses were 21.2 million representing 14.2% of sales. The operating expense overall showed a modest increase of 0.4 million from the December quarter, but leverage from the improvement related to the growth in revenue resulted in a 1.1% improvement. Microchip's total inventory position at March 31st was $88.6 million, which was down approximately 3.7 million from the prior quarter levels. Inventory turns are 3.3 compared to turns of 3.1 in the prior quarter. Days of inventory now represent 110 days going from a high of 127 days earlier in the fiscal year, and we anticipate that days of inventory levels will continue to decline over the next several quarters. At March 31st, Microchip's receivable days were 47 days compared to 48 days at the end of the December quarter and to 50 days a year ago. Our total receivables were 80.7 million, which showed an increase of 60.2 million from the prior quarter or approximately 25%. The increase in receivables reflects the overall increase in sales for the quarter and the non-linearity of shipments in the March quarter compared to the prior period. Overall, receivables beyond 90 days continue to be at extremely low levels. Also at March 31st, our cash position was $280.6 million and there were no borrowings against any of our lines of credit. This represented an increase in our cash position of approximately $35 million in the quarter. We anticipate growing the cash balance by approximately $90 million over the next 2 fiscal quarters. Capital spending was approximately 8.5 million for the March quarter. Depreciation expense for the same period was 25.9 million versus 26.2 in the same quarter last fiscal year and 27.6 million in the December quarter. Capital spending for fiscal 2002 was approximately $44 million. Capital expenditures for fiscal 2003 are expected to be between $110 and $140 million, and depreciation expenses for fiscal 2003 are anticipated to be approximately $150 million. Microchip shipped approximately 6200 new application development systems to customers during the second quarter. We now have shipped approximately 227,000 systems, representing one of the largest user bases of development tools in the semiconductor industry. With that, I'll it pass it on to Steve who will comment on our performance last quarter and outline the guidance for the next 2 quarters.

Thank you Gordon, and good afternoon everyone. I would like to take a few minutes and expand on a few elements of our business. First the distribution inventory. There has been much speculation in the media about the phenomena called inventory restocking. Our channel inventory, which was at an all-time low at the end of December quarter, actually fell modestly at the end of the March quarter which means at the end of March quarter it was at another all-time low. Therefore, our sales growth of 5% last quarter was driven by strong end product demand. Next is visibility. Now visibility gets talked about a lot. The measure of good visibility versus bad visibility tends to be the amount of backlog at the start of the quarter. So when we had 62% turns to take last quarter, up from 51% the quarter before, we got plenty of questions from investors and analysts. I have personally not subscribed to this definition of visibility. I have always maintained that visibility comes from design win tracking, inventory tracking, customer run rates and shipments, and a host of other indicators that we look at. This is demonstrated by the fact that we beat the March quarter despite the poorest visibility in the last 6 quarters. Having said that, for those interested in the numbers, April 1 backlog required 57% turns order filled to make the June quarter, down from 62% achieved in the March quarter, hence representing significantly better visibility if you subscribe to that definition. Bookings were quite strong in the March quarter up 35% over December quarter.

Book-to-bill ratio was in excess of 1.1. The strength in the bookings has continued through April, and the remaining turns orders required have been reduced to approximately 34% as of today's discussion. The 57% turns required at the start of the quarter has been reduced to 34% as of today's discussion. This really means that with about 29% of the quarter passed, we have taken about 40% of the required turns, which means we have a pretty good head start here. The strength in bookings and backlog is also broad based. We expect all major product segments to show sequential growth. We are also seeing a more general pattern of price firming in the Serial EEPROM arena. We expect prices to be flat to up a few percentage points in the current quarter. Next I will comment on capacity. Based on the trends we are seeing in the business, I have instructed Microchip employees to shift to an up-cycle dynamics from recessionary dynamics. This involves the following. Number one, we are increasing capacity in both our fab 1 and fab 2 fabrication facilities. We expect capacity utilization by end of this quarter to be about 80% and by the end of September to be about 88% of our current installed capacity. I want to remind the investors that we have clean room space beyond that in which we can add additional equipment to expand capacity. So when we get to about 88%, it does not mean we only have 12% left. We have 12% left of the installed capacity today. Number two, we are ramping our backend assembly and test facility in Thailand, and we are pulling in production equipment that had previously been pushed out during the recession. Number three, we are analyzing our longer term demand and capacity, and we are evaluating the need for startup of our fab 3 facility, as well as expansion of our assembly and test plant in Thailand. Number four, the impact of this increased utilization will be a 100 basis points gross margin increase in both June and September quarters. Number five, we are increasing Serial EEPROM prices selectively, wherever the prices will stick. And finally, number six, we are maintaining controls on discretionary spending so that the gross margin percentage increase, along with the expense leverage, can flow to the bottom line. Guidance. Let me provide you some forward-looking guidance. We are increasing our guidance for the June quarter from 5% growth that we had guided about 3 weeks ago to approximately 6% growth or approximately $157 million to $158 million for the June quarter. We are also providing a guidance for the September quarter for a 6% growth. The earnings per share post stock split is expected to be ¢14 and ¢16 respectively for the June and September quarters. Please take this earnings per share guidance and understand it carefully. Minutes after our earnings release there were already rumors on the web that Microchip guided it down. Remember our ¢19 earnings per share announced in the March quarter is pre-split, and our guidance of ¢14 and ¢16 for June and September quarter is post-split.

Please make a note of that. With that, Natalie would you please poll for questions.

Yes sir. The question and answer session will be conducted electronically today. If you would like to ask a question, please press the '*' key followed by the digit '1' on your touchtone telephone. Due to time constraints, we ask that you please initially limit yourself to one question and re-queue later in the session if you have further questions. We will take as many questions as time permits, and we will respond in the order that you signal us. Once again, please press '*1' if you would like to ask a question. We'll pause for just a moment to assemble our roster. Our first question comes from Chris Danely with Merrill Lynch.

Hey guys. Congratulations on a great quarter and a strong outlook. A couple of questions, are you seeing lead times start to move out for any of the MCUs or EE products?

Chris, the Microchip lead times are not moving out. Some selected competitive lead times on selected devices and selected geographies are moving out. It's lot of moving parts.

Okay, you don't anticipate going forward any issues with that?

What is that?

You don't anticipate going forward any issues with that mostly on the EE side, not really on the MCUs? That's my only concern.

Well, I don't really see an issue from a Microchip prospective. I think it only gives us opportunity when the lead times are longer or there are scattered problems from the competition and if our inventory is in very good mix and our lead time is assured, then we are in a much better position to capitalize on it, and I think we kind of already are.

Sure.

I think right now we are seeing strength across the board. We had a very strong growth in analog and microcontroller last quarter. The Serial EE actually declined at 3%, but there was only 6% ASP decline, so units were up. We are seeing now the ASP flattening to really moderately moving up combined with the unit growth, so we expect a lot better performance on the Serial EE, and we expect the micro and analog performance to continue. So my feeling is that we are looking at relative strength really on all three product lines. Obviously, the proprietary product lines are much more under control, which are design win driven. You can see from bottoms up what is in the funnel, what has been designed in, who is going to production. On a commodity products, you've got to win the business from the street, and as a result, it depends a little bit more on the market dynamics and prices and lead times and everything else. So I can say it more confidently on the micro and analog, and based on the opportunity I see in the marketplace with serial lead times moving out, I think Serial EE may be in pretty good shape also.

Okay, great. And then do you have some sort of goal in terms of inventory days that you're shooting for.

The inventory days ideal for microchip will be somewhere in the range of 90 to 100 days, and we are at about 110 days now. Our inventory has been declining both in dollars, as well as in days of inventory, and what we are trying to do over the next couple of quarters is to slow down the decline of the inventory in dollars but continue to have it go down in days of inventory based on stronger sale. However, with beating the numbers last quarter and then guiding even slightly higher even after ramping in manufacturing facilities, we are seeing both days of inventory, as well as the dollar of inventory, continue to decline, and I think over the next couple of quarters we will see that we will get into that 90 to 100 days range.

Great. And then last question, when do you need to start bringing up the pre-op fab?

A comment from a little bit earlier, we are basically doing the analysis and trying to understand the longer term demand for really the calendar year four and trying to establish that this traction is really going to stay for a little longer period of time. We have not made that decision, but I think we are within 60 to 90 of really making that decision.

Okay, great, thanks a lot.

Welcome.

Hans Mosesmann with Prudential Securities has our next question.

Hi, it's actually Tracy [_______________] for Hans. Congratulations on the quarter. I was wondering if you could talk about, given the changes that have been occurring at Motorola, if you could discuss what you would estimate your marketshare to be on the microcontroller side, and what perhaps you think you might have gained on Motorola?

What changes do you know are taking place at Motorola?

Just in terms of assuming that lead times have expanded and to whatever degree that, whether SAP closures might be impacting their ability to deliver?

Okay, these are general issues. Tracy, the marketshare information is not really available on a quarterly basis. The basic source of marketshare data we present really comes out of Dataquest on a yearly basis, and their calendar year 2001 data will be available in late June-July, which is almost 6 months after, because they've got to compile the data from every manufacturer and massage it and all that, so we don't really have that. So anecdotally we can tell you that if you look at last year, Microchip's microcontroller business in the calendar year 2001 was only down about 10% in dollars, while we believe the industry was down 35% in semiconductors and 8-bit micro could be down in the similar range, maybe down 30%. So we know we gained a lot of market share, and if you do the calculation the market share will fall somewhere in the 12% range or so, maybe slightly higher depending on who did what. As far as the growth is concerned in Q1, the calendar Q1, the March quarter, we do not know any breakdowns from any other supplier at how their 8-bit microcontroller did. Going into the year, there were forecasts of only 5% to 6% growth for the entire calendar 2002. Semiconductor Industry Association was forecasting a 6% growth for the entire year and we have done that in the first quarter. So I don't know whether that means substantial market share gains or that means everybody else grew that too, I think that data is not available. However, we know from market dynamics that we're doing extremely well.

Okay, great. And then also if you can, I'm just looking out maybe a year or 2 in the future, how would you expect your product mix to look. Would it be similar in terms of percentages given that pricing in EEPROM was just followed on as it is, but how do you expect your product mix to change over the next couple of years in terms of analog versus microcontroller versus EEPROMs, and then perhaps your additional signal controllers?

I think, you want to take that Gordon?

Yeah. I think the product mix we expect analog to really be the fastest growing segment of our business over the next several quarters, several years. It's a significant market opportunity, and it's an area where we are just starting to see some traction. So from 7% and 8% that it is currently we expect to see that make some movement over time to be a larger portion of our business, moderately over the next 6 to 8 quarters. From what we see in terms of the continuing design win status in our microcontrollers, that is going to continue to be a largest portion Tracy in the 80 to low 80% range. Serial EE is tougher to be able to look at, it can have differences at particular points in time because of market dynamics, because of lead time issues from our competitors. So it will potentially fluctuate a little bit more over time dependent on the opportunities and what we see as available for our own business model.

Great. Thanks so much.

From Shaker Investments, Edward Hemmelgarn.

Yeah. Just one question. You gave the number, you talked about, you've got 227,000 software seats that you shipped over time. Do you have any idea of what might be the actual installed base now, I mean, what people are using?

There really isn't a very good indicator of that. You know, many times the older systems that we shipped 10 years ago may be obsolete and no longer being used. We have shipped higher versions of the systems which were supersets, making some of the older systems obsolete, and there's really no way to do that. Several years ago, we had done an experiment of couple of 100 customers by direct calling, and a very large percentage of them were still in production. I couldn't give you any guess right now. I think if 50% of them were in production that would still be an outstanding number. Thinking that we have over 30,000 customers today, if 100,000 customers actively designing on a development tool, it would be awesome.

I would agree. Can you talk a little bit Steve, about what any of your end-use markets you might have shown more strength than others in the first quarter or in the March quarter and what you're seeing in the June quarter so far?

No, I think with two-third of our business, 65% of our business really in distribution and the balance being direct, spread out in 30,000 customers, it's very difficult to remake a quarter-to-quarter assessment of really the end markets. A lot of the end markets are, what's a radar detector, is it automotive or is it consumer? What's a radio in a car, is it consumer or automotive? And we don't really make an attempt of doing that from quarter-to-quarter. We change these numbers once a year roughly and the numbers can be changed very little by a percentage point here and there. So, in general, from a bookings perspective, we have seen, booking strength comes from across the industry, across the geographies, and across our product lines, and if I were to break one out I would say the telecom segment, the cell phone segment, its still not as strong as anybody would like, but the rest of it, I see it in pretty good shape. And our analog strength the last quarter came from our broad base of customers, not that it's cellular phone, which was one of the highlight in the analog growth.

Okay. Thanks.

Welcome.

Once again please press '*1' if you would like to ask a question. Moving on to Doug Lee with Banc of America.

Hi. Great quarter guys. Just a couple here, maybe it's another crack at the end market question Steve. I noticed that Europe was up stronger than the other regions. Is there any particular markets that are maybe more geared for the European region, just help us understand some of the end market trend?

No. There is no real particular end market there Doug. Europe tends to have, traditionally, a very strong first calendar quarter. It's a full production period for them compared to the December period when there is an obvious slowing as we get past the Christmas period. So, it's really across the board in all of the regions in Europe and across all of the particular end markets that are the attractive markets for us in that overall region.

Okay. Understood. Moving onto just kind of the outlook in September, you guys are one of the few companies to really give pretty precise guidance for the September quarter. I'm curious to know if you guys are beginning to see any backlog build beyond the 90-day window here?

We are seeing some backlog build past the 90 days window, but it's not huge and that's really not where the guidance comes from. As I mentioned in my commentary also, you could have a larger amount of business in the backlog and take smaller number of turns or you do the other way around, and I think what you need to know is, what's moving, what the customer run rates are, a bottoms up analysis, who is committed to doing what by distributor, by sales person, by account. So, it's clearly a much, much larger analysis at Microchip, which gives me the comfort to be able to give the guidance, not just the backlog.

Okay. I think that's fair. And just to wrap up that piece of the question, when things kind of normalize and you have more of a comfortable run rate, if there is such a thing, what kind of turns do you normal require Steve, is it in the 35% range, 40% range?

Like you said, 25 is in the business, I haven't found what normal is. What happens at Microchip is our percentage of the turn required for the quarter have peaked in the 62% to 65% range. In the last cycle back in 1998 they peaked at 65. And this cycle they peaked at 62, which was the last quarter. And the current quarter, the June quarter, required 57% of the turns. That's starting to come back down showing the inflection point. On the other end, I think the lowest number of turns we had was about 20%, which was in late 2000.

Okay. Great. And then just lastly, on the understanding that you guys are still working through your fab 3 decision, the capex guidance you gave of this 110 to 130 million, does that range because of fab 3 or if you decide to go forward with it, more aggressively, could we see that number rise? Thank you.

It has some of the fab 3 capital requirement. The only decision is exactly when we start, a quarter in, a quarter out, but most of that money still is really going to be in that fiscal year. Remember there is a fair amount of equipment for the initial setup in fab 3 already. When we were trying to bring it up about a year and a half ago, there are some more facilities work needed, there is some other work needed and that capital is allocated in this number. Now obviously if we see its strength much beyond that and we decide to ramp it harder and put the subsequent step also in fab 3 then the number could rise, but the initial start up capital is in there.

I got it. Okay, great. Thanks a lot. Great quarter again.

Thank you.

From Morgan Stanley, Mark Edelstone.

Yes. Thank you. This is Gary Chia for Mark. Great quarter. There's really not a whole lot of questions. But one clarification first please, Steve. You said bookings were up how much in the quarter?

Bookings were up 35% sequentially from December to March.

Okay. Thank you. And, Gordon, would you be able to provide an application breakdown for your revenues?

Application?

Yeah, consumer, auto, applications?

I think Steve commented on that. Just with the broad base of customers that we have, it is not possible to give you a quarter-on-quarter update on that Gary.

Okay. What about distribution? What portion of revenue came from that particular channel that we do know?

That's in the 60% range approximately Gary.

Okay. Right. Is it more in the mid 60s or low 60s?

It would be in the low 60s. I believe it was 60% in December, and it's in that same range here for March.

Okay. Great. And finally, would you be able to comment a little bit on the attached rate trends for your analog product to the microcontrollers?

This again were the 30,000 microcontroller customers and our 6000 analog customers, hours after the earnings announcement, we don't really have that kind of analysis. We look at it a little bit more during the quarter after all the earnings and all that are done, and maybe I would make a comment at a future conference, but no, we don't have that today. However, I did comment that the growth in analog is coming from much broader base customers where we have gotten our analog design in, in the last year, which will point towards some of the growth coming from the attached rate rather than coming from all cellular phone business from Motorola and others. So qualitatively, yes, it's getting better. Quantitatively I couldn't give you a number right now.

Okay. Fair enough. Thank you very much.

Hi. This is actually [_______________] for Tim Mahon. Just a quick question on the fab in Washington, when you do plan to come out with that and ramp that facility up, what technology do you plan for that to be on?

There are several scenarios being looked at, and before I actually make that decision, I'd rather not share that.

Okay.

Natalie, are there any more questions?

Yes, you have a question from [_______________] from Robertson Stephens.

All my questions have been answered. Thank you.

Moving onto a follow up question from Edward Hemmelgarn.

Two questions. Gordon, the February, trying to jog my memory again, the 62% in turns was what you needed in the January conference call to make the March quarter. Right?

That's correct.

Okay. So your actual turns then was 67%-68%?

I didn't do the calculation, but it's all the new business, the business above a flat quarter had to come from turns, by definition.

In the, this is Steve, in the January conference call, in fact, the number we gave was we needed 61% turns to take, and the actual turns that we took was about 62. Now, the average question comes in that our guidance was up 1% to 2% and we did up 5%, what if the turns only go from 61 to 62, and that's why there are a lot of moving parts depending on what's shipped out from distribution. So the distribution POS sales were stronger and sometimes that is where you can easily be up 2% or 3% in turns.

Okay. Right. That makes sense. Then my last question is you talked about you got extra clean room capacity at fab 1 and fab 2 that you just haven't filled up yet. On a percentage basis of the overall capacity then, how does that compare to the existing clean room capacity that you're utilizing?

Do you have the number?

I don't have a specific number I could come up with. I think it somewhat depends on the mix of technologies that we put into that clean room space and how efficiently we bring that up on those particular technologies. But certainly, in terms of the capacity we have in place, that is the clean room capacity we have in place, that's part of the deliberation that we have when we're thinking about the timing for [_______________] and how those will both sell together. But specifically a percentage I couldn't give you at this point.

I think the reason we're a little elusive on that is this, there are several technologies, and if the demand comes in perfect mix in those technologies, we can get up to a, if I throw a number out, $220 million plus range. There is no guarantee that the demand comes in that perfect mix. If the demand comes more on the hiring technology and less on the older technology, which we were on 2 years ago, then the number could be slightly less than that because there isn't a space available for some advanced capacity and water and things needed for really that advanced capital and hence may require in fab 3 earlier. So there are a lot of moving parts, but usually we have been pretty good in being able to provide that capacity as we get there.

Good. How much lead time do you need from the time that you make a decision to start ramping fab 3 till when you would actually start producing usable chips?

Usually we can bring Fab III up in about a year.

Okay.

From the time that they go, we can make production in a year. It doesn't take a rocket scientist to figure that out. It's all in exercising triangulation towards what the ultimate capacity is, what our June and September quarter next year may look like. We get this question a lot, and neither we have the visibility one year out and what the revenue will look like one year out, nor there is a good visibility in what mix the demand may come in. So it really becomes a very difficult exercise, and that's the exercise we're going through and trying to come up with several scenarios, what if scenarios, and to ensure that we have enough capacity for our customers no matter what mix the demand comes in. And that's why we can't really tell you plainly that the decision gets made on day 1 and this is the capacity and this is where we can get to, because the visibility of the business is not that good, going out, one year out.

Understandable. Thanks.

Once again, please press '*1' if you'd like to ask a question. It appears there are no further questions. I'll turn the call over to you Mr. Parnell for any closing comments.

Thanks very much Natalie. We appreciate everyone's attendance at our conference call this afternoon. Steve and I will both be available for the balance of the day for any other followup questions. Thank you very much.

That does conclude today's conference call. Again, thank you all for your participation today.