Executives
Bob Okunski - Director of Investor Relations
Bertrand F. Cambou - President, Chief Executive Officer, Director
Dario Sacomani - Chief Financial Officer, Executive Vice President
Analysts
Glen Yeung - Citigroup
Ahmed Seraf - Credit Suisse
Aaron Husock - Morgan Stanley
Daniel Amir - Lazard Capital
JoAnne Feeney - FTN MidWest
Betsy Van Hees - Cowen & Company
Bobby Gujavrty - Deutsche Bank
Spansion Inc. (SPSN) Q3 2007 Earnings Call October 16, 2007 4:30 PM ET
Operator
Good day everyone and welcome to the Spansion third quarter 2007 earnings results conference call. Today’s call is being recorded. At this time, for opening remarks and introductions, I would like to turn the call over to the Director of Investor Relations, Mr. Bob Okunski. Please go ahead, sir.
Bob Okunski
Thanks, Dustin. Good afternoon, everyone and welcome to Spansion's third quarter 2007 earnings conference call. This is Bob Okunski, Director of Investor Relations here at Spansion. Joining me are Bertrand Cambou, President and CEO; and Dario Sacomani, Executive Vice President and Chief Financial Officer. As for procedure on this call, Bertrand will start out with a high level view, followed by Dario who will give some additional color on our performance and then turn it back to Bertrand for guidance. We will then open up the call for questions.
Before beginning today’s discussion, I need to spend a few minutes remind you of the Safe Harbor limitations of our discussion. During this call, we will make forward-looking statements that are made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995, including but not limited to statements regarding the expected production at SP1, expectations that Spansion will complete the Saifun acquisition and its associated future impacts, expected fourth quarter sales, gross margins, ASPs, and expected fourth quarter costs in research and development, sales, general and administrative expenses, net interest expense, taxes and capital expenditures.
Investors are cautioned that these forward-looking statements in this conference call involve risks and uncertainties that could cause actual results to differ materially from the company’s current expectations.
For risks that the company considers to be important factors that could cause actual results to differ materially from those set forth in these forward-looking statements, the company urges investors to view in detail the risks and uncertainties in the company’s Securities and Exchange Commission filings, including but not limited to Spansion Inc.’s annual report on Form 10-K for the fiscal year ended December 31, 2006, and quarterly report on Form 10-Q for the quarter ended July 1, 2007.
Also, a copy of this press release is available on our website and this live call is being recorded for replay purposes and can be access on our investor relations website at www.spansion.com.
Now, let me turn the call over to Bertrand Cambou, Spansion's President and CEO. Bertrand.
Bertrand F. Cambou
Good afternoon. Thank you, Bob and thank you for joining us. I will focus my comments on the state of the business. Dario will go through the financials.
The two most significant highlights for the quarter were the stabilization of ASPs and improvement in the book-to-bill ratio. First, let us talk about ASPs. In the third quarter, we saw a return to a more stable pricing environment, reversing the difficult business conditions that we experienced in the first six months of the year.
For the first time in more than one year, price per bit rose sequentially, reflecting stability across product line, leveraging the strong demand for MirrorBit solutions. This is in contrast from previous quarters where price per bit declined sequentially 18% in Q2 and 12% in Q1. We are very encouraged by this recent trend as we enter now the seasonally strong fourth quarter.
Let us talk about the book-to-bill. In Q3, book-to-bill ratio was at 1.3. This is the highest level we experienced in more than one year and is giving us confidence that the overall demand for our product is robust as we enter Q4.
Booking strength was across product lines and regions, except in wireless Japan, where we experienced an inventory correction due to the prior build-up of inventory in the early part of the year.
Demand for 90-nanometer MirrorBit was particularly strong and we intend to take advantage of this momentum with Fab 25 now fully ready at 90-nanometer and 300-millimeter 90-nanometer wafers available out of TSMC.
From an operational standpoint, revenue and gross margin increased in the quarter while total operating expenses, which include R&D and SG&A, declined, despite the accelerated start-up of Spansion 1. Dario will provide further detail on this.
I am now going to comment on our results by division. The consumer set-top box and industrial division business, CSID, had another record quarter, with revenue up 8% to $294 million. Strength was general and ASP per bit stabilized. In fact, in this division, blended ASP increased. We again leveraged the transition to high density MirrorBit solutions, gaining particular traction with the one-gigabit NOR offering.
We further penetrated the serial, peripheral and interface segment with backlog almost doubling compared to last quarter.
Q3 was another good quarter for CSID, with continuing traction in market share. Book-to-bill in Q3 was again greater than 1 for CSID, reflecting a strong backlog to finish the year.
In the wireless solution division, WSD, revenue was down 7% to $317 million, due to the inventory correction in Japan that reduced revenue by approximately $40 million of high ASP, high density product. Note that we had anticipated this correction and reprioritized manufacturing to other businesses to minimize the financial impact.
We were glad to see our business outside Japan growing quarter on quarter by $20 million, as a result of successful penetration of top five handset OEMs and success in China.
ASP per bit outside Japan stabilized and worldwide unit shipments went from 82 million units in Q2 to 96 million in Q3, indicating that we are making good progress in unit share.
Book-to-bill in WSD was extremely robust, higher than 1.5, driven by the acceptance of our 90-nanometer MirrorBit technology solutions, positioning us for an upside during the holiday season.
Other significant events during the quarter include the announcement of the official opening full production of Spansion 1, a factory fully dedicated to 65-nanometer MirrorBit technology. We will use Spansion 1 to produce MirrorBit at 65-nanometer this year. And then we will qualify our entire NOR product family productions as early as early ’08, including the new Eclipse architecture.
We are glad to say that this facility is the only 300-millimeter 65-nanometer NOR factory in the world that has been designed to take advantage of MirrorBit technology. Our intention is to migrate quickly Spansion 1 to 45-nanometer toward the end of ’08.
This path will accelerate our cost leadership and leverage for high density NOR device architectures.
Before closing, I want to reiterate how excited I am about the announcement last week on the pending acquisition of Saifun. This transaction will have three impacts on our company. First, to secure comprehensive intellectual property around our successful MirrorBit technology; second, it will enable us to enter into the technology licensing business, expanding our margin and accelerating our path to profitability; and third, expand our engineering by increasing the size of our design team by 50%.
The Saifun team brings a lot of experience, depth of expertise in areas like four bit per cell, SPI, data storage, and specific design expertise around MirrorBit. We expect the transaction to be closed in about three months and are looking forward to working together as a combined company.
With that, I would like to turn the call over to Dario Sacomani, our CFO, to discuss our financial performance in greater detail. Dario.
Dario Sacomani
Thanks, Bertrand and good afternoon, everyone. Thanks for joining us. Just a few specifics of our financial performance for the quarter; net sales for the quarter were $611 million, up from $609 million last quarter. And as Bertrand mentioned earlier, our revenue performance was impacted by the inventory correction in certain parts of the Japanese handset market, which is primarily high ASP, high density 90-nanometer products.
The third quarter gross margin was up 50 basis points sequentially to 18% as our improvement in manufacturing efficiency was affected by our lower density product mix. I would say it’s probably in the range of 50 to 70 basis point impact associated with the lower density product mix.
In addition to that, we had some factory output issues associated with the earthquake in Japan, which was in the range of 30 to 50 basis points. So actually, although we improved in gross margin, I think that there was an opportunity to do better than we did.
For Q4, we expect gross margin to be approximately flat sequentially as the better manufacturing execution and the more stable pricing environment is most likely going to be offset by costs associated with accelerating the production ramp of SP1.
We are going to continue to focus on reducing our manufacturing expenses through further implementation of our wafer level test initiatives, which we’ve talked about before, increased 90-nanometer output, and better rationalization of our back-end test costs. All of those three initiatives we’ve talked about before as being our key cost reduction efforts.
Research and development expenses were flat at about $111 million, or 18% of revenues, as the increase in the SP1 ramp-up costs were offset by benefits related to our cost control initiatives.
Based on the current SP1 production timing, the majority of the start-up costs for Q4 are going to remain in R&D. With that, we anticipate R&D expenses to increase about 3% from Q3 to Q4. Now, any changes in these, the timing of these production qualifications are merely going to result in a reclassification of expenses from R&D to cost of goods sold.
Q3 sales marketing, general and administrative expenses were $58 million, which represented 9.5% of revenue, which is down from 10% of revenue in Q2 of ‘07. We expect the SG&A expenses to be approximately 9% of total revenue in Q4.
Operating loss for the quarter was $59 million, down 9% versus last quarter, as we saw a slight improvement in our gross margin and benefited from our continued focus on our cost control initiatives.
Net interest expense for the quarter was $17 million and reflects the elimination of certain one-time benefits that we talked about in Q2 of ’07, primarily related to real estate gains in Asia.
We anticipate that net interest expense will be approximately $18 million in Q4. We also recorded a tax benefit of $4.3 million for the quarter related to adjustment of allowances in our Japanese operations. We expect taxes to be nil in Q4.
Q3 net loss was $72 million, compared to a $67 million loss in the second quarter of ’07, and Q307 net loss per share for the quarter was $0.53.
Moving on to the balance sheet, we remained focused on cash management. DSOs declined to 52 days from 58 days last quarter. Days payable increased to 85 days from 72 days, reflecting the continued impact of our large CapEx spend for SP1 in Q2.
Inventory days were 97 days, up from 87 days in Q207 and reflects our increased build of 90-nanometer parts for anticipated seasonal strength in Q4 of ’07.
CapEx for the quarter came in at approximately $195 million, down from $595 million last quarter, as we continued to spend on our SP1 facility. Approximately 75% of the CapEx spend in the quarter was for 65-nanometer and 45-nanometer development, primarily at 300-millimeter wafer diameter, as well as our improved testing initiatives. We still anticipate CapEx to be around $1 billion for the year, depending on the timing of deliveries. And depreciation for the quarter was $130 million.
At the end of Q3 2007, our cash and short-term investment balances were $529 million, down $189 million as we paid for the SP1 equipment that we purchased last quarter. Net debt was up $387 million due to the decline in cash and the partial draw-down of our GE equipment loan related to our SP1 investment.
Revolver capacity at the end of the quarter was approximately $180 million, as well as approximately $200 million to be available under our GE term facility.
EBITDA for the quarter was approximately $71 million, which is actually up $8 million from Q2 if you exclude the benefits of the Asia land sale. And cash flow from operations adjusted for the accounts payable fluctuation related to SP1 in Q2 was approximately $70 million. If you’ll recall, we had a big increase in cash flow from operations last quarter and a decrease in cash flow from operations this quarter, but if you adjust the two quarters for the anomaly associated with the CapEx spending in Q2, it’s about $70 million worth of cash flow per quarter.
And with that, I will turn it back to Bertrand to discuss our expectations for Q4. Bertrand.
Bertrand F. Cambou
Thank you, Dario. Looking forward, Q4 is a seasonally strong quarter in this industry and all indicators point toward a typical pattern this year. ASPs are stabilized and we anticipate ASP erosion to be back to a more typical level, with ASP per bit down 6% to 7% sequentially.
Given these trends, we estimate that Spansion revenue in Q4 will be in the $640 million to $700 million range; gross margin will be approximately flat sequentially, as the benefit from the revenue increase will be offset by the extra burden of starting Spansion 1 in volume production.
Our cost control programs are still in motion and we expect all the cost parameters to be rather good.
In conclusion, the stabilization of ASPs in Q3 was encouraging for us, as the very difficult environment that we observed in the first half of the year seems to have improved. We are also very encouraged to see that our booking rate points toward growth and continued acceptance of MirrorBit technology, in particular at the very cost effective 90-nanometer node.
Finally, with both the successful acceleration of our new 300-millimeter factory Spansion 1 at 65-nanometer and the agreement with Saifun are significant events that are going to position us well for success in 2008 and beyond.
With that, I would like to open up the call for questions. Thank you.
Question-and-Answer Session
Operator
(Operator Instructions) We’ll go first to Glen Yeung with Citigroup.
Glen Yeung - Citigroup
Thanks. Dario, questions for you to start with; I’m not sure I understood what you were saying about R&D. Your suggestion that R&D in the fourth quarter will be up 3% to 4% because of some of the expenditures for starting up SP1 will still be R&D expenditures at that point?
Dario Sacomani
Correct. The timing of when we turn it on, it could be early in the fourth quarter, it could be the beginning of December, it could be towards the back of December, so what I did is I guided you guys as though it was all going to be for the quarter in R&D. If in fact we get earlier qualification than that, than some of that is going to just shift from R&D to COGS, but I tried to guide you with a flat gross margin and a 3% increase in R&D. If that switches a little bit due to qualification timing, you will still be okay on your bottom line.
Glen Yeung - Citigroup
I guess the question is why wouldn’t gross margins be higher than in higher revenues if you are not putting the expenses of SP1 into COGS?
Dario Sacomani
I anticipated that we would turn on depreciation to something in the range of $7 million to $10 million, and how much of the R&D material actually gets reclassed up into COGS. Like I said, I didn’t know yet so I tried to guide you so that you would get the right bottom line.
Glen Yeung - Citigroup
I’m sorry, I’m still a little confused here. So are you suggesting that gross margins could in fact be higher than flat and depending on what happens to R&D, or is there a reason that gross margins would have been lower in the fourth quarter if not for the issue of SP1?
Dario Sacomani
No, absolutely not. I think given the ASP environment, I think that they would have been up but I am assuming in that number I gave you around $7 million to $10 million of incremental depreciation and I’ve assumed that all the engineering material continues to be in R&D. If in fact some of it transfers up to COGS, I’m expecting a lower gross margin. But no, I think excluding SP1, I think it would have improved.
Glen Yeung - Citigroup
Okay, I understand. I think I got it. Thanks. The other question I had was just thinking about the mix of consumer versus wireless and maybe Bertrand, for you; as we look into the fourth quarter, what do you expect that mix to look like? How much more business are we seeing in consumer? And maybe if you could just answer that also in terms of revenue mix but also, what’s the relative profitability there? It sounds like the pricing in consumer is a little bit better and as we shift towards consumer, does that help or hurt you?
Bertrand F. Cambou
Let me try to give you -- of course, we are looking at the guidance right now here, but what we are looking at is the CSID has been kind of every quarter going up, I would say like a clock in the sense of gaining market share. Obviously as we are gaining a lot of market share, we start to get to a point where now market share is going to be -- I don’t have the number here but it is not going to be very far from 40%. And compared with essentially 25% 18 months ago, which means we are getting to a point of keep increasing is not going to be as easy as it was.
I anticipate CSID to increase a little bit in Q4 but the bulk of the growth is going to be coming from the wireless outside Japan into the big OEM, where actually if you look at my margin compared with CSID, this is similar. This is actually a good business that we have lined up. This is 90-nanometer MirrorBit. That’s the bulk of the growth in wireless.
I feel that this growth is going to be good quality business for us.
Glen Yeung - Citigroup
And you are confident that it’s really what’s in your backlog right now?
Bertrand F. Cambou
Yes.
Glen Yeung - Citigroup
You’re confident why Japan will come back, it’s because the backlog is so --
Bertrand F. Cambou
But Japan is not going to come back.
Glen Yeung - Citigroup
Sorry, not going to come back.
Bertrand F. Cambou
Japan is not going to come back. Japan is going to be as-is. We are actually modeling it flat. It could be a bit up, yes, but we are not going to count on it. All the growth is essentially the big top OEM, a little bit in China, but it is really solid, you know, big account, you know -- and this is using our most cost-effective technology.
Glen Yeung - Citigroup
I just have one quick last question for Dario, which is when do you think gross margins can start to actually go up? I guess that depends somewhat on start-up charges for SP1 go away?
Dario Sacomani
I think as we get to -- you know, we are trying to get ourselves to about 2,000 wafer starts a week, which I think we should probably get to by the end of Q1 of ’08. I think at that point in time, I think we’ll probably see some start-up, some pressure on gross margin in Q1 but I would suggest that perhaps once we get into Q2, we start seeing an improvement in gross margin again.
Bertrand F. Cambou
To echo what Dario said, 2,000 wafers a week is what we have lined up in the plan and right now, we are looking at the end of Q1. At that point, of course, in SP1 is going to be competitive with what we are currently achieving and getting the kind of break-even at that point. After that, of course we are going to keep ramping up Spansion 1 and that is going to be a help rather than a drag.
Glen Yeung - Citigroup
So flat gross margins again in Q1, you think?
Bertrand F. Cambou
Well, in Q1 obviously it is too early for us to guide you because we want to better understand the seasonality pattern, which means that it is too premature, Glen. Ask us the question next quarter.
Glen Yeung - Citigroup
Okay, thanks.
Operator
We’ll go next to John Pitzer with Credit Suisse.
Ahmed Seraf - Credit Suisse
Thank you. This is Ahmed Seraf calling in for John. I was wondering, could you elaborate a little more on your market share comments and give us a competitive landscape update? Specifically, could you address -- one of your Asian competitors last week said they were taking market share as well. Have you seen any impact from that?
Bertrand F. Cambou
If we look at the market share, the CSID business has been kind of very steady, stable gain and if you look at the last 18 months, we had a huge gain. It has been quarter after quarter of going up a significant number every quarter, to a point now that we are almost 300 million a quarter on CSID.
On this particular one, the company in Asia that you are referring, are not being very effective so far. We know that they are trying to penetrate some big account. Within the consumer space, they are kind of a pretty much of a niche.
Now, in the wireless space, if I look at some of the major wins that we have in our backlog that has been created, this 1.5 book-to-bill that we reported, some of that has been a direct hit taking market share away from them.
We think that in the near future, we are going to take some business away from them in wireless with our 90-nanometer family that has been very well accepted into the top OEM company.
With that being said, obviously the competition is intense and I don’t know who is gaining market share on whom here, but my sense is as a company, we are very focused, like we say, on the CSID and into the major account in wireless and in those accounts, we are kind of [inaudible] in our approach to market here and we think that we are gaining traction on that space.
Ahmed Seraf - Credit Suisse
Okay, and a quick follow-up for your Q4 guidance, what pricing assumptions do you have going into that or for that guidance? Also, what have you seen quarter to date for pricing?
Bertrand F. Cambou
Like I was alluding during my prepared remarks, we are anticipating a typical level of ASP erosions of 6% to 7% per bit per quarter. That is kind of the hypothesis we took and that is in line with the -- this is in line with what we are looking at in the industry.
Ahmed Seraf - Credit Suisse
Okay, and if I could just ask one last quick one, for your Saifun acquisition, looking at the external customers, have you got a chance to go to them and get a sense of are they renewing their contracts, even though they might be sourcing from what they perceive as a competitor now? Do you have any confidence around those external customers?
Bertrand F. Cambou
I actually received several -- I’m going to meet some of them later on here but I received a few contacts from some of them and they are pretty happy because obviously one of the most pressing issues they had is the know-how with Spansion property, and for them to fully enjoy the Saifun IP, they were kind of constrained by not infringing Spansion IP, which is right now we have an opportunity to increase the satisfaction of some of those Saifun customers going forward.
And of course, we are going to work on a case-by-case, but by and large, we received some relief from those accounts which are looking at the merge as an opportunity for them to freely win in that space.
Ahmed Seraf - Credit Suisse
Sounds great. Thank you.
Operator
We’ll go next to Aaron Husock with Morgan Stanley.
Aaron Husock - Morgan Stanley
Great. Thanks for taking my questions. I guess first, I was wondering can you give us a sense for how much the unit impact was from Japan wireless? Because you still put up a very good handset unit number.
Bertrand F. Cambou
The unit in Japan is always very low and ASP very, very high, which means that you have to look at these units as being kind of a small number. If I was to say a couple of million units, that would be kind of a close call. Over the 96 that we shipped, as you can see, this is not a big, big number.
Of course, on the revenue standpoint, that’s a $40 million impact, which is given you the idea of the type of ASP that we are talking about.
Aaron Husock - Morgan Stanley
I guess looking at the total content per handset that you sold in the quarter, it was down about 20% sequentially, so even if I add back the Japan impact, it looks like it was probably still down meaningfully.
Bertrand F. Cambou
No, if you actually take the $40 million away, we grew by $20 million and units shipment too were 40 million up, which means that the mix was slightly less but I would say this is 5% to 10% type of number. But the bit was also less, which means that the ASP was, like we say, per bit was pretty stable.
Like I was indicating to the caller previously, what we have on the backlog is all 90-nanometer and that we have in the wireless space, 1.5 book-to-bill, and this is high density that we are going to essentially sell during the holiday season, which means that you can look at the little [pause] on the content for us in Q3. However, on the backlog is going to be quickly reversed as we are going forward.
Aaron Husock - Morgan Stanley
Okay, good, good. Can you give us what your ORNAND sales were in the quarter?
Bertrand F. Cambou
The ORNAND sales went down by essentially $40 million, because everything we were doing in Japan was ORNAND. However, a couple of very encouraging news on the ORNAND.
We currently receiving Q3 a backlog of $10 million of ORNAND Quad and we are very encouraged. You know, $10 million in a quarter is not any more negligible and for us, the plan is to go very, very had at 65-nanometer directly out of Spansion 1 for the first essentially three to six months. The bulk of Spansion 1 is all going to be ORNAND 65-nanometer, which means that altogether, we see the ORNAND revenue decreasing a bit in the second half of ’07 to essentially rebound quite well in 2008 with the combination of the quad that is now getting more and more acceptance and the 65-nanometer 300-millimeter wafers.
Aaron Husock - Morgan Stanley
Okay. That makes sense. I guess for Dario, roughly how much in R&D dollars a quarter is related to the SP1 start-up costs at this point?
Dario Sacomani
It was up $1 million in Q3. It was about $13 million compared to $12 million in the prior quarter.
Aaron Husock - Morgan Stanley
Okay, and is that -- okay, got it. And just one last one, what was options expense in the quarter?
Dario Sacomani
It was about $4.5 million.
Aaron Husock - Morgan Stanley
Okay, great. Thanks, guys.
Operator
We’ll go next to Daniel Amir with Lazard Capital.
Daniel Amir - Lazard Capital
Thanks a lot. Thank you for taking my call. A couple of questions; first of all, a follow-up on the ORNAND -- I mean, considering the Eclipse coming out next year, I guess what, end of Q108, what type of momentum do you expect in the ORNAND for next year in terms of year-over-year growth?
Bertrand F. Cambou
Obviously ORNAND is a new architecture right now. It is receiving some fantastic customer receptions. One of the opportunities is going to actually set it into a NOR socket and that will be potentially major for us, as a significant piece of NOR could switch to Eclipse.
Now, in the ORNAND space, the 65-nanometer is going to be potentially bigger than the 90 because this is -- the cost structure is much more competitive. The 90-nanometer, we had the 200-millimeter wafers versus 300 for 65 and all density are very price competitive, which means that what we expect is a way -- I don’t know what, if we can give you a number here but it is going to be significant -- twice as much or more, I don’t know yet but this is in this order of magnitude.
Daniel Amir - Lazard Capital
Okay. The second one, you are currently -- I guess there’s currently the negotiations going on a bit for next year on the NOR pricing side. Can you comment a bit where does this stand right now, what’s kind of your impression maybe compared to last year?
Bertrand F. Cambou
Obviously we have a case right now where some customer need us more than they need last year. In particular, with the merge between our two major competitors where suddenly some companies find they have only one source and they are talking to us right now, and that’s good timing for us, as they are essentially either new customer or new as a desire to go much higher.
Across the board, I would say that the situation we have right now is normal, which means that we have to give some price concessions, no more than we usually do but this year, we have a lot of cost reduction in the pipe between the next generation technology. The 65-nanometer is coming very hard and the large wafer diameters we have a lot to offer here, which mean actually we can, on a negotiation standpoint, we think that we have -- I don’t want to say the upper hand, but we have an opportunity to do rather well we think this year.
Daniel Amir - Lazard Capital
Okay. And then my final question, what do you think CapEx will be for ’08? I guess around $1 billion here in ’07 -- what’s the -- obviously it’s coming down in ’08 but any ballpark figure, Dario?
Bertrand F. Cambou
I’m going to let Dario answer but before he answers, all he can say is it is going to be a function of how we want to or fast we want to go to Spansion 1.
Dario Sacomani
I was just going to add, it is; it’s a function of demand, which we are working through right now, demand and affordability and that’s what’s the pacing item for SP1 and we’ll hope to give you a lot better guidance on that during next quarter’s call.
Bertrand F. Cambou
To give you some information here, if I can, a document, right now we essentially have baked a plan to go to 2,000 wafers a week, like Dario said. With an extra $400 million or so, we can go to 4K per week. Now, the question here -- in Spansion 1.
Now the question here is do we need that much, and we are looking at scenarios right now maybe doing only 3K per week or, you know, and so on and so forth, which means -- there is a $400 million incremental spending in Spansion 1 and we have not decided anything right now.
Daniel Amir - Lazard Capital
Okay. Thanks a lot.
Operator
We’ll go next to JoAnne Feeney with FTN MidWest.
JoAnne Feeney - FTN MidWest
Thanks. Good afternoon, folks. Just a further question about the manufacturing side of things, just trying to understand the timing of the transition from 65 to 45, but perhaps before we get to that, maybe you could summarize for us where you are doing your manufacturing, how much you are using in house, Fab 25, how much you are still using foundry capacity, the old JV fabs, and how much TSMC? Just so we can get a feel of where your capacity is coming from and what kind of process technology you are emphasizing at this point.
Bertrand F. Cambou
Let me quickly say here, I’ll start, JV1 and JV2 that we sold to Fujitsu, we are keeping the facility at constant. It is loading right now, essentially full. The demand is there. JV3 is all converted right now essentially to 110-nanometer and it’s full, and there is no changes at 110. We are still using TSMC as well and that one next year may go down. As we are transitioning to 90-nanometer, we are not going to need as much foundry at 110.
At 90-nanometer right now, Fab 25 is full. We are using TSMC foundry at 90-nanometer and as we are going to migrated to 65-nanometer next year, we have several options here and we are going to study that. But one option here would be to reduce our foundry, convert Fab 25 to 65, and of course going to 65 on foundry as well. Of course, this is a work in process right now that we are going to essentially -- we have built the company right now with quite a bit of flexibility and we are spending our money into the leading edge technology node, and obviously this is where we wanted to be because then we can respond to ups and downs of the business and to maximize our response to the customer in an appropriate way.
Dario Sacomani
JoAnne, just to quantify a little bit about what Bertrand said, about 35% of our starts in the quarter were 90-nanometer and of that, TSMC was about a third of it. So overall, TSMC is getting close to about 15% of the wafer starts. So we’ve got flexibility.
JoAnne Feeney - FTN MidWest
If I could follow-up, it sounds like then you are not in as much of a hurry to decrease use of the old JV fabs and switch them over, switch that capacity over to SP1. Is that correct?
Bertrand F. Cambou
The old JV, this is all the small densities, commodity parts, 16, 32-meg, 64-meg -- this is all the legacy parts. Eventually, we are going to migrate that to 110-nanometer, which is JV3. We are not going to transfer that to Spansion 1.
Spansion 1 is essentially either cost reductions of mid- to high-density, I would say starting at the 256 megabit and above, as well as aligning us to participate into the emerging high-density part of the industry.
JoAnne Feeney - FTN MidWest
And then, it seemed like the move to 45-nanometer has been slowed a little bit. It seemed like that was originally planned for early or middle of ’08 and now you are talking about the end of ’08. Is that because you are just not seeing quite the uptake in terms of aggregate demand or there’s a lot of capacity in the industry at this point, so you don’t see any reason to rush that? How do we understand that?
Bertrand F. Cambou
The way you have to understand that is we don’t want to essentially confuse our manufacturing organizations. Right now, we are loading factory Spansion 1 at 65-nanometer 300-millimeter, and we want to have that transition to be flawless. And the 45 is going to follow right after that.
Now are we about the schedule where we are planning to be? Yes, I think as an industry, if you look at the business we are in, the 40, 45-nanometer migrations, these are late ’08 and I believe we are going to be consistent with the best-in-class memory company.
Don’t forget that at that point, we need immersion lithography. Those too are not cheap and there is some kind of maturity that we have to master before we go to high volumes. We are very pleased with the shrink ability of MirrorBit at 45-nanometer. We are very, very active on the program as we speak and are planning that to happen in the ’08 timeframe.
JoAnne Feeney - FTN MidWest
Okay, so it sounds like then you feel like at 65-nanometer, your costs will be very competitive and so you don’t see the reason to rush to 45 at this point?
Bertrand F. Cambou
At 65-nanometer using 300 millimeter wafers and MirrorBit, we believe that we are going to have a best-in-class cost structure in our industry and that is obviously going to be the case for the 2008 timeframe.
Dario Sacomani
And as we’ve talked about before, I don’t know if you’ve heard us talk about it before, JoAnne, but we get a real significant benefit in costs as we go to 65-nanometer because we’ve got a microcontroller on board every chip and it allows us to do built-in self-test, so to your point and to just echo what Bertrand said, 65-nanometer is a great cost reduction node for us.
JoAnne Feeney - FTN MidWest
Okay, terrific. Thanks very much.
Operator
We’ll go next to Betsy Van Hees with Cowen & Company.
Betsy Van Hees - Cowen & Company
Thank you very much for taking my call. I just had a couple of questions. First, can you give us a breakout by technology, floating gate, MirrorBit, ORNAND, your revenue in Q3?
Dario Sacomani
MirrorBit percent?
Betsy Van Hees - Cowen & Company
Yeah, the percent of revenue.
Dario Sacomani
Seventy-one percent, Betsy, was MirrorBit.
Betsy Van Hees - Cowen & Company
Okay, and ORNAND and floating gate?
Bertrand F. Cambou
I think ORNAND we did what, $55 million?
Dario Sacomani
Close to 50.
Bertrand F. Cambou
A little bit less than 10% on ORNAND.
Dario Sacomani
That’s in the MirrorBit number, Betsy. So it’s 71 --
Bertrand F. Cambou
And then 30 or 29% would be floating gate.
Betsy Van Hees - Cowen & Company
Okay, great. Thank you. And then I also had a question in regard to wireless. What percent of wireless business is multi-chip packaging?
Bertrand F. Cambou
Seventy-five percent.
Dario Sacomani
At least.
Betsy Van Hees - Cowen & Company
At least 75%? All right, and then on the memory, on the addition memory that is being packaged in that, that’s essentially a pass-through to the customer, to the end customer, correct?
Bertrand F. Cambou
Essentially.
Betsy Van Hees - Cowen & Company
Okay, great. Thanks. When looking at the balance sheet, can you give us a little color as to it went up quite a bit sequentially, up over 10%. Can you give us a little bit of an idea; is that largely due to the Japan product being pushed out and is that in there and when we can expect that to ship?
Bertrand F. Cambou
You are talking about the inventory right now?
Betsy Van Hees - Cowen & Company
Right, inventory.
Bertrand F. Cambou
Okay, no, inventory actually -- no, we did not build inventory in Japan because we anticipated -- what we had in Japan -- let me explain it.
The customer, which is the telephone operators in Japan, is not selling their phone and that stuff has been rippling through the channel but we saw it coming. Already last quarter we switched capacity into what we need to do, which is currently because the business is going up with the book-to-bill of 1.3 on average, 1.5 in quarter, we have to build ahead, if I can say, to be able to increase revenue in Q4. We should have the revenue, the inventory that we essentially created in Q3 is to be successful in serving our market in Q4. Nothing to do with Japan.
Betsy Van Hees - Cowen & Company
Okay, great. And then if I understood you correctly, you said that Fab 25 was full, meaning that you are at capacity?
Bertrand F. Cambou
Totally at capacity, yes.
Betsy Van Hees - Cowen & Company
Okay, totally at capacity. Given that you are at your capacity basically constrained at Fab 25, is there any jeopardy to the guidance that you’ve given us or can we expect there could be some potential --
Bertrand F. Cambou
No, because the guidance that we are giving you incorporates what we can deliver and by the way, there is a lot of improvement right now at Fab 25. We are essentially investing a bit in Fab 25. We want to increase the run-rate in Fab 25. We have TSMC as a buffer that is going to be much bigger going forward. We have Spansion 1 that is going to start to contribute pretty soon. And then of course, we have the inventory prepared to meet the number.
The guidance is actually totally supported by our manufacturing plan right now.
Betsy Van Hees - Cowen & Company
Okay, great. And then Dario, just a question for you on the income statement for next year. Can you give us an idea of how we should be looking at taxes for you guys next year?
Dario Sacomani
I think you are probably safe to say -- it depends on the mix from a jurisdiction perspective but I would say just to model anywhere between 25% and 30%.
Betsy Van Hees - Cowen & Company
Okay. All right. Thank you so much for taking my call.
Operator
We have time for one final question and that will come from Bobby Gujavrty with Deutsche Bank.
Bobby Gujavrty - Deutsche Bank
Thanks, guys, for squeezing me in. Last quarter you mentioned that the DRAM pass through had some impact on revenues and margins. Did that have any impact this quarter or negligible?
Dario Sacomani
It was negligible. That’s why we didn’t mention it. It was pretty much flat.
Bertrand F. Cambou
Yes, exactly because we had the exact same DRAM content quarter on quarter.
Bobby Gujavrty - Deutsche Bank
Okay, fair enough and back to inventories; do you expect them to be down next quarter?
Bertrand F. Cambou
We are going to watch Q1 very carefully and as you know, Q1 we have the Chinese New Year, which is a very exciting time. The problem with the Chinese New Year -- this is not a problem but this is a reality that is -- we need to have product in January because the Chinese are going to take vacations on February and we need to be sure that everything is done before that vacation, which means that we need to have inventory in place for the first part of the quarter to respond to Chinese New Year, which means that we are currently asking our sales team to place all the backlog for Q1 to be sure that we are ready to sell to market. And that is going to essentially dictate our inventory at the end of the quarter.
Bobby Gujavrty - Deutsche Bank
So flat, maybe something like that?
Dario Sacomani
Even flat would be down to like 87 days from 97, so from a days perspective, even if it is flat, the volume increase in Q4 is going to reduce the days by about 10 days. But like Bertrand said, it is going to be dependent on the need we see for January.
Bertrand F. Cambou
On the inventory standpoint, I have to add that as we are watching inventory in the channel to our distributor right now, we are at the low end of where they want us to be, which means that there is definitely -- I don’t want to say a shortage in the channel but we are in an environment right now where we need parts. Of course, we need the right parts and the right mix and -- but so far, it is a pretty good business environment right now for us.
Bobby Gujavrty - Deutsche Bank
Okay, fair enough. And just on the margins, it sounds like -- I mean, it sounds like you guys are doing pretty well on cost reduction. Depreciation seems to be robbing you a little bit from the margin improvement on an income statement but your cash margin is clearly improving.
Do you expect that maybe depreciation will be a bit of a headwind in the first half of next year as well, in terms of gross margin improvement? I understand your cash margin will continue to improve but just on an income statement basis.
Dario Sacomani
Like we just were talking about earlier from one of the other questions, just to -- you’re right. I mean, it’s going to -- we are probably going to have some SP1 start-up costs in Q1 putting pressure on gross margin but like we talked about, from a P&L perspective, once we get that thing running at a clip of 2,000 a week, hopefully towards the end of Q1, we should kind of get over the serious incremental margin pressure.
But you are right; it’s going to be a headwind, particularly as we get the start-up of SP1, which is the driver of the incremental depreciation.
Bobby Gujavrty - Deutsche Bank
Okay, and just a last question, more of a market kid of commentary; do you see any impact -- DRAM prices are obviously quite low and it looks like NAND prices are following them. Do you see that that has any impact on kind of the DRAM/NAND combination indirectly becoming more cost competitive? Does that put any impact on NOR or they are just very different market segments and there are very few where there is overlap between the two solutions?
Bertrand F. Cambou
It seems to be that on the NOR market, we received a beating so far this year. How much lower do you want us to go as an industry? We kind of were the leading -- I don’t want to say dog here, but we kind of took a lot of beating on that one which -- we think we are at a point right now with the NOR, the NOR is right-priced and the customer knows that the balance, supply/demand and the cost environment is such that we are -- I don’t want to say we have reached a plateau here because who knows, right? Things can change but there is an overall understanding here that room for cost reductions that have been out of control at the beginning of the year is most like over. But again, sometimes you have double-dip. What you say may happen but so far, that’s not what we are observing and that’s definitely not what we observed in Q3.
Bobby Gujavrty - Deutsche Bank
Okay. Thanks, guys.
Operator
At this time, I would like to turn things back to our speakers for any additional or closing comments.
Bertrand F. Cambou
We would like to thank everybody for spending time with us and listening to the Spansion story. Thank you. Bye-bye.
Operator
Again, that does conclude today’s conference call. Thank you for your participation. You may disconnect at this time.
Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.
THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.
If you have any additional questions about our online transcripts, please contact us at: transcripts@seekingalpha.com. Thank you!