庫力索法 (KLIC) 2009 Q4 法說會逐字稿

Greetings, ladies and gentlemen and welcome to the Kulicke & Soffa fourth fiscal quarter and fiscal 2009 year-end results call. At this time, all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation. (Operator Instructions). As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Mr. Tom Johnson, Director of Investor Relations for Kulicke & Soffa. Thank you, Mr. Johnson. You may begin.

Thank you and good morning, everyone and welcome to Kulicke & Soffa's 2009 fourth-quarter and fiscal year-end conference call. For those of you who have not seen the results announced this morning, they are available on the Investor Relations section of our website at www.kns.com. An audio recording will be made of this entire conference call, including any questions or comments that participants may contribute. This recording may also be accessed from the Kulicke & Soffa website for a limited time.

During today's call, we will make reference to non-GAAP financial measures. Reconciliations to those measures or to the most directly comparable GAAP results are also posted to the Investor Relations section of the website by the GAAP to non-GAAP reconciliations link.

The content of this conference call is owned by Kulicke & Soffa Industries and is protected by US copyright laws and international treaty. You may not make any recordings or other copies of this conference call and you may not reproduce, distribute, adapt, transmit, display or perform the content of this conference call in whole or in part without the written permission of K&S.

Today's remarks are governed by the Safe Harbor provision of the 1995 Private Securities Litigation Reform Act. Actual results may turn out significantly better or worse than indicated by any forward-looking statements we may make this morning. For a complete discussion of the risks associated with the operations of Kulicke & Soffa, please refer to our SEC filings, particularly the 10-K for the year ended September 27, 2008 and our other recent SEC filings.

It is now my pleasure to introduce the host for today's call, Scott Kulicke, CEO and Chairman of the Board. Scott?

Thanks, Tom. Good morning and welcome to our call, the purpose of which is to discuss K&S's financial results for the September 2009 quarter. Considering the year we have all been through, results for the September quarter were quite good. Revenue grew to $110 million, more than doubling from the June quarter. With this revenue growth, coupled with our previous expense reductions, K&S has returned to profitability.

The semiconductor industry and the electronics industry in general have rebounded from last winter's unprecedented downturn. For K&S, that took the form of increased revenue in all our established productlines with ball bonders leading the way. Improved demand came from both IDMs and subcontractors and across every market segment with the exception of memory. Our memory customers continue to defer backend CapEx. Although as memory unit volumes continue to grow, they will either have to start spending or to put their incremental units into subcontractors. K&S should benefit either way. We also see continued strong demand from the LED market.

I will talk about our expectations for the December quarter and the trends we see in the industry in a moment, but first I would like to turn the call over to our CFO, Mike Morris, for a detailed review of our financials. Mike?

Thank you, Scott and good morning, everyone. My remarks today will include non-GAAP measures as a supplement to our GAAP results in order to provide a better view of our financial performance. The items we exclude to determine our non-GAAP measures are explained in our earnings release and are also provided on our website.

As was the case last quarter, all operating results associated with our former wire business are reported as discontinued operations and are not included in the current or prior quarter's discussion. On today's call, I will compare the September quarter to the June quarter and will refer to non-GAAP numbers unless otherwise noted.

As Scott mentioned, results for the September quarter were good. Net revenue for the period was $110.5 million, up $58.4 million from last quarter. Volumes were up across all of our productlines except die bonders as we continue to phase out our legacy die bonder models, and while our new iStack die bonder completes its customer qualifications.

The revenue increase for the period was driven mainly by ball bonder volume. Although heavy wire wedge bonder and expendable tool volumes were also up significantly. Ball bonder unit sales were weighted toward subcontractors who comprised about three-quarters of our ball bonder shipments.

Gross profit was $47.2 million, up $27.5 million from last quarter. Our gross margin was nearly 43%, up roughly 5 percentage points from the June quarter as the volume pickup drove improved absorption of our fixed manufacturing costs.

Operating expenses were $35 million, up $3.4 million from the June quarter and were better than the projection we provided in September. The expense increase was driven by costs associated with our higher sales, including employee incentive compensation expense and higher commissions to our sales representatives.

As a measure of our operating leverage, 41% of our incremental revenue this quarter fell through to operating profit. As we have discussed in our previous calls, our operating expense reduction plan has generated roughly $22 million in annualized savings and is essentially completed. For this reason, we will not provide operating expense guidance going forward.

Turning to the balance sheet, we ended the quarter with total cash and investments of $144.8 million, an increase of $27.5 million from last quarter. Working capital, defined as accounts receivable, plus inventory, less accounts payable, grew $25.5 million due to the significant increase in sales. From a days perspective however, our working capital performance was good. DSO was 78 days, down 9 days from 87 days last quarter. DSI was 59 days, down 58 days from 117 days last quarter and our days payable was unchanged at 57 days. Our ROIC this quarter was nearly 21%.

Finally, you will recall that we sold 8 million shares in our August follow-on offering, raising $38.7 million in net proceeds. This stock sale leaves us with a solid liquidity position going into 2010. Scott?

Thanks, Mike. As we look into the December quarter, we expect more moderate growth with revenue to increase to $115 million to $120 million. While it is too early to call the March quarter, historically, it has been seasonally down for much of the industry. The first calendar quarter notwithstanding, many forecasters are predicting a significant increase in IC units into 2010. The LSI for instance is currently forecasting 9.6% growth in semiconductor unit volume. Historically, IC units have been the principal driver of demand for K&S's ball bonder and expendable tools productlines. Any cyclical increase in IC unit demand should also drive growth in our recently acquired heavy wire wedge bonding business where sales are driven by unit volumes of power management and high-power application ICs.

Layered on top of this is the demand from the LED packaging market. LED unit volumes continue to grow at a rapid pace. While K&S is a new entrant in the LED market, the same strength that has made us the market leader in IC assembly has allowed us to quickly carve out a reasonable market position in LED. Revenue for LED assembly is all incremental to our traditional business.

Next, we believe that the industry's transition from gold wire to copper wire has the potential to drive a significant wire bonder replacement cycle. Our view is that a sizable portion of today's installed base of wire bonders is not suitable for conversion to copper wire. These older bonders lack the process capability to reliably bond copper and conversion is not economically feasible. The high price of gold has accelerated our customers' efforts to convert to copper wire. For at least one of our subcontract customers, early adoption of K&S's copper wire bond technology has led to marketshare gains. Conversion to copper wire has the potential to create significant incremental demand for ball bonders over the next few years.

We are also excited about the potential for incremental growth through our recently introduced iStack die bonder. Our initial demos and quals have proven that K&S has reset the standard for stacked die bonding both in terms of productivity and of accuracy. We expect to start to convert quals to purchase orders and then shipments during the March quarter.

Our optimism isn't just about revenue opportunities. We are midway through a series of previously announced actions designed to deliver manufacturing cost reductions in both our equipment and expendable tools productlines. These actions include our newly opened subassembly and supply chain facility in Malaysia, which has started building selective subsystems for our Singapore final assembly factory, as well as the ongoing consolidation of (inaudible) manufacturing in China.

In 2010, we will move heavy wire wedge bonder manufacture to Asia as well with subsystems built in Malaysia and final machines shipped from Singapore. Taken together, these various [programs] should continue to drive down our manufacturing costs.

As you know, K&S's stated strategy is to simultaneously pursue the goals of being the technology leader and the cost leader in each of our major productlines. Our success in executing against these goals is evident in both the success we are having in the marketplace and in our return to profitability. We expect more of both in 2010. Claudia, we would be happy to take a few questions.

(Operator Instructions). Gary Hsueh, Oppenheimer.

(inaudible)

Hey, Gary, are you out there?

Yes, can you hear me?

Yes, now we can.

Okay. I am on a cell phone and the connection is (inaudible) here in New York, but first question I have is just about some of these new businesses that you are getting into in 2010. How does that (inaudible) in terms of (inaudible), particularly with the die bonder (inaudible)?

Hey, Gary. I am sorry. You are breaking up and I am not really hearing you.

Okay, I am going to jump off and dial again. Sorry about that.

Okay. Well, please come back. We would like to hear your question.

Krish Sankar, Bank of America-Merrill Lynch.

Yes, hey, guys. I have a couple of questions. Number one, you didn't mention -- you briefly touched upon the price of [challenger] 1Q seasonally weak. Wanted to ask do you think you're going to see normal seasonality this time are coming off a deep (inaudible), the seasonality might not be around this time?

Well, that's a tough question. We argue a lot among ourselves about normal seasonality because our data says this business is seasonal except when it is not. So I am not sure that I can say for starters what normal seasonality is.

Secondly, there is nothing about this year that has been normal in my experience. I have been doing this for a long time. Both the downturn and the rebound are without precedent.

I am not sure that we are not all using seasonality as an excuse for lack of visibility. And we are all just sort of plugging the idea that March could be weak and that is going to be our excuse.

I know, for us, an awful lot of our planning revolves around looking past the March quarter because, after all, it is only a quarter and it is a quarter where you have the tail end of Christmas plus the lunar new year.

But when we look into 2010, we see a lot more good things on the horizon than bad things. As we mentioned in our opening remarks, most of the forecasters that we are listening to are projecting significant increases in IC units. That is all good and should drive demand. Layer on top of that the LED business, which is developing to a nice solid business for us. That is incremental demand. Layer on top of that in the ball bonder space the potential for significant replacement cycle. That is incremental demand over the units. Then you have die bonders, then you have cost reduction and we feel pretty good about 2010 because of all of that.

Right, okay. And if I look at your calendar Q4 guidance, the December quarter guidance for the core semiconductor business, do you think though the sales will move higher from here or do you need memory to come back or do you think the core business can grow just based on end demand from the OSAT guys?

Yes, it can grow just from the OSAT guys. We have at least one OSAT customer who is one of our biggest customers who is very bullish. Now part of their bullishness is they are taking share from their competitors, but they have got a really aggressive plan and they are big for a reason. And we take some confidence from their confidence.

Got it. Okay and if I can squeeze one last, on the LED side, bonder side, how do we think of the overall unit -- the number of bonder units run rate for 2010, 2011 for the industry?

Very round numbers. Our view is that the LED segment of the industry is taking -- I am extrapolating from what we've got and what we think our share is. It is taking -- in total from us and from ASM Pacific who is the other big supplier, it is probably taking around 1000 bonders a year, that order of magnitude. They used to get 100% of that. We think we are up in somewhere the 30%-ish marketshare over the last couple quarters and we are still driving for further penetration into their market.

Andy Schopick, Nutmeg Securities.

Good morning. How are you, Scott?

Hey, Andy. It's good to see you. We missed you the last couple calls.

The last couple of calls, I have had some conflicts, but thanks for mentioning it.

You mean we are not first on your priority list, Andy?

Not always, Scott. Unfortunately, I have got to juggle a lot of balls.

I am hurt.

I have got to ask you a few questions here. Tax rate expectations for the current fiscal year, cash and book, what are they going to look like?

Mike?

Yes, I would go with a 10% book and a 10% cash tax rate for the next quarter and for the full fiscal year.

Okay. Scott, back to you. Capacity utilization at your customers. Can you kind of give us a little bit of an update as to what you have seen over the course of your fiscal year?

Sure. Well, first, Tom, is this on the website?

No, it is not.

Well, is there a version of it on the website?

The most recent would be the (inaudible) conference.

Okay. We will update the chart on the website. The one that you have got probably goes back to the beginning of October -- end of August? That long ago? Okay. We have seen -- okay. Right now -- I won't give you all the history of the year, but, in general, about this time last year, capacity utilization plummeted from what was already trough of cycle levels. That is to say from the middle 70%s. It plummeted down to 45% and since we have been keeping track of this data, it has never been that low.

Interestingly, it came back almost as quickly in the spring and it is now bouncing around in the middle, low 80%s, so 84% to 82%, which is traditionally peak of cycle kinds of levels. Now -- and that is absolutely consistent with the idea that customers have been buying bonders at almost peak of cycle kinds of levels as well. Current quarter is a very strong quarter for bonders and for expendable tools, so it all fits.

Okay. I want to come back to another segment of your business, the wedge bonding business. You haven't really talked too much about that on the call today. How is that performing relative to your expectations since the acquisition of that product area?

Well, certainly the last few quarters have been horrible, like the rest of the business. But they have started to come back. The September quarter was a much better quarter, the best quarter since the acquisition. So they are getting back to reasonable levels.

A lot of our focus with the wedge bonder business -- and by the way, some of their customers' base, which is automotive driven and outside kind of traditional electronics, is still depressed, so there is still upside in their revenue stream. A lot of our efforts with them are around the cost reduction. We think there is significant cost-reduction opportunities. We know there is significant cost-reduction opportunities as we move their productline to Asia next year. The pace of that move has been dated mostly by their inventory position. They have a ton of inventory. It is one of the reasons why our DSI, while it is improving, still isn't where we want it to be and we see significant improvement in this kind of operating metrics, significant improvement in gross margin for wedge bonders. And again, as I said before, still upside potential on the revenue side of their business.

Scott, would you attribute the poor performance that you have seen to certain industry verticals such as auto perhaps as really being the main depressant?

Not just auto. I mean nobody escaped the carnage earlier this year. And I don't think you can -- we were all in that together. That they are coming back a little slower than the rest of the Company I would attribute to the analysis you just gave.

Okay. Another thing here about industry estimates and projections. I believe that you primarily use the LSI data and it appears to me, from some notes that I have from the past, that clearly the ball bonder and die bonder markets have fallen well short of their more recent estimates over the course of the last few years. Do you have any general sense realistically of what the ball bonder and die bonder businesses will look like for calendar 2009 and how that will relate to your data for 2008?

Off the top of my head, absolutely not. I would be completely winging that. I will say I think the LSI gives the best granularity of any of the forecasters in terms of different equipment segments. So I still say they are the people to use for that kind of data. That everybody had a badness in 2009 is I think pretty obvious.

Can you share with us what their estimates are for the ball bonder and die bonder businesses for 2010?

Not off the top of my head. I mean we can get that for you though. I'm sure if you call them up directly, they would also give it to you. But I don't have it off the top of my head.

Okay. One last thing for Mike if I can. You have got a convertible debt payment coming due. I think it is $49 million next June. What is the likely scenario for retiring that convert?

We are good for the convert. We have been good for the convert for a while.

Meaning that you are just going to let it mature?

Yes, we are going to pay it off.

Pay it off? Okay, thank you.

Andy, let me say on that in general, the Company has, I guess, an announced policy to continue to delever and we are going to keep going down that path.

Yes, well, it is the right path, as well as was the decision to sell the gold wire business.

Thank you. Okay, do you have a last question?

No, that's it.

Okay, thanks, Andy. Good to have you back.

Thank you.

Lee Simpson, Jefferies & Co.

Hi, good morning, everyone. Just one or two quick questions. I missed the start of the call, so apologies if you did cover this earlier. I wanted to sort of get an idea for what do you think the replacement rate will be on ball bonders given the progression to copper? And could that be -- and what does that really translate as a driver for sales growth next year? Maybe if you can maybe try and characterize what is the tipping point for customers as far as costs? You will see them moving into this replacement cycle and maybe are we in the early stages of this already?

Interesting questions, Lee, good questions. Hard to answer with any degree of precision. First, the tipping point. We are long past the tipping point on gold price. On whether gold was at $900 or $1000 or $1100 an ounce, the move to copper is a gigantic relative to the cost of packaging, a gigantic cost reduction. And it is a one-way transition. Gold could go back to $800 an ounce and people wouldn't go back to gold. They will stay in copper once they have made the transition.

The cost reductions are primarily in the copper wire versus gold wire, but there is also a small cost reduction available in the wafer fab. As you go away from evaporated aluminum pads towards electroplated or electroless plated copper or nickel palladium pads, which are harder and more robust and necessary to deal with the hardness of the gold ball.

We hear customers talking about significant transitions. One of our big IDM customers has a stated goal of 80% converted to copper by the end of next year. I think that is pretty aggressive and so they miss it by a quarter or two, it still gives you some sense of the magnitude of the potential shift. And we see a couple of our subcontract customers, AFC in particular, also being very, very aggressive in the copper transition.

When you look into the installed base -- again, I am talking very round numbers -- we use a planning estimate of about 80,000 total ball bonders in the installed base and round, round numbers again. So no great precision. Use this as order of magnitude only, estimates. About half of those are not, in our opinion, convertible. So that is 40,000 bonders. They are older, slower bonders generally, so those 40,000 bonders -- again, very round numbers -- might require 20,000 new bonders to replace them.

(inaudible) 20,000 incremental bonders over the demand curve, even if it is over the next five years, that is a material improvement, a material increase in bonder unit shipments. In a good year, the industry takes 10,000 bonders or so for an average year. So it is a big bump in shipments. I am not going to forecast which quarter it starts, what quarter they are going to be in and how much comes in 2010 and 2011 because nobody knows. They are all questions still to be resolved.

But anyway you look at it, we think it is significant incremental volume over the next few years. We expect that we will be one of the major beneficiaries of it. Unfortunately, I think ASM Pacific will also benefit from it. But I think the two of us will increasingly emerge as the two guys that are serious players in the ball bonder space.

Great. I mean the split between the two beyond this replacement cycle should reflect the current marketshares do you feel or do you think there is a chance for this option in the end-market shares?

Well, our goal every day is to eat their lunch. So yes, we are certainly planning to disrupt the current marketshares, but I am sure if you were to ask them, they would say the same thing as well.

Fair enough. Fair enough. Again, as I say, I missed this on the call, I just wanted to check, did you give any sense for how the market splits between subcons and direct customers?

Yes, subcons were about three-quarters of the shipments. It was in Mike's part of the thing.

Okay, great. I dare say he gave a guidance on or a number for the LED split in ball bonders because it was due to drop this quarter as I understood.

No, actually we didn't give it, but LEDs were a little under 10% of total bonder shipments. That is down from last quarter, but that is not because the units were down, that is because -- that's not because LED units were down, but because the denominator of that equation, that is the number of bonders we shipped, especially to IC subcons, was way up.

Yes, okay. And then maybe one final question. The wedge bondings upturn, if you like, that we see in the third quarter in arrears in the rest of the business, what sort of longevity at this point would you give -- you give the order on here? Does this run for three, four quarters or is it something less short than that?

It is not clear to us right now. The wedge bonders have historically had their own rhythm, a little bit different than the ball bonders, lags the ball bonders and we are still trying to make sense of that. Certainly the automotive and hybrid side of their business continues to lag. The part that has turned up has been the power management, IC power management side where the strength is in the current quarter.

I mean if you speak to some of the mix signal chipmakers in power management, the power savings guys over in Europe, Infineon, Zylog, [Sami] and [Satchaw], it sounds like the segment is powering on. There is some real hope for next year around that end market. Is there any way we could translate end chip units to wedge bonder ordering at this point?

I wouldn't want to go out on that just yet. It is also complicated by what we think we see as a move for more of those parts into the subcontract chain as opposed to being built by IDNs, assembled in IDM factories. A lot of the wedge bonder shipments last quarter and this quarter are to one of the big subcontractors who is taking a big -- getting a big footprint in that area in those packaged formats.

Great. All right, thanks, guys.

Peter Wright, GC Research.

Great, thank you and congratulations, guys. Most of the questions were answered that I had, but one is on the wire bonder replacement cycle, who are the main customers that are participating in that?

Okay, Peter. Here is -- we talk about a perspective cycle, so participating is a little bit misleading and I don't want to mislead anybody. But if I look at who are the people that are most aggressive on the copper conversion, on the IDM side, I would point out in particular KI who is incredibly aggressive and TI gets the credit. They were one of our development partners in the whole process. We spearheaded an informal -- consortium is too strong -- but an informal multicompany working group to deal with molding problems, bond pad robustness problems, wire bonder problems and TI was the lead on that in terms of doing the bond pad work because they controlled their own fabs. So TI I clearly mention is way out in front on the IDM side. ST is also very aggressive.

On the subcontractor side, I would say that ASE is clearly in the forefront, still is working hard at it, but we have, I don't know, 60 odd companies that are working on copper conversion right now. But those four are the ones that I think are probably most noteworthy.

Fantastic. That is great color. So that answers even more specifically I guess what I was looking for. It is more the logic side than the memory side though, is that the --?

The memory side, interestingly, hasn't shown a lot of interest in copper conversion. We talked about it as recently as yesterday. They say, ah, our wires are real short and we've got bigger fish to fry. It surprises me. My guess is that someday somebody is going to say, no, we have got to convert and they will all move in a herd to convert rapidly. But right now, the memory guys continue to be laggards in the copper wire conversion.

Fantastic. And then my last question --.

Peter, if you can get a better answer from them than we did, please give me a call because we don't understand it.

I am sure that would be impossible, but my last question is on the LED side. What do you think your current share is in that market?

It is a market we are still learning, so I am going to be pretty loose on our estimate, but we think we are still only at 25% to 30%. That doesn't sound like very much, but remember we only introduced an LED, formally, an LED bonder last quarter. We have only really been participating in the market for about a year. So we are really happy with the slope of our marketshare curve and we think we have got more to go.

That's fantastic. And there isn't a lot of mom and pops; it is pretty much just you guys?

In terms of the bonder side, yes. ASM, who is absolutely the dominant player, SHINKAWA, a little bit with Japanese guys. But now us and we are targeting targeting across the board, a broad range of applications, both little tiny, mom-and-pop shops, right up to the Samsung LED type guys.

Fantastic. Thank you and congratulations.

Thanks, Peter.

Wenge Yang, Oppenheimer & Co.

Hi, (inaudible) is going to get on the line, so let me ask a couple of questions. Regarding your equipment operating margin, I did a quick calculation, it is coming in at around 3.4% and 40% (inaudible) rate. So this compared to the last three pieces looks pretty low. And could you give some explanation on why the equipment side operating margin remains still pretty low at this point?

Sure. You have to remember that we are carrying all the operating expenses of the die bonder business and no revenue to speak of. Now you have to front-end load those expenses, the engineering expenses, the development expenses, all the sales and service to support your quals and we don't expect to get first orders, first revenue in there until the March quarter. So that significantly drives down the operating margin on the equipment segment.

So do we see notable improvement in Q2 in the June quarter next year?

You're really asking me what is the rate of the revenue increase, how quickly will we penetrate that. I am going to duck that question. It won't be a step function. We are the new guy on the block. People are not going to bet the farm on us just yet. We are going to have to prove ourselves. The quals have taken a long time, although we have done very well in every qual. But typically customers will qual an application and will do well and then they will say let's try a different application, we have got to go do the application part and then we run that application and then they are going to buy a couple units and see how that does and then they will buy some -- place some real volume orders with us. So we expect it will be a while for that revenue curve to peak. And it is through that, yes, we will have an operating expense problem that will drag down total margins, but it will get better.

Okay. Back to copper conversion. Between the copper conversion kit and new equipment purchases for copper wire, what is the percentage between those two buckets in terms of total conversion and what kind of growth margin profile you are looking at for each bucket?

There is not a significant difference in the margin profile -- well, significant -- I mean the copper machines or copper-converted machines have an improved margin, but it is within the noise of normal customer to customer variations. I guess about a third of the -- no, almost half of the machines we shipped last quarter -- we shipped -- okay, it is a hard question to answer. I am not trying to duck it. It is that we ship -- we basically sell people a machine and a kit and sometimes we install the kit and sometimes they install the kit. So if I look at number of kits versus number of bonders, about half -- the number of kits is about half the total number of bonders we shipped in the quarter. That number is growing in the current quarter.

Okay. Last question, so when you deal with the subcons and maybe I didn't get it, what is the current utilization rate on the ball bonders? And do you have any visibility on their inventories at this moment and moving forward?

Okay. No, we almost never have inventory visibility on our customers. So the answer is I can't help you with that one. We don't break out subcon utilization from IDM utilization, but we do talk about total utilization and total utilization across the companies we survey is about -- and we survey, I don't know, about 40% of the installed base when we do our utilization number. So we think it is statistically significant and it is running in the middle low 80%s right now.

Okay. Great. Thanks.

Thank you. Tell Gary we missed him.

There are no further questions at this time. I will turn the floor back over to Scott Kulicke for closing comments.

Okay. We would like to thank you all. If you have any follow-on questions, please give Tom Johnson a call and he will track down either Mike or I -- Mike or me to answer it. Thank you very much.

Ladies and gentlemen, this does conclude today's teleconference. You may disconnect your lines at this time and we thank you for your participation.