07.19.07 SanDisk Q2 Conference Call
Eli Harari, Chairman of the Board, and CEO
Judy Bruner, Executive Vice President, and Administration & CFO
Sanjay Mehrotra, President, and COO
Lori Barker Padon – Senior Director of IR
****
Lori Barker Padon, Senior Director of Investor Relations
Thank you. Good afternoon and welcome to the financial teleconference for
SanDisk Corporation for the second quarter of 2007. I’m Lori Barker Padon, SanDisk’s Senior Director of Investor Relations. Joining me is Dr. Eli Harari,
Chairman and CEO of SanDisk, Sanjay Mehrotra, President and COO, and Judy Bruner, Executive Vice President of Administration and CFO.
The agenda for today’s teleconference is as follows: Eli will start with an update on the industry and SanDisk’s overall strategy, Sanjay will follow with operational and business updates and Judy will end with our second quarter financial results and future guidance. We will conclude the teleconference with your questions for Eli, Sanjay and Judy.
[Safe Harbor]
Now I’d like to turn the call over to our CEO, Eli Harari.
Eli Harari, Chairman and Chief Executive Officer:
Thank you Lori.
This quarter was the first quarter of operations under our new organization structure that integrates the legacy businesses of SanDisk and msystems. In managing this more global and diverse organization and business, Sanjay, our cofounder, President and COO, is leading day-to-day management of the company, and that allows me to focus on strategic issues facing the company and the industry. Therefore, Sanjay will be covering the operational and business updates in our quarterly calls, while I will address the overall strategy and our big picture outlook. Please do not view this as a signal that I plan to retire anytime soon. To the contrary, these are exciting times for SanDisk, and I am energized by the many strategic opportunities and challenges ahead of us.
The second quarter started under very difficult market conditions but improved markedly as the quarter progressed. April and May were characterized by excess supply but July has come into balance and there is a distinct possibility that demand for high capacity Flash products may outstrip industry-wide supply in the second half of this year. If that scenario plays out then the second quarter will turn out to have been the bottom of the current severe down cycle, which in turn would signal the beginning of the next industry upturn. If that is the case then our strategic investments in low cost captive capacity at Fab 3 and Fab 4 will prove to be exceptionally well timed.
I believe there are two key factors that are driving the fortunes of the Flash industry. First is the torrid pace of product innovations where Flash is an enabler for mobile products. For example, the iPhone is the industry’s first product where the only two models available are categorized by either their 4GB or 8GB iPhone for flash storage. This is remarkable. The iPhone through its innovative features has the potential to be an industry transforming product in the fastest growing and most competitive mobile market for multimedia handsets. iPhone validates everything we have been saying about flash storage in multimedia handsets and it may have far greater impact on the flash market in 2008 than the launch of the iPOD Nano had in 2005. Apple’s own consumption of NAND flash may become substantial in 2008, but also major competitors are accelerating deployment of high capacity flash in new multimedia handsets to mount a ferocious challenge to the iPhone. We believe that in the next 12-18 months we will see the mobile storage requirement in multimedia phones bifurcate into an OEM business that is driven by high capacity embedded storage, such as iNAND and Multichip Packages or even MegaSim on the one hand, and high capacity cards sold primarily through retail on
the other hand, with a gradual decline in the OEM bundled low capacity cards. SanDisk is uniquely positioned to gain from this expected shift as we have built strong sales organizations that are separately focused on mobile OEM and Mobile retail worldwide.
The second key factor is rapidly growing global markets in the new consumer economy, which is fueling surging demand in consumer electronics and mobile markets. The US economy used to be the engine of growth for the rest of the world. Today, the world economies are charging ahead across many geographies, with Europe and Pac Rim countries expected to fuel global growth of consumer demand for the foreseeable future. SanDisk will be a benefactor of these global economic trends because we are becoming a trusted global consumer brand, and because we continue to invest aggressively in global distribution of our products and in a global logistics supply chain. In the second quarter we saw a year-to-year doubling in units sold in Europe and we are expanding our sales activities to tap the huge potential for us in international markets in the coming years.
NAND flash has become one of the most strategic enabling technologies for portable consumer electronics and mobile products. NAND MLC in particular has brought affordable storage to the sweet spot of this market. According to Gartner, a market research firm, in 2001 NAND Flash was 2X more expensive per megabyte (MB) than DRAM. In 2007 DRAM is 4X more expensive per MB than NAND MLC. This is the result of DRAM pricing declining by 5.5 times between 2001 and 2007 while NAND MLC declined by 45 times. I say that again, 45 times, not percentage points. While NAND scaling will become more difficult with smaller geometries in the coming years, NAND may indeed pose an increasingly more disruptive threat to hard disk drives used in mobile applications.
On the IP front, the ITC case against STMicroelectronics went against us and, given ST’s announcement about spinning off their NOR and NAND businesses, we decided not to appeal. Both we and ST agreed to dismiss our patent infringement cases in Texas, but still have the California cases pending. We also signed a card license with Taiwan-based Ritek, and we plan to aggressively pursue licensing other card manufacturers. In addition, our SD card licensing program shows continuing growth with the increasing popularity of microSD.
Concerning the demand-supply situation in the second half of the year, prices are firming up and there are market signs that the pendulum may have begun to swing back in our direction. The market has now had two full quarters to absorb the industry-wide transition to MLC late last year which caused the abrupt inventory bubble that had to work its way through the channels. That was a one-time occurrence that should not repeat itself at the end of this year. With price elasticity and new products driving megabyte and unit demand worldwide, we believe the industry will likely be tight on supply of high density MLC components as we move into the seasonally strong back half of the year. We are fortunate that we control our destiny through our captive supply of low cost NAND MLC at Fab 3 this year and Fab 4 in 2008. I am optimistic that our long term strategy for market leadership should translate into improving financials in the second half of 2007 and in 2008.
I will now pass it on to Sanjay.
Sanjay Mehrotra, President and Chief Operating Officer:
Thank you, Eli.
SanDisk had a solid second quarter even though the overall industry climate remained challenging. We achieved positive results in driving demand and market share for our products, managing the product supply chain, growing international sales, increasing average product capacities, diversifying product mix, reducing unit costs and controlling expenses. First, I will highlight the key developments in some of our main markets.
The digital imaging market experienced surprisingly strong sequential revenue growth which was particularly enhanced by growth in our international geographies and by sales of our high performance cards such as Ultra and Extreme. As of May 2007, total unit shipments of digital still cameras worldwide grew about 25% year-over-year, according to the Camera and Imaging Products Association. The strong DSC market helped drive more than a 2.5-fold year-over-year increase in our bits sold through the digital imaging market in the second quarter. You may have seen the announcement yesterday of our worldwide marketing campaign with Ducati: we are making a strong branding push emphasizing speed and performance as we step up customers to higher margin products.
In the mobile market we saw strong growth in retail revenue but results were light in the OEM channel due to aggressive pricing and reduced market demand for certain handset models at one of our customers. Our mDOC revenue was also light in the quarter with the industry beginning to shift towards more advanced MultiChip Packages (MCPs) that combine NAND with mobile DRAM. We believe MCPs will represent a growing segment of the embedded mobile storage market with revenues starting in the first half of 2008. For iNAND, we are also engaged in strong design-in activities at key Tier 1 mobile OEM accounts for our second generation at densities of 4GB, 8GB, and 16GB. In the retail channel sales were strong with both units and revenue increasing more than 50% sequentially. Further, average retail mobile card capacity grew about 20% sequentially reflecting continued adoption of high-capacity cards for mobile applications. Unit demand for high-capacity mobile cards was exceptionally strong and we launched the
industry’s first high-capacity 4 GB Memory Stick Micro (M2) card and the industry’s first 6 GB and 8 GB microSD cards.
For the computing market, we launched our 64GB solid state drive for notebook PCs to meet the projected growth in demand for high capacity SSDs. Furthermore, we announced that our SATA 5000 2.5-inch SSD will be used by IBM for data storage within the BladeCenter® HS21 XM. This is the industry’s first blade server to feature a solid state storage solution and we are proud to be part of that achievement. While the SSD market is still in its infancy, and we don’t expect significant revenues this year, the continued adoption of our SSD products, going forward, by PC, server, and other system solution providers positions us well to be a key player in that market segment.
We have made substantial progress toward pursuing our strategy of increasing our market share outside of our traditionally strong U.S. market. According to GFK, a market research firm, SanDisk’s overall revenue market share in Europe, across all product categories, increased by almost 4 percentage points during May compared to March of 2007. Also, unit sales of our products doubled year-over-year in both Europe and Asia. I believe that the new dedicated international sales teams we are putting in place will give us greater ability to drive our business in specific regions and I am optimistic about continued gains in international sales of our products.
Solid execution remained the hallmark of our manufacturing operations. Fab 3 continued to ramp with 56-nm process technology reaching cost-parity with the 70-nm technology early in the second quarter. Given the healthy ramp of the 56-nm technology, and the excellent yields we have been able to achieve in Fab 3, we are now more confident that we can meet our demand, even in the seasonally strong fourth quarter, with minimal reliance on non-captive supply. Fab 4 is on schedule with initial production planned for the end of the year and ramping up during 2008.
In summary, the new organizational structure and the market expansion initiatives that we discussed at last February’s analyst meeting are already bearing fruit, giving us what we believe to be strong momentum for continued growth. The innovation engine at SanDisk is strong and we are excited about the work we are doing to bring differentiated products to market. We are seeing good progress on unit cost and our strong execution on the 56-nm production should manifest itself favorably in our financial results in the second-half of 2007.
Let me now turn the call over to Judy.
Judy Bruner, Executive Vice President, Administration and CFO:
Thank you Sanjay and Eli and good afternoon everyone.
Our year-over-year increase in product revenue of 13% reflected an increase in megabytes sold of 217%, and a decline in ASP per megabyte sold of 65% as well as an increase in non-memory product revenue. On a sequential basis, our megabytes sold were up 50% and our ASP per megabyte was down 26%. Adjusting for the fact that we are no longer consolidating the Twinsys joint venture, which added $53 million to Q1, our total product revenue increased sequentially by 13%. Looking at our revenue by channel, we returned to a more traditional mix with retail comprising 67% of our revenue.
Our retail sales exhibited very strong seasonal recovery with retail revenue up 41% sequentially, megabytes sold up 74% and ASP per megabyte down 18%. The 18% retail decline in ASP per megabyte compares very favorably to 35% in the first quarter, and the 74% retail megabyte growth compares to 45% in Q2 of last year.
We believe the favorable Q2 retail megabyte growth reflects three factors: 1) the elasticity benefit of price reductions over the last several quarters, 2) the exaggerated seasonal softness in Q1, and 3) the growing retail mobile market. We saw very strong sequential retail sales growth in our imaging, mobile handset and MP3 products.
On a year-over-year basis, our retail revenue grew 9%, reflecting robust growth in megabytes sold of 236% and a steep 68% decline in ASP per megabyte. We saw the strongest year-over-year retail growth in our mobile business followed by growth in MP3 and USB sales.
Looking at OEM sales on a sequential basis, and removing the effect of Twinsys in Q1, OEM revenue was down 22% with megabytes sold approximately flat to last quarter and ASP per megabyte down 27%. As Sanjay discussed, our mobile OEM revenue was down sequentially due to aggressive price declines coupled with a reduction in unit demand from one of our handset OEM customers and a decline in mobile embedded sales
On a year over year basis, our OEM sales were up 23% with megabytes sold up 166% and ASP per megabyte down 57%. The year-over-year OEM revenue growth primarily reflects the addition of product lines from the msystems acquisition, such as mobile embedded and industrial embedded products and non-memory products such as SIM cards sold in the mobile network operator channel.
While OEM sales were weak this quarter, retail sales were particularly strong, and our overall product revenue performance demonstrates the benefit we derive from the diversity of our product lines and customers.
License and royalty revenue for the second quarter of $107 million increased 30% year-over-year and 11% sequentially and reflects the addition of our license agreement with Hynix, increased card based royalties, and higher than forecasted component based royalties. The component royalty increase is primarily the result of higher royalty rates on one of our licensees’ Q1 sales, resulting from certain sales thresholds. These higher royalty rates will not be repeated in the remaining quarters of the year, resulting in annual royalty rates by product line that are consistent with the past.
Our Q2 non-GAAP product gross margin of 19.0% improved slightly from 18.5% in Q1 primarily because our costs decreased slightly more than our prices on a per megabyte basis. The amount of inventory-related charges in Q2 was similar to Q1 as a percentage of sales and did not contribute to the gross margin improvement. These Q2 inventory charges primarily relate to certain products where our sales forecasts are less than originally anticipated. Our Q2 shipments were made with approximately 95% captive memory, consistent with the Q1 captive shipment mix.
Non-GAAP operating expenses were $169 million, $8 million less than in Q1, and significantly below the forecast I had provided of $190 to $200 million. About half of this difference is due to a shift in certain Fab 4 start-up costs from Q2 to Q3. This does not reflect any delay in Fab 4, just the timing of when the expenses will be incurred. The other reasons for lower spending in Q2 included a shift in certain sales and marketing expenses from Q2 to Q3 and better than anticipated savings from our cost containment actions.
Other Income of $38.5 million included $4.1 million, a one-time gain related to the sale of certain equity investments.
Turning to the balance sheet, cash and investments decreased $267 million to $3.2 billion. We utilized $90 million of cash in operations, primarily due to an increase in product and royalty receivables. Days’ sales outstanding are at a favorable level of 39 days. Inventory, net of reserves, increased slightly to $601 million, remaining above 90 days of inventory. We are comfortable that this inventory will serve us well to meet the expected demand in the second half of the year when non-captive supply may be more costly. We provided $123 million in cash to Flash Partners for Fab 3 equipment in the form of a note receivable, and we received $13 million from FlashVision in final payback of the FlashVision note receivable. We also invested $54 million in SanDisk capital equipment which included assembly and test equipment for our captive factory in China as well as some captive fab equipment that we now own at a foundry for our 3D products and processes. The last key use of cash was $55 million for repurchase of our shares.
I’ll now turn to our outlook. Traditionally megabyte growth in Q3 is less than in Q2 due to slower summer seasonality, and we expect a similar pattern this year. Over the last 3 years our Q3 megabyte growth has ranged between 20% and 40% sequentially and we believe this is a reasonable range to expect for this Q3. At the same time, the pricing environment is expected to be significantly more favorable for us than in the first half of the year. We have few planned price reductions in retail for the summer months however our ASP per megabyte will still decline due to price reductions implemented in June which will impact all of Q3. With our OEM customers, some of the expected Q3 business is locked in at prices quoted during Q2 which reflect a sequential quarterly reduction. Our expectations for megabyte growth and pricing lead us to forecast Q3 product revenue in the range of $750 million to $825 million. Looking at the year, we now expect bit growth for 2007 of 180% to 200%, up from our previous forecast of 170%+.
We expect Q3 license and royalty revenue between $105 and $115 million based upon the forecasted sales of our licensees. For 2007, we are raising our forecasted license and royalty revenue from our previous forecast of $400 million to a range of $420 to $430 million.
Turning to gross margin, we expect cost declines to exceed price declines leading to forecasted non-GAAP product gross margin of 20% to 24% in Q3, and we currently anticipate a continued gradual increase in gross margin for Q4 as we benefit from a higher mix of 56 nanometer sales. We expect Q3 GAAP product gross margin to be lower than non-GAAP by approximately 3 percentage points due to the inclusion of stock compensation and acquisition-related purchase accounting charges.
We forecast Q3 non-GAAP operating expenses to be in the range of $205 million to $215 million reflecting a sequential increase of approximately $20 to $25 million in Fab 4 R&D start-up costs, and increased investment in future technologies as well as in sales and marketing programs related to demand generation and expanding our international footprint. We currently expect non-GAAP operating expenses to increase again in Q4 due to seasonal sales and marketing expenses, while Fab 4 start-up costs are forecasted to be at approximately the same level in Q4 as in Q3. Q3 GAAP operating expenses are expected to be higher than non-GAAP by approximately $34 to $38 million due to stock compensation and acquisition related intangible assets.
Other income for Q3 is expected to be about $10 million lower than in Q2, due to the one-time investment gains in Q2, anticipated lower cash balances as a result of fab investment and a shift in our investment portfolio toward more tax-exempt securities which yield us a better after-tax return.
On the balance sheet, we forecast that Q3 inventory levels will likely increase in preparation for Q4, and we expect to make cash investments in Flash Partners and Flash Alliance of approximately $350 million, about half of which we originally expected to fund in Q2.
In summary, we are optimistic that the improved supply/demand balance in the NAND market will continue, leading to a return to more normal price declines and a gradual improvement in our margins in the second half of this year.
We will now open the call for your questions.
****
Q&A
Daniel Gelbtuch, CIBC: Hey congratulations on a great quarter and great guidance.
Eli Harari: Thanks Daniel.
Daniel Gelbtuch, CIBC: First of all I was wondering if you could give us some color about the Ritek deal and do you think this is going to be a model for additional deals in Asia with regards to many of the ODMs and OEMs or smaller ones out there?
Eli Harari: Yes, we do. We have put together a, frankly, a new license contract for third party card resellers, a card including USB flash drive, MP3 players and so on. And Ritek is the first one that in fact has signed up for that license agreement. We expect it to be, [pause] we will certainly offer it to a wide range of other potential licensees and hopefully they will sign up.
Daniel Gelbtuch, CIBC: Ok, and then turning to SSD. Obviously with the pricing going up for a component, I guess not for you guys, but for everyone else in the world, I was wondering where do you think we should see the inflection point for SSD coming? Is it going to be function of how NAND compares in price to magnetic? Is there some trigger point that you think we will see a tipping, or a ramp of SSD?
Eli Harari: Well, we said, I said several times, that SSD really needs to shift to multilevel cell, MLC flash even, at basically at any technology node. In order for the SSD to become attractive at 32 GB or 64 GB for the consumer space, it certainly needs to go to the 56 nanometer or 50 nanometer or so, a generation of MLC and this is what we are focused on.
In the enterprise space, the service space, the pricing is substantially higher, the requirements are different, although the market there is relatively small, I mean smaller than the notebook market, it’s still a very substantial market and as you know that the disk drive people make all of their margins, the majority of their margin in the enterprise space. So, with the announcement about IBM server design win, we’re basically signaling that we are going to be addressing both the consumer SSD market and the server market.
Daniel Gelbtuch, CIBC: Ok. Thank you very much and congratulations.
Eli Harari: I would say, really as Sanjay said, we expect this to be really marketed, to be very design win intensive over the next 12 months and really start generating material revenues in 2008.
Daniel Gelbtuch, CIBC: Thank you.
Amit Kapur, Piper Jaffray: Great, thanks a lot. I was wondering if you could provide your views on what kind of, what types of densities you expect within the mobile handset applications over the next say 12 to 24 months? Most of the phones that we see out there seemed to be capped off at 2 gigabyte memory card support. Do you see any new applications or potential iPhone impact on other OEMs breaking through that barrier soon?
Sanjay Mehrotra: We certainly see the average capacities in mobile handsets continuing to increase. If you look at last year to this year, average capacities for retail have doubled in the mobile handset business. We certainly continue to see the same trend forward. By end of the year in retail we would expect the average capacities to be about 2 gigabyte.
On the embedded side as well with our iNAND we are seeing strong design wins and these are iNAND design wins of high capacity 4 GB and 8 GB. And on the card side for mobile handsets, the support for high capacity cards is certainly coming through and then in near future we will see, we will begin to ship 8 GB cards as well later on in the year. And we will have the handset support for high capacity microSD and other high capacity cards as well. So all in all, we continue to see strong trend both on the embedded as well as usage of cards for high capacity, in the handset market.
Eli Harari: I would add that in the retail market, on the high capacity, that is 4 and 8 GB, the key issue that is being resolved as we speak has been designing the new handsets to accommodate the high capacity, the SDHC standard and that is happening right now. So moving forward, I think certainly in this whole retail season and certainly next year, the after-market will move very much to 4 and 8 GB to basically meet the iPhone phone 4 and 8 GB capacity points.
Amit Kapur, Piper Jaffray: Ok. Great. Maybe one quick follow-up. Mobile TV seems to be gaining support with some of the wireless carriers out there. But the ability to store content seems kind a limited. What are some of the barriers do you see in mobile video storage and when do you expect those to be resolved?
Eli Harari: The issue there is very much the infrastructure to deliver content, both mobile TV as well as movies in reasonable time because of bandwidth allocation. The wireless bandwidth allocations are still very severe. So this is slightly [inaudible]. But then there is also the protection DRM schemes for protecting valuable video content, movies and so on.
We think that the industry on the TV side is beginning to move very much away from selling you TV episodes for $1.99 or so, to ad supported free downloads of TV programs. We think that that’s a great move for the industry and of course for us, because, this will definitely simplify the whole infrastructure for delivery of these TV programs. In terms of downloads, we believe that TV programming will far, far exceed downloading of movies at least for the next 12 months or so.
This is a very exciting situation right now, and definitely the entire content providers are looking now very eagerly at how to move in a big way to download content, not just to standalone players, PMP players, video players, and music players, but also to handsets. Handsets is eventually going to be the biggest application for video and in fact iPhone has demonstrated or will be demonstrating, what a great platform, the handset, the multimedia handset can be for video programs. So we are very early, but this is going to be a very, very large market.
Amit Kapur, Piper Jaffray: Great, thanks a lot guys.
Paul Coster, JP Morgan: Thank you. Eli you seem quite positive about the outlook for the beginning of next year. This time around we are not going to get that one time effect of the MLC ramp, but is there anything in the new products that also gives you confidence that you will see kind of overflow of demand, and I’m thinking here will the SSD, which is presumably not a seasonal sale have any impact in the first half of next year, or for that matter the attached, the aftermarket sale of cards in these new multimedia handsets?
Eli Harari: Yeah I don’t think that SSD can absorb, can make a big difference frankly in Q1, Q2 next year in terms of the demand supply balance, but I do think that the two key factors, if you will, that will influence the demand supply balance in Q1 will be Vista on the one side which is really been directed through the DRAM market. If Vista finally really drives demand for DRAM that will definitely impact the supply of NAND through non-conversion of DRAM capacity to NAND.
And the second one is what we said the market’s reaction to the iPhone that by the first quarter next year should start kicking in our first quarter next year, the 3GSM Conference and everybody is furiously working on products that will be very attractive, significantly less expensive than $600 and will definitely use a lot of flash. And this is where we see the biggest possibility for creating a demand supply balance or in fact an imbalance in favor of demand.
Paul Coster, JP Morgan: And the second question, Judy the Fab 4 ramp, at what point do the start-up costs get rolled into gross margins? And how should we be thinking about the gross margin impact from that event when it happens, timing its etc and magnitude?
Judy Bruner: Sure. We currently expect the impact of the Fab 4 start-up cost to begin rolling into cost of sales in the fourth quarter. We expect they will continue to be start-up costs in R&D in the fourth quarter at approximately a similar level to that of the third quarter, but then additional start-up costs beginning to hit cost of sales in the fourth quarter. So that would have some muting affect on the gross margins in the fourth quarter. But as I said we’re optimistic in terms of the pricing environment for the second half and the effect that by the fourth quarter of this year we’ll begin to have much more contribution from 56 nanometer in our sales.
Paul Coster, JP Morgan: Thank you.
Satya Chillara, Pacific Growth Equities: Hi, good afternoon. So Eli at this point are you in a position to talk about cell phone revenues for the year now that it seems you have locked in lot of OEM as well as you’re seeing some retail sales. What is the prediction by the end of the year, cell phone would be for SanDisk?
Eli Harari: What do you mean?
Satya Chillara, Pacific Growth Equities: The cell phone revenue as percentage of your total product sales from cell phones?
Eli Harari: We are not providing this kind of prediction. I could tell you that in terms of unit revenue for Q2, we sold approximately 30 million units combined and we certainly expect the unit count and megabytes count to increase in the second half of the year.
And as we’ve indicated that consists of a decline in bundled cost and an increase in retail cost as well as a decline in low capacity units and an increase in high capacity units.
So it’s moving in the right direction for us. But I would expect that this year mobile will be our biggest revenue generator for the year with one proviso, and that is that as we’ve said for this second quarter the digital still camera market was significantly strong in retail than we thought and if that trend continues and that too will be a very strong market for the rest of the year.
Satya Chillara, Pacific Growth Equities: Ok, Eli, one other question on the licensing side. In terms of STMicro, you’ve gone with STMicro even though you said you are dropping it right now in terms of appeal, you have lost the case two times and contrast that with Samsung back in 1997. What’s the message I think will be, we are struggling here, why did you loose the STK’s at ITC and basically in future, how airtight is your intellectual property?
Eli Harari: We think we have very, very strong intellectual property. We lost the case, frankly the same partner has contested three times at the ITC with three different judges. The outcome in the three cases was in the first case we won on infringement and validity. In the second case, we won on validity and non-infringement. I guess we lost on validity and non-infringement. In the third, we won on infringement and non-validity.
So this is the part of the frustration that you have when you are dealing with a patent case. I would say if there was one error that we made, a strategic error is that we tried to go for a very quick resolution and therefore relied on a single patent which we thought was definitely a very strong, a pretty fundamental patent that we create. And if I had to do it all over again I would definitely bring significantly more patents to play in here.
When you bring the same patent, what happens is that the other side learns all the arguments and all the counter arguments and that was the mistake that we made. We will learn from that mistake.
Satya Chillara, Pacific Growth Equities: Ok, thank you.
Eli Harari: Thank you.
Gurinder Kalra, Bear Stearns: Thanks. Just a couple of question from my end. Now Eli, you talked about the potential of demand exceeding supply in second half of this year. Now, it looks like both Samsung and Hynix are having some yield issues with their current process transition, and it will be resolved at some point in time. How does that influence your supply projections and your potential forecast?
Eli Harari: We are not counting on them to fail. We are counting on them to succeed and they have been very good in the past at delivering whatever the market needs within limits. More importantly is how much of their capacity do they allocate between NAND and DRAM. In my opinion that’s more important than yield issues. I think that they will, [pause] my assumption is that they will overcome those yield issues.
Gurinder Kalra, Bear Stearns: Ok, thanks. Secondly, can you update us on when you plan to go into volume production on three bits per cell, x4, and also any progress on the Matrix side?
Sanjay Mehrotra: So, on the three bits per cell as we said before, we will have initial product shipments of three bit based technology early in 2008. On the four bits per cell technology, we will have samples in second half of 2008, and the three bit technology we are working on [pause] and the 3D technology, we are working on transitioning that to the next generation technology node and we are continuing to make good progress there.
On the OTP side as well as working on developing the read-write version of that technology which Eli had mentioned is the R&D efforts that will take two to three years to really fully materialize.
Gurinder Kalra, Bear Stearns: So, on the four bits per cell line, you are talking about samples in the second half of the year or for next year?
Sanjay Mehrotra: Excuse me.
Gurinder Kalra, Bear Stearns: So on the x4 you said you are having samples in the second half of next year?
Sanjay Mehrotra: Yes.
Gurinder Kalra, Bear Stearns: So that would be a slight delay from your previous statement?
Eli Harari: I said last time that we had decide to apply x4 on a more advanced, the leading edge technology node rather on an existing or trailing edge node and that basically means that we scrapped product on the 70 nanometers technology, which we actually have and are focusing on a more advanced, on really the leading edge technology in that time frame.
Gurinder Kalra, Bear Stearns: Thanks very much.
Craig Ellis, Citi: Thanks for taking the question. Just first a question for Eli. Eli, you have talked about the number of supply/demand factors into early next year. How much of your confidence in an upturn in NAND is predicated on the challenge that industry will base as it moves off of 200 millimeter capacity in the first half of next year and the dampening effect that will have on overall bit growth?
Eli Harari: I think as I have said and I think now lot of people are saying, in fact, I think I believe the Koreans are saying the same thing now, that they plan to not drive advanced technology transitions in NAND under 200 millimeters. So that will have an impact on reduced supply on the one hand, on the 200 millimeter.
On the other hand, they are investing aggressively in 300 millimeter. So I think that, I am looking more frankly at this point to the surge in demand we talked about, this global increase in demand that we, it’s a real phenomenon, and the fact that we have such a relatively small market share in many of these geographic markets is the staff optimism, that once they put in place, the kind of resources that we have, we build in the United States that we can do significantly better and that will drive at least for us, we believe continued growth and demand and account for a 100% of our output from our captive supply.
Craig Ellis, Citi: Okay. Thanks for that Eli. And then Judy, just a couple of quick ones. First on royalties. For a couple of quarters now we are in the low teens as a percent of revenue. The long-term model is really high single digits to 10%. Is there anything unusual other than the one-time item you mentioned in the second quarter royalty that’s going to impact how we think about the longer term model?
Judy Bruner: I would say really the only other factor is that we are seeing success in terms of growing our card based royalties. But at this point, I would not change our long-term model.
Craig Ellis, Citi: Ok. And then secondly, I wasn’t clear on the gross margin commentary in the outlook. It sounded like there was a similar degree of inventory write down pressure in the second quarter, as you saw in the first quarter. What is assumed in the third quarter guidance with regards to inventory write-down?
Judy Bruner: Well of course it’s difficult to predict what inventory write downs might be in any future quarter. But, we are optimistic that inventory write downs in the second half will not be at the level they were in the first half. And that could be a factor and in gross margin improvement. But the primary factor that we are looking at is the pricing environment, and then as we get further into the second half, the higher contribution of 56 nanometer product.
Craig Ellis, Citi: Ok. Thanks to everybody and congratulations on the quarter.
Eli Harari: Thank you.
Jim Covello, Goldman Sachs: Hi guys, good afternoon thanks so much for taking my question. Couple of quick things some of what you alluded a little bit too on the remarks up until now. But I guess first you have the advantage now where you have the captive supply as your competitors, your card competitors costs are going higher, how do you strategically think about the decision of increasing market share versus increasing margin?
Eli Harari: I think we should try to maximize our margin of course but keep our output at a 100% utilization. And at this stage it’s a little bit academic, it’s very difficult to frankly, the list is like this, non-captive supply today is very, very expensive and we definitely at this stage, a bit difficult to justify to bring additional captive capacity overall to gain market share at depressed or close to zero or may be even negative margin, it’s not what we would want to do at this time.
Jim Covello, Goldman Sachs: And I mean I guess another way to ask it would be with your captive supply that you have, how do you just make the decision between being price aggressive to try to go out to gain share with your captive supply versus just trying to have higher margins because, your supply is cheaper than your competitor’s supply?
Eli Harari: Well, first and foremost, we make sure that we address our strategic customers and don’t behave opportunistically. On the other hand, we as well as the industry, the supply side, have really suffered through this price decline and we do think, I have said many times that the kind of price decline we’ve seen in the last two years are really unsustainable and that to justify future investments you really needed to get to significantly better gross margins than what we have experienced in the last two quarters.
So you can improve your gross margins primarily by holding costs, by bringing cost down faster than your pricing comes down. And in some cases you want to bring your prices up because they were forced upon you to be way too low because of crazy competitive situation. We will definitely try to regain some pricing wherever we can, but not opportunistically not point of abandoning or upsetting low trend on strategic customers.
Jim Covello, Goldman Sachs: Okay that’s helpful and that kind of leads into my second question which is, we are now in a very good position in the industry but in large part because supply got cut so much, our new supply growth got cut so much because of the move to from NAND to DRAM and because of the decrease in NAND orders from some of your competitors to the equipment suppliers.
How do we get comfortable with now that pricing is looking better, that everything is not just going to come flying back on line, that your competitors aren’t going to just turn on spigot again and that you guys along with Toshiba aren’t just going to increase the Capex again because that could be what brings all this to a faster halt than we would all hope for? How are you thinking about it and how can you help us to get comfortable with the continued discipline on the supply side?
Eli Harari: It’s a good point, good question and definitely it seems to go south in the future and there will definitely be periods where we have these kind of situations, where industry as you know go through these cycles. But what’s working for us is price elasticity, and demand elasticity is very strong right now. It finally caught up with the pricing and its much, I mean we are talking about 220% increase in megabyte, tripling our megabyte shipments year after year, so we have the large number of, the theory of large number is actually now in our favor and that it’s becoming more and more difficult to double the megabytes output because the basic numbers are now becoming very large. It’s very difficult for new players to come in and make a dent.
So, I think that the industry, and everybody has suffered frankly through this experience of the last two quarters, but frankly we’ve had six quarters lost, pretty severe quarter-to-quarter ASP declines, and I would say the six quarters is the longest down cycle the NAND industry has had. And I think that the industry, all the suppliers would like to see this trend kind of turn around and much more moderate price declines, but of course the key to that is to get the demand through elasticity, through new products, through existing markets and new markets and that is what we are working on.
Jim Covello, Goldman Sachs: That’s helpful, thanks. And then my final question and then I’ll go away. In terms about the royalty stream, you’ve referenced earlier like at the analyst meeting that Samsung, you would continue to negotiate with them as they contract, that’s called the current contract, it gets closer to expiring at the end of 2009 and you had said that that would be a difficult negotiation but you thought it would ultimately be successful. And then, so any update there? and then on Micron, Micron seems pretty adamant that they’re never going to have to pay you guys a license, a royalty fee and I’m curious as to your opinion about that? Thank you.
Eli Harari: Yes, I will repeat what I said, I don’t need to repeat this basically what I said about Samsung. We still believe that that would have been the best interest of Samsung and send it to a, come to an agreement before the expiration of the current agreement.
With regards to Micron, nothing has changed as far as our thinking which is of course not the same as their thinking and we will continue to work on that one.
Jim Covello, Goldman Sachs: Thanks very much. Congratulations.
Eli Harari: Thank you.
Daniel Amir, WR Hambrecht: Thanks a lot. A couple of questions here. First, what’s the percent of production will you have a 56 nanometer by year end?
Sanjay Mehrotra: As we have said before, by end of the year we will have majority of our bit production in terms of output from the fabs in 56 nanometer. And that the 56 nanometer ramp is going extremely well, we feel pretty confident in our ability to meet that previous guidance.
Daniel Amir, WR Hambrecht: Ok. And then the question, a bit about your OEM mobile business as it relates I guess to the mDOC/ the iNAND. Can you a bit think about how does this compare with your previous expectations, I guess before the M-Systems acquisition? And is there a following against the iPhone trend here? I mean is there a change I guess in the mindset which direction it’s going here in terms of embedded NAND in handsets?
Sanjay Mehrotra: So, embedded NAND on the handset as we said in the prepared statement, really the trend is in two directions. One primary trend is toward multi-chip packages and with our partnership with Qimonda, in the joint venture with Qimonda, we plan to have multi-chip packaged solutions in the first half of next year.
Also on the embedded side, we are seeing the trend is moving toward the iNAND type of solution. A high capacity iNAND design wins, and with that we will see increase in our revenues later on in the year because iNAND embedded solutions design wins in terms of production take somewhat longer. By end of the year we will be seeing increases in embedded revenue coming in from the iNAND. So, the mDOC part is really shifting towards multi-chip packages, as well as, toward iNAND design wins.
Daniel Amir, WR Hambrecht: And where does the MegaSIM kind of fit into the strategy here?
Sanjay Mehrotra: The MegaSIM also is an attractive semi-removable solution and that is also something that we are offering to the customers and we are seeing increasing design wins in that area as well.
Eli Harari: On the MegaSIM, of course, the target customers for MegaSIM are not the handset manufacturers, but the mobile network operators, because they sell the SIM today. They own that and in their endeavor to increase their ARPUs, they are moving to more services than they could [pause]. Those services could be much richer, if they were able to store a lot of content on the MegaSIM.
Daniel Amir, WR Hambrecht: Ok. Thanks a lot.
Eli Harari: Thank you.
Steven Chin, UBS: Thanks. This is Steven Chin calling behalf of Alex. I just had a couple of more questions related to the iNAND and mDOC product lines. So as far as relative positioning going forward, it sounds like iNAND and other MCP type products will have more preference or usage in embedded cell phone applications.
Just given that, does that mean mDOC sales would begin to decline at a certain point and then you would have a crossover between iNAND sales and mDOC sales? And secondly, for iNAND MCP type products, would you guys typically be a single source for those sockets, or would it be dual source and more because of the standardized footprints for some of these products?
Eli Harari: So, really you hit the point on the second point which is the standardization. So, what is happening, in the mobile space for embedded is that there is rule among the main suppliers towards the JEDEC standard, JEDEC 64, which in itself is not just one product, but assembly of products that migrates towards eventually a bootable NAND working with a mobile DRAM and that is a direction that we are moving with iNAND and Samsung is moving with moviNAND, Toshiba has the GB NAND and its trying to satisfy the handset manufacturers that perhaps are tired of dealing with a very large number of proprietary solutions that are not really the way to go for very high volumes.
Bennett Notman, Davenport: Thank you for getting me in at the buzzer here. It seems like since you’ve locked up the bulk of the NAND manufacturing marketplace in terms of the licensing opportunity, you are now diversifying and going into sort of ancillary and related opportunities. And could you just talk a little bit about the potential for licensing outside of the manufacturers and into the other NAND market areas and how you look at that opportunity versus the magnitude of the opportunity you’ve got with the manufacturers?
Eli Harari: We’ve said that we think we have most of the NAND manufacturers and NAND MLC manufacturers covered and we of course have by virtue of SD, MiniSD, microSD being a very large part of the card business, we have that covered through the SD association.
We don’t have the other card manufacturers and USB flash drive covered in this. This is a target area that we are pursuing through this licensing program. It will be significantly smaller, than the component business, but let’s say in the next one to two years; however we’ve said that eventually we think that most flash memory will be sold as a card or as a system solution as opposed to a component and at that point, we’d like to capture both aspects of the product.
Bennett Notman, Davenport: And do you have incremental opportunities in controller, in the area of controllers or just other portions of in the enabling technology within the system or would it be just be sort of one license for the end card maker?
Eli Harari: You would rather license the card makers than the controller makers. The controller maker’s patent law, is such that we cannot license the controller and also separately double-dip if you will. If you license the controller manufacturer, and license the flash manufacturer, you basically have licensed, you’ve licensed the card. We’d rather go after the card and separately from the memory, and not license the controller manufacturer.
Bennett Notman, Davenport: Thank you.
Eli Harari: I’d like to thank everyone here for listening today and we are looking forward to seeing a number of you during upcoming conferences. Be well and thank you very much. Goodbye.
SanDisk etc 