21 July 2008 SanDisk Q2 2008 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, Senior Director of Investor Relations: Thank you. Good afternoon and welcome to SanDisk Corporation’s financial teleconference for the second quarter of 2008. I’m Lori Barker, 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.
Now I’d like to turn the call over to our CEO, Eli Harari
Eli Harari, Chairman and Chief Executive Officer: Thank you Lori, and good afternoon everyone.
I am disappointed with our Q2 results which are clearly not satisfactory. Our team is adapting quickly to the deteriorating economic climate and we are taking aggressive steps to protect our balance sheet and improve our competitiveness through these difficult times. Let me tell you how we are planning to move forward to get us back to profitable growth over the next several quarters. I will then discuss the recent announcement regarding our cooperation with Toshiba on 3D rewritable non-volatile memory (3D R/W) and its potential implications for the post-NAND flash era.
Starting with the current conditions in our markets. I believe that SanDisk and our industry are currently facing excess supply because of two factors. First, while the demand for Flash megabytes has continued to grow, the rate of growth is slower than expected due to the weakening U.S. economy. Second, supply has been growing rapidly because of the high productivity of 300-millimeter (mm) dedicated Flash fabs transitioning successfully from 56/60 nanometer (nm) technology to 43nm. This divergence has resulted in unsold inventories which has led to depressed or negative margins.
For SanDisk, we are implementing the following decisive actions to curtail our rate of supply growth until market conditions improve markedly:
First. We will push out the start of the next major phase of Fab 4 wafer output increase to no sooner than April of 2009. This will moderate the growth in our captive capacity starting in early 2009.
Second. We will put on hold our investment decision for Fab 5 until market
Third. Over the next 1-3 years we plan to increase our non-captive purchases of Flash memory from currently close to zero to 15%-20%. This will give us more flexibility to better manage our inventory.
These actions are expected to reduce our capex outlays starting in 2009 and to
improve our cash flow.
I hasten to say that these are our own actions, and are independent of any actions by our development and manufacturing partner Toshiba.
On the technology and operations side, we are staying focused on cost reductions while at the same time driving for global sales growth and product innovation. I believe our manufacturing technology execution is first class and that we are ahead of competition in our manufacturing transitions to 43nm and to 3-bits per cell. In 2009 we plan to introduce our 32nm technology and the first x4, that is, 4-bits, per cell NAND production. During Q2, we restructured to streamline, consolidate and make more responsive our customer-focused businesses and global sales. This resulted in a modest headcount reduction. We are watching carefully our opex, however we continue to invest in the following strategic areas: first, our R&D spending is staying at the current level as we aggressively pursue the development of future generations of NAND flash, while concurrently developing the 3D R/W memory that we believe will replace NAND flash sometimes in a the next decade when it [NAND] can no longer be economically scaled. We are also investing aggressively in accelerating our Solid State Drive development efforts, where we have some catching up to do. Second, our sales expense is growing slightly because we are expanding our retail footprint in faster growing geographies, principally Brazil, Russia, India and China. And third, our legal expenses reflect a more aggressive stance to enforce our storage systems patents. In our system level suit at the ITC (International Trade Commission), last week we had favorable Markman rulings by the administrative law judge and we believe we are well positioned for success at that trial.
Our marketing strategy is clear and simple: we are focused on three young markets with incredible growth potential ahead of us, and we intend to be market leaders in all three. The first of these is: storage-centric mobile consumer electronics, which encompasses imaging, USB flashdrives, portable content including music and video, portable gaming and GPS. In this space we are already market leaders in the U.S. and see major opportunities in growing our share in global markets. Second, is what I have called “the mother of all growth markets” – mobile storage in handsets, the ultimate convergence device. Here we are leading the market with our microSD and M2 card standards and we are pioneering the new MegaSIM and embedded storage solutions. Third, in the brand new market for SSD storage in portable computing, where we believe our system expertise with MLC and rich IP will allow us to catch-up and be a market leader. We aim to take full advantage of the phenomenal price reductions of Flash over the past two years, to accelerate these major new markets.
I want to briefly comment on 3D R/W memory. In the second quarter, we signed a strategic long-term agreement with Toshiba to jointly develop and commercialize 3D R/W. SanDisk has been making good steady progress since our acquisition three years ago of Matrix Semiconductor who pioneered this technology starting in 1997, and we currently have more than 200 issued patents that cover key elements of 3D rewritable memory technology. Many of these patents extend beyond 2020. Commercialization of 3D rewritable memory represents a considerable long-term undertaking with significant challenges yet to be overcome. We believe this effort is worth the prize, as 3D memory is a potential game changer and has the real potential to not only succeed NAND flash but also achieve the cost structure to disrupt HDD in the coming decade. Toshiba’s joining force with us will, we believe, accelerate our development activities and considerably improve the likelihood for success.
In summary, we are responding to the economic realities by taking the aggressive actions that are necessary to protect our balance sheet during these difficult times. Although the near-term outlook remains unsatisfactory, we are well positioned with our market initiatives, technology and IP to weather the downturn and be a strong leader in the next industry upturn.
I will turn it over to Sanjay
Sanjay Mehrotra, President and Chief Operating Officer: Thank you Eli,
This was indeed a very challenging quarter for SanDisk.
While our pricing actions were much more moderate than in the first quarter, our sales were not as strong as expected and margins were impacted by the lower volume of product sales and inventory write downs. In hindsight, we believe we should have moved pricing more aggressively in Q2 to sell additional product and lower inventories.
On the sales side, while retail sales in the Americas were weighed down by weak U.S. consumer sentiment, in EMEA we saw relatively good retail unit and revenue growth quarter-over-quarter. OEM demand was softer than expected in the mobile business due to certain top tier handset manufacturers scaling back production of high-end models.
In good news, our Fab 3 and Fab 4 production wafers continued to provide excellent yields and we continued to reduce our cost structure through our transitions to 43nm MLC and the industry’s first commercialized 56nm 3-bits per cell technology.
For the second half of 2008, we remain extremely focused on reducing the cost of memory and non-memory components in our products. We are on track to produce two-thirds of our total bit output on 43nm by the end of the year and more than 20% on 3-bits per cell by the end of 2008. In 2009, we expect approximately 50% of our fab output to be 3-bits per cell. We are making good progress on 4-bits per cell development and we expect to introduce initial products with this technology in the first half of 2009, well ahead of the competition. With our memory technology transitions, yield improvements and non-memory cost reductions, we are on-track to meet or exceed our targeted manufacturing cost reductions in 2008 and this will position us well in improving the profitability of our business once our current excess inventory and the industry demand / supply balance improves.
Speaking of new market opportunities, at the June Computex show in Taiwan, our new pSSD designed with NAND MLC for ultra low-cost netbook PC’s received excellent coverage and strong interest from leading netbook OEM’s. We also believe that the cumulative effect of price reductions will accelerate the adoption of SSDs in notebook PC’s starting sometime in 2009 and we plan to be a leading player with our MLC SSD products. Flash storage has now been adopted widely in personal GPS systems and is rapidly penetrating into digital camcorders. We are well positioned with our product line-up to continue to grow these markets and we recently announced a 16 Gigabyte 4 hour video card for the camcorder and the video enabled Digital Still Camera market.
On the sales side, our pricing strategy in the third quarter will be more aggressive than in the second quarter and the pricing actions are intended to increase unit sales and drive average capacities across our core products. Based on our early orders, robust mobile handset product refreshes, and the ripple effect of the iPhone 3G, we are optimistic that mobile card sales will improve from the Q2 levels in the second
half of this year. Our team is focused on aggressively driving sales in the second half, including maximizing our opportunities in the Q3 traditional back-to-school period, the Q4 holiday season and in the international markets.
Now, I will turn the call over to Judy.
Judy Bruner, Executive Vice President, Administration and CFO: Thank you Sanjay. I’ll start with our Q2 results and then address Q3 guidance.
Our second quarter product revenue of $688 million was down 5% on both a sequential and a year-over-year basis and came 61% from the retail channel and 39% from OEMs. Our ASP per megabyte declined 15% sequentially and 55% year-over-year, in line with our expectations, however demand was weaker than we had anticipated with megabytes sold growing 14% sequentially and 120% year- over-year. We did not experience the typical second quarter seasonal lift in the U.S. for moms, proms, dads and grads.
Our retail revenue was up 1% sequentially and down 13% year-over-year. Retail average capacity increased by a healthy 15% sequentially, however retail unit growth was weak at a 6% sequential increase. Our retail sales in units and dollars were down sequentially in the U.S. and Asia, but up in Europe. U.S. consumer demand was quite weak but we believe we maintained our market share. In Asia, we lost some share due to aggressive pricing, while Europe retail delivered solid revenue growth from Q1 to Q2.
Our OEM revenue of $266 million was down 13% sequentially and up 13% year- over-year. OEM weakness came primarily from the mobile handset vendors where we experienced a sequential decline in unit sales of bundled mobile cards and flat average capacity.
License and royalty revenue for the second quarter of $129 million increased 2%sequentially and 20% year-over-year. The growth in license and royalty revenue came from both variable NAND component based royalty and card royalties.
Second quarter non-GAAP product gross margin of 5.7% was 10 points below the middle of the gross margin range that we provided in April. Price reduction and product cost reduction were as we had estimated. The margin miss is attributable to inventory reserves, lower volume and regional mix. Seven percentage points of the gross margin miss was due to higher than forecasted inventory charges relating to lower than expected sales, higher Fab 4 output, and our outlook that price decline in the second half will be more aggressive than it was in Q2. The majority of the inventory reserve was related to lower of cost or market. The remaining three percentage points of our product gross margin miss is attributable to volume and mix. Because of the lower than expected sales volume in Q2, a smaller percentage of our sales came from the more recent lower cost production, and fixed cost of sales represented a larger percentage of our lower sales. In addition, our Q2 sales included a higher mix of international revenue where prices are more competitive and gross margins lower. We recognize that our gross margin results are not acceptable and that we must improve our capacity planning and inventory management. We are taking actions and are extremely focused on improving our performance on these fronts, but the overhang of the current excess capacity will take several quarters to correct.
Non-GAAP operating expenses of $224 million were down 2% sequentially and included a $4 million restructuring charge related to employment reductions which impacted approximately 130 employees. Our lower sales volume coupled with higher inventory charges resulted in a non-GAAP operating loss for the second quarter of ($57) million or (7%) of revenue. Other Income of $20.5 million was as forecasted, and the non-GAAP loss per share for Q2 is ($0.10) compared to a non-GAAP profit of $0.21 per share in Q1 and $0.30 per share in Q2 of last year.
On the balance sheet, cash and short and long-term investments decreased from Q1 by $484 million to $2.5 billion. Cash flow from operations was a negative $327 million, primarily reflecting the Q2 operating loss, growth in inventory, and increase in royalty receivables which are paid bi-annually. Net inventory increased sequentially by $101 million reflecting the strong output from our captive fabs, and at $796 million represents more than a quarters’ worth of inventory. Channel inventory ended Q2 between 8 and 9 weeks, somewhat higher than at the end of Q1 and the end of Q2 last year, again reflecting a slow-down in sell-through, particularly in the U.S. Capital fab investments for the quarter included a $97 million investment in the Fab 4 joint venture and the return of $24 million of investment from the 200 millimeter joint venture. We also guaranteed $282 million of incremental operating leases for purchase of equipment for Fab 4.
I’ll now turn to our outlook for Q3. Please note that non-GAAP to GAAP reconciliation tables are posted on our website for all applicable guidance. Given our high inventories, we currently plan for our ASP per megabyte to decline more in Q3 than it did in Q2. We believe this will help move the market to higher capacities, however we believe unit demand will continue to be weak given the macroeconomic environment. We are forecasting third quarter total revenue between $750 million and $850 million including license and royalty revenue approximately equal to the Q2 level. We now expect that growth in megabytes for the year will be below the low end of our previously guided range, or below 150%, and we expect that the decline in ASP per megabyte for the year will be above the high end of our previous guidance range – we are estimating approximately 60% decline.
Turning to gross margins, we forecast that Q3 is likely to include additional sizeable inventory reserves for lower-of-cost-or-market given our expectation that inventory will increase again in Q3 and given our current forecast of price decline beyond Q3. The level of our inventories also means that the cost benefit from new technology production will come later than originally expected. Based on these factors, we forecast third quarter non-GAAP product gross margin to be approximately zero. Of course in these market conditions, any forecast is subject to a high degree of variability, and when gross margins are thin, a small change in the rate of price decline or cost decline can cause a big difference in the required inventory reserves.
We forecast Q3 non-GAAP operating expenses to be between $225 and $235 million, with the sequential increase primarily in sales and marketing expense focused on driving additional international market penetration. We estimate Other Income for Q3 to be between $13 and $16 million.
In terms of fab investments in the second half of 2008, we expect cash fab investments to be approximately $150 million, which will be partially offset by the continuing return of cash from the 200 millimeter venture. Total capital investments for 2008, including fab and non-fab investments, are now expected to be less than predicted at our February analyst day due to improved fab equipment productivity and deferral of the next phase of the Fab 4 expansion.
Our decision to defer the next phase of investment in Fab 4 and to postpone the Fab 5 investment decision is expected to moderate our inventory growth and reduce our cash usage in 2009. We believe that this reduced capacity growth will allow us to maintain a strong balance sheet with ample liquidity even if the demand and pricing downturn were to extend another year.
I’d like to mention two other accounting topics. First, due to the recent decline in our stock price, we performed a goodwill impairment test during the second quarter which resulted in no impairment. If our stock price stays at current levels, which is below our net book value, we may need to perform a further impairment analysis in Q3, and it is possible that this could lead to a non-cash impairment charge. The second accounting topic is a reminder that in 2009 the accounting rules are changing for cash settled convertible debt securities such as our $1.15 billion convertible debt. We estimate that this accounting change will result in approximately $55 million in non-cash interest expense in 2009. We currently expect to exclude this non-cash accounting charge from our non-GAAP results.
We are clearly disappointed in our Q2 results and our forecast for Q3. Our near-term focus is on demand creation while we slow down further capacity expansion, both of which should help to achieve a better supply/demand balance in 2009.
We’ll now open the call for your questions.
Paul Coster, J.P. Morgan: Thank you. First of all, in delaying your Fab 4 investments, what is being forfeit in doing that?
Eli Harari: What do you mean, what is being forfeited? Paul? I don’t understand. What do you mean, what is being forfeited?
Paul Coster, J.P. Morgan: In making the decision to postpone investment in production capacity, you are presumably pushing out your cost curve. Is that a correct statement?
Eli Harari: No, we don’t think so. I think that the cost curve of Fab 4 is at a very, very attractive position right now and adding more capacity later on will not degrade that. You could say that the ROI on these fabs depends on when you invest, make the investments and how long technology can continue to run in a given fab, so over a five-year period, there may be some return on investment impact, but in terms of cost per megabyte, we don’t see that.
Paul Coster, J.P. Morgan: Okay, got it.
Sanjay Mehrotra: It’s a pretty large fab. It’s a pretty efficient fab at the current scale, so it has good cost efficiency already.
Paul Coster, J.P. Morgan: Okay, got it, thanks. And then in terms of the inventory composition, Judy, can you give us any details there?
Judy Bruner: I would tell you that the inventory on our balance sheet increased in all categories, meaning raw material with hand-finished goods and clearly is coming from the very productive output from Fab 4. But it increased in all categories.
Paul Coster, J.P. Morgan: Okay, last question. International/domestic split, can you give us the percentage?
Judy Bruner: Sure. In terms of the retail split, hold on, you know, why don’t I just come back to that once I get it there. Hold on. The [pause] well, I’ll give you overall revenue by geography was about 38% North America and 62% international.
Paul Coster, J.P. Morgan: Got it. Thanks very much.
Daniel Berenbaum, Cowen & Company: Thanks for taking my question. I wonder if we could go back to the discussion about shifting more of your buys to non-captive capacity. How long do you think it will take to get up to that 15% or 20% level, non-captive? And then, are those going to be long-term contracts with your existing JV partner, or is this something that you are just going to go and buy on the open market?
Sanjay Mehrotra: So with respect to the non-captive mix question that you are asking, our first priority at this point is to sell the inventory that we have got. And this will take us some period of time, a few quarters to work through this. So with respect to getting more non-captive, I don’t expect that to be happening for a few quarters. We’ll be getting to the targeted 85% to 15% to 20% mix in a one- to three-year timeframe, and the sources of that supply, just like we have leveraged those sources in the past, will continue to work with those sources and as the time comes for the need of non-captive supply, we will evaluate those sources and determine from whom we buy and how much.
Daniel Berenbaum, Cowen & Company: So then how do you think about what that does to your cost structure? I mean, obviously you’ve invested a lot of money into joint ventures with the rationale of having internal capacity at least in part because of your own manufacturing ability and the cost reduction you get from the internal fabs, so what do you think this does to gross margin as you look out over the next one to three years? And then more strategically, do you think about ever reducing your internal capacity to say 50% or even lower?
Sanjay Mehrotra: First of all with respect to the cost structure, we will of course continue to utilize our captive capacity first and that captive capacity is our lowest cost and it has cost leadership in the industry, we’ll be in excellent position with respect to the cost structure of the captive capacity.
Of course, as we purchase more non-captive mix, that non-captive will come at generally the market kind of prices and will, in the course of time, impact the gross margins. But we believe that the benefit of flexibility, the benefit of being able to manage inventory with non-captive supply is a tremendous plus and that’s why we want to target to get back to 85% and 15%, 20% kind of range in a one- to three-year timeframe.
Eli Harari: I think that right now, the situation basically demonstrates the fact that this kind of captive/non-captive model has its advantages. We have a situation where our captive supply is doing phenomenally well and is out-producing, and the demand, because of the economic conditions, has slowed down substantially and that has created a mismatch. If we were [pause] if we did not have sufficient captive capacity and were relying on non-captive for the incremental demand, we would be able to cut down that non-captive and not end up with the excess supply that we have today.
Daniel Berenbaum, Cowen & Company: And if I could just speak on, the last in on the follow-up; based on your commentary, it wasn’t clearly to me. You were going to slow down the investment in Fab 4. Is Toshiba still going ahead with the full investment? Can you give us some more color on what their plans there are?
Eli Harari: You know, we have flexibility in the agreements that we have with Toshiba for the two companies to manage their capacity requirements independent of each other. Of course, they [pause] you have to talk to them on their conditions. Their requirements are different than ours. I would say that the entire industry has this problem of excess supply. We are not unique in that and basically when they make their own decisions, they will, I’m sure, discuss those.
Edwin Mok, Needham & Company: Thanks. Just one question. What’s your Fab 4 capacity right now in terms of wafer starts and where do you expect it to end by the end of this year?
Sanjay Mehrotra: As we have said before, Fab 4 capacity by the end of this year, we expect it to be over 110,000 wafers per month.
Edwin Mok, Needham & Company: Okay, great. And then [pause]
Sanjay Mehrotra: And it continues to ramp toward that at this point.
Edwin Mok, Needham & Company: Great. Regarding your [pause] you have your in-house test and assembly facility as well, and you have said before that you target 30% of your production by the year-end to come out of that facility. Are you still targeting that and are you still planning on ramping or investing more money into that facility?
Sanjay Mehrotra: Yes, our assembly and test facility in Shanghai is actually ramping up very nicely and we expect to be achieving our target of 30% total unit production out of that factory late this year, in the fourth quarter. And that captive capacity definitely gives us again flexibility in terms of manufacturing product for us, as well as gives us cost benefit, particularly for multi-die stacked products, particularly for the mobile formats with multiple chips in them.
Edwin Mok, Needham & Company: Great, and one last question regarding your agreement with Toshiba on the 3D; I understand from their filing that there is a payment that Toshiba will pay you. I’m not asking you guys to give us the number there but at least can you give some color there or how long will that payment last and when do we expect to see that payment in the royalty stream?
Judy Bruner: Edwin, those payments will extend over multiple years. The exact timeframe is confidential in the agreement but the payment to SanDisk will be recognized on a straight-line basis over the period of the agreement, and will be recognized as license revenue.
Edwin Mok, Needham & Company: I see. And is that starting from the next quarter or when does that actually start?
Judy Bruner: You know, because of confidentiality clauses, I can’t give you the exact timeframes but we will take it into account in the future guidance that I provide.
Edwin Mok, Needham & Company: Great. Thank you.
Armand Pagosi, Credit Suisse: It’s Armand [Pagosi] in for John Pitzer today. A couple of questions about the bright spot, the SSDs. Could you elaborate a little further about what you think about the penetration, especially in the enterprise side of the business? And also a little bit of color on your positioning there and the expected increase to the leadership position?
Eli Harari: Yes, I think I commented on that last quarter. The enterprise space is moving very nicely. There’s a lot of activity in the enterprise to adopt, to embrace a solid state disk for reasons of power and performance. And that is a market where price per megabyte is not an issue at all, so we think that that [pause] however, the requirements in the enterprise of course are much more stringent than in the consumer space in notebook computers, for example. So that is a niche market that is developing very nicely. Of course, most of the volume in the market for solid state disk we expect to come more in the notebook and even in the desktop as well as the ultra low cost NetBook computers, and those are the ones that are very price sensitive and that is where MLC, which we believe is going to start making significant in-roads in 2009, is going to make the big difference.
Bob Gujavarty, Deutsche Bank: Great. Thanks for taking my questions. Just a little more color on the CapEx. I have the analyst day slide in front of me. So in 2008, you don’t see much impact to the $2 billion in total JV fab investments, but in 2009, that $2 billion will come down a little bit. Am I understanding that correctly?
Judy Bruner: That’s correct, Bob. Actually, if you look at the total capital investment that we predicted at analyst day, including fab and non-fab, it was about $2.4 billion for 2008 and about $3 billion for 2009, and I actually expect both of those numbers to come down. In 2008, the number will come down a little bit in the fab category and it will also come down in the non-fab CapEx category. I would expect that we would be closer to $2 billion than to $2.4 billion in 2008. And in 2009, it will come down in the fab category and also in the non-fab CapEx category.
Bob Gujavarty, Deutsche Bank: Any idea where. compared to $3 billion, any idea about the sizing of how much that could come down?
Judy Bruner: I would expect that it would come down at least $0.5 billion.
Bob Gujavarty, Deutsche Bank: Got it. Great, thanks. And just one more question; I know the Japan FX can impact your gross margins. Was that part of your gross margin hit as well or was that pretty neutral?
Judy Bruner: Good question. It actually was not a part of the miss from our guidance because when we gave our guidance for Q2, we of course already knew the approximate rate at which the wafers were purchased in Q1 or before that would be sold in Q2, but it is a reason for the decline in gross margin on a year-over-year basis and also on a quarter-over-quarter basis.
Bob Gujavarty, Deutsche Bank: Great. Thanks, guys.
Craig Ellis, Citigroup: Thanks for taking the questions. First of all, a number of presenters have addressed the inventory issue, which looks like it will be with us for some time. When will the on-hand inventory position be at its worst? Is it the third quarter or will it continue to build into the fourth quarter?
Judy Bruner: I definitely expect that it will increase from Q2 to Q3. I am hopeful that it will decrease in dollar value from the end of Q3 to the end of Q4, given the holiday season. But it is very possible that it would still be higher at the end of the year than it is right now.
Craig Ellis, Citigroup: Okay. And Judy, turning to the third quarter gross margin outlook, can you quantify how much of an inventory write-down is included in the 0% guidance that you gave us?
Judy Bruner: I would tell you that we expect it could be similar to what we included this quarter.
Craig Ellis, Citigroup: Okay, and then my last question is for Eli. Eli, given the rationalizing actions that you are taking on the supply side, when does that actually benefit industry? Because it seems like it would really probably be in the first or the second quarter of next year. Is that the right thinking?
Eli Harari: Yes, it’s not going to have much of an impact in the near-term but I think it could make a significant impact next year. Of course, it will also depend on the actions or lack of actions by some of our largest competitors. But Fab 3 and Fab 4 combined are very, very substantial megabyte generators, and I think that this at least our 50%, slowing that down will I believe have a qualitative impact on the demand/supply balance in 2009.
Craig Ellis, Citigroup: Thanks, everybody.
Kevin Vassily, Pacific Crest Securities: I want to go back to again the comments on Fab 4. Does your decision to delay investment right now at all change the equity stakes or ownership stakes you have in the fab? Is there anything in your agreements that would trigger a lower ownership and ultimately a lower capacity output that you need to take from that by doing this?
Eli Harari: No, I think we do have the flexibility in there and we do have, without going into confidential specifics, we do have the ability for a party that decides to delay over time to catch up.
Kevin Vassily, Pacific Crest Securities: Okay, so the long and the short of it is at some point when Fab 4 has reached its complete investment cycle, you’ll have reached your commitment that was agreed upon when the deal was signed? Is that the best way to think about it?
Eli Harari: I think at every step of the way, we are meeting our commitments within the flexibility terms that are given to each party. It’s always been assumed, these were agreements that were originated 10 years ago and have held us, both companies, in very good step that there will be times where the requirements of the companies diverge within a boundary. I mean, you can’t just go all over the place, and we are well within those boundaries.
Kevin Vassily,Pacific Crest Securities: Okay, great. Thank you very much.
Atif Malik, Morgan Stanley: Thanks for taking my question. When you look at your wafer starts or wafer outs in the Fab 4, where do you see an inflection in quarter-over-quarter wafer outs change? Is it in Q3 or is it in Q4?
Sanjay Mehrotra: That ramp-up in fab 4 will continue through Q4, so we will continue to see wafer output increases in Q4 as well, and then as Eli has mentioned, that we have decided to stay our wafer output increase, at least until April 2009. So from the end of Q4 through Q109, there will be no further increase in the wafer output. Remember, the technology transition will of course continue to be implemented in the fab, to give us [pause] to continue to produce the lowest cost per megabyte wafers.
Eli Harari: But that also applies to Fab 3, which is about 150,000 wafers a month, 300-millimeter. And that too is undergoing conversation from 56nm to 43nm.
Atif Malik, Morgan Stanley: Okay, and can you provide some comments on the competition in the SSD area? You know, where is your relative cost structure and pricing with respect to guys like Intel and Micron and other guys?
Eli Harari: This is really a very, very young market. At this stage, revenues for all suppliers are relatively miniscule and we are really into first generation products, most of which are based on SLC and therefore really not meaningfully representative of the potential of Flash memory.
Our cost structure, we believe SSD more than probably any other product, will drive capacity: 64GB 128GB into 256GB (?) and therefore we’ll be most closely tracking the cost per gigabyte at the flash component level and there we think we have certainly the lowest cost. Certainly we’re in the league of top notch, lowest cost producers. And that we expect to remain certainly in 2009 and as we introduce 32nm technology, we expect to be able to maintain a highly competitive cost structure into 2010 and beyond.
Atif Malik, Morgan Stanley: Okay, thanks.
Vijay Rakesh,ThinkPanmure: Just a couple of questions. On this 43nm transition, what percent of your output was at 43 in Q2 and where do you see it in Q3?
Sanjay Mehrotra: So by end of the year, we have said we’ll get two-thirds of our bit production output during the quarter, during the fourth quarter in 43nm. We just started production output in second quarter. Our production output in second quarter in terms of bit output from the fab was approximately 15% and this will continue to ramp toward the two-third that I mentioned in the fourth quarter. So obviously it will increase in third quarter.
Eli Harari: I would add to that that the cost per megabyte on the 43nm two bit per cell is already very competitive with the best cost that we can get from 56nm.
Vijay Rakesh, ThinkPanmure: Okay, and on the inventory, you mentioned that retail, it was eight to nine weeks. How does it compare to what you normally see at this time of the year?
Judy Bruner: I would say that last quarter and the year-ago quarter, it was more like about seven weeks. Eight to nine weeks is not unduly heavy. it’s just a little bit heavier than it was last quarter and a year ago.
Eli Harari: I think it represents about maybe two weeks of unsold June, typical retail sales that we were expecting that did not happen in the U.S.
Vijay Rakesh, ThinkPanmure: Got it. Okay, and the last question, on SSDs, do you have any tier one OEMs as [pause] would you [pause] do have any tier one OEM wins here for the second half or for the first half of next year?
Eli Harari: We are working with a number of large OEMs, as well as with a whole tier of OEM manufacturers and we really cannot comment on specifics.
Vijay Rakesh, ThinkPanmure: Okay, thanks.
Eli Harari: But there’s like [pause] there’s 10 top tier and then there’s a whole number of second-tier players, and then of course there’s enterprise, and we are probably doing one thing or another to just about all of them.
Vijay Rakesh, ThinkPanmure: Okay, thanks.
Daniel Amir, Lazard Capital Markets: Thanks a lot. Thank you for taking my question. Eli, just if we look kind of in the next 12 to 15 months, what needs to happen in your opinion to get out of the downturn here in the NAND industry in terms of demand/supply compared to the steps that have been taken already by some of your competitors on the supply side, and kind of how you view the demand side here in the next 12 to 15 months?
Eli Harari: Good question. The current environment is really untenable. I said really three years ago that it is unsustainable for any manufacturer to have three [pause] you know, to have a 60% price reductions every year and we’ve had it now this is the third year. So cost reductions just can’t keep up with that kind of rate of decline, even with three and even with four bits per cell and if you are leading edge and playing with very highly efficient fabs.
So this thing really will slow down either because people decided this is not a good business to build new fabs for, or because demand really picks up and the demand we believe will definitely pick up. These are [pause] you know, every market projection shows that the industry demand for gigabyte is going to increase by 20-fold or so. Not 20% [pause] 2000% over the next three to four years, so there is no question that there is growth in demand ahead. The question is, is it worthwhile investing in the supply side? And the answer has to be that this needs to be a profitable business in order to get the kind of return on investment, and that means that the demand/supply balance needs to come into much better balance and that the cost reductions and price reductions must come into track with each other.
From our point of view, we think that the key of what we need to do is to drive the attach rate in handsets because that’s the biggest market out there and we need to drive the capacity per card in the handsets. We think that that’s going to happen now. I think the iPhone 3G is a very, very good step for the industry and will drive cell-phone manufacturers on the high-end to adopt 8- and 16GB Flash storage.
So I think that there is [pause] there’s a lot of very bright spots on the demand side but we are still struggling through excess supply because no one really foresaw the economic slowdown, the thing in the economy that really [pause] you know, the sub-prime lending and the high price of oil.
Daniel Amir, Lazard Capital Markets: Okay, thanks. And one other question on the OpEx side; SanDisk has taken a number of steps in the restructuring side, so the question is have you completed your restructuring? And second, I mean, considering that for the next few quarters, we’re going to see a challenging pressure on your business model, are you expecting further restructuring here beyond that what you’ve already been talking about last quarter?
Eli Harari: We’re very happy with where we are today. We think we have a very lean and very, very focused organization. We believe that the best way for us is to grow our way out of this current situation and that means introducing new products, innovation, cost reduction. Most of our cost advantage, if you will, is in the manufacturing and technology, not in headcount. Our R&D is very, very efficient and frankly, we are developing in parallel two major technologies. NAND Flash, we are right there at the leading edge and 3D read/write is, you know, we’re investing more and more and accelerating that investment.
Daniel Amir, Lazard Capital Markets: Okay. Thanks a lot.
Doug Freedman, American Technology Research: I was wondering if you guys could talk a little bit about the controller technology required to get the SSD adoption jump-started, and where you stand with your efforts there.
Eli Harari: Controller technology depends on the application. In very low-end, ultra low-cost PCs, existing controllers can get the job done for 8-, 16-, and 32GB storage because these are relatively unsophisticated and demanding requirements. As soon as you get into Vista applications in notebook and desktop, you start running into very demanding applications because Vista is not optimized for flash memory solid state disk. And the next generation controllers need to basically compensate for Vista shortfalls. In the enterprise, you have proprietary operating systems, proprietary architectures, and therefore manufacturers, enterprise manufacturers, server manufacturers can take advantage of flash with existing controllers that have been developed mostly for disk drives.
Doug Freedman, American Technology Research: Okay, and I guess my question really is where do you guys stand and how do you feel about your internal controller efforts and whether your SSDs are going to be built with an internal technology or a technology that you are going to incorporate from external sources?
Eli Harari: We have very good internal controller technology, as you know. We are at this stage 100% internal controller for just about everything we make, and that’s been a 20 year tradition in here and we have a huge number of engineers that really understand the Flash issues and firmware issues and so on. That said, I’d say that we are now behind because we did not fully understand, frankly, the limitations in the Vista environment. When we acquired msystems, msystems had very, very good industrial grade solid state disk and this is our first generation products. Unfortunately, their performance in the Vista environment falls short of what the market really needs and that is why we need to develop the next generation, which we’ll start sampling end of this year, early next year.
Doug Freedman, American Technology Research: And do you believe the products that will be sampling, those will be the ones that will drive adoption? Or are we still looking at another generation required before we get into the high volume?
Eli Harari: I think it will start driving adoption but this is like everything that we’ve done. Eevery generation gets better. You learn from the previous generation, you learn from customers and this is going to be an ongoing thing. It’s very similar to our Ultra Extreme, where in the early years of imaging our cards were not the fastest in the market and we undertook a major effort and through two or three generations of controllers, we developed the industry’s fastest and have stayed ahead. So these things don’t happen overnight. You need to be patient. You need to invest and we are up to the task.
Doug Freedman, American Technology Research: All right, and then you made a comment about trying to increase the penetration in the mobile market. What is the early [pause] you know, there’s more and more slots out there this year, clearly. The density of that adoption, the retail card capacity crossing over the 2 gigabyte numbers, clearly a positive but in particular to mobile, what are you seeing in the mobile market for the Micro SD as far as density? And what attach rates are you presently seeing?
Sanjay Mehrotra: With respect to the attach rates, at the end of Q1, we have said that attach rates on the retail side of the business are in the high 20s and at this point, those attach rates are about the same.
With respect to the trends in the mobile business, we are absolutely seeing more and more interest in higher average capacities, whether it’s on the OEM side or on the retail side. The iPhone 3G, the 8 gigabyte and 16 gigabyte is actually only pushing the rest of the MNOs and the handset manufacturers in terms of offering higher capacity solutions, and that’s being done [pause] it’s being looked at. For the OEM side in terms of bundling, we expect increasing higher capacity bundling in the future. Also from the mobile network operators, there is increasing utilization of their network related to content and that again bodes well for higher capacity usage for our cards. And in that regard, we recently actually introduced Ultra line of mobile cards, high-speed line of mobile cards which will actually help speed up the download of the content.
So on the retail side, definitely continued to see nice increases in average capacity and the iPhone 3G effect is also encouraging the OEMs to look at higher capacity bundling, such as in fact even 8 gigabyte bundling, and we expect that trend to increase over time.
Doug Freedman, American Technology Research: All right, great. Thanks for taking my questions.
Craig Ellis, Citigroup: Thanks for taking the follow-up. Eli, we clearly have a very challenged U.S. economy, and in other parts of the world as well. But from the data that you see, what gives you confidence that that’s all there is at play and we’re not simply seeing market saturation in product like USB drives and MP3 players?
Eli Harari: The specifics for our industry, I have no doubt about it and I have [pause] this is not just my optimism. I think the data will support it. The markets have near-term not shown the kind of elasticity that we have seen in the past. I believe that with all this cumulative pricing that we will see a more [pause] certainly USB Flash drives to 4, 8, 16 and even 32GB in the next 12 months. I think that GPS is going to move to three dimension mapping and that will definitely drive the 2GB to the 8GB phone. We introduced the video card at 4 and 8GB. We are selling more 8GB video cards than 4GB cards, which is very, very unusual. We’ve never seen that and that’s because people would like to have a two-hour recording rather than a one-hour recording. And of course, SSD is there.
So I think we have our work cut out for us but we know very, very well what we need to do. There’s really no big mystery in here. We’ve got to [pause] currently, we’ve been through a worse down cycle than this one. 2001 was [pause] you know, this one is not an easy one but 2001 by comparison was much worse and we will get through it. We are very, very focused. And I think our industry is [pause] we’re not unique. We’re in a very challenging time but this team really is taking the decisive action that will get us to renewed profitable growth and I am very, very confident that we are heading in the right direction and I think we really [pause] I think we’ve gone [pause] Lori is pointing I need to finish.
Kate Kotlarsky, Goldman Sachs: This is Kate Kotlarsky for Jim Covello. A quick question regarding the status of the royalty agreement with your primary licensee; just curious whether there’s been any progress there over the last couple of months, or whether things are pretty much status quo.
Eli Harari: Status quo.
Kate Kotlarsky, Goldman Sachs: Okay. And can I also ask Judy, at the analyst meeting, you laid out some longer term targets, I believe out to 2010. Does the push-out of Fab 4 and Fab 5 impact those?
Judy Bruner: Are you referring to the P&L financial model?
Kate Kotlarsky, Goldman Sachs: Yes.
Judy Bruner: You know, clearly the P&L financial model that I shared at analyst day, which was an operating margin of 13% to 16%, is very dependent on the product gross margin and achieving that type of operating margin is really dependent on getting our product gross margin back to a more normal level, back to a more acceptable level, which I suggested at analyst day was a 24% to 28% product gross margin.
So a number of things need to happen to get us back there but getting the industry and ourselves in a better supply/demand balance and therefore reducing the rate of price decline is one of the key factors, and so yes, in that regard, slowing down or pausing for the moment on the rate of growth of capacity expansion should help us move back toward that kind of model. But at the moment, that model is not there in the foreseeable future and is really dependent on getting our product gross margins back in place.
Eli Harari: I’d just like to wrap it up here. I want to say that we are responding to this current environment. It came rather quickly, not forecasted by anybody, really, at the beginning of the year. And what we are doing is basically building for the future but protecting the present and really making sure that our balance sheet remains strong and making the right decisions for, let’s say, for the next 12 to 18 months.
Thank you very much. We very much appreciate you listening and your support and we’ll talk to you in a quarter. Thank you.