Q1 2008 Semiconductor Manufacturing International Corp. (SMIC) Earnings Conference Call - Final

Corporate Participants

* Theresa Teng SMIC Corp. - Head, Finance and IR * Morning Wu SMIC Corp. - Acting CFO * Dr. Richard Chang SMIC Corp. - CEO * Gareth Kung SMIC Corp. - Director, Treasury Division

Conference Call Participants

* Daniel Heyler Merrill Lynch - Analyst * Steven Pelayo HSBC - Analyst * Pranab Sarmah Daiwa Securities - Analyst * Bill Lu Morgan Stanley - Analyst * Randy Abrams Credit Suisse - Analyst * William Tong UBS - Analyst * Eric Chen BNP Paribas - Analyst * Donald Lu Goldman Sachs - Analyst

Presentation

OPERATOR: Welcome to the Semiconductor Manufacturing International Corporation's first quarter 2008 webcast conference call. Today's conference call is chaired by Dr. Richard Chang, Chief Executive Officer and President and Ms. Morning Wu, Acting Chief Financial Officer and Ms. Theresa Teng, Head of Finance and Investor Relations.

Today's webcast conference call will be simultaneously streamed through the Internet at SMIC's website at www.smics.com. Please be advised that your dial-ins are in listen-only mode. However, at the conclusion of the management presentation we will be having a question and answer session, upon which you will receive further instructions as to how to participate.

The earnings Press Release is available for download at www.smics.com.

Without further ado, I would like to introduce you to Ms. Theresa Teng, Head of Finance and Investor Relations with a cautionary statement.

THERESA TENG, HEAD, FINANCE AND IR, SMIC CORP.: Good morning, everyone. Welcome to the SMIC first quarter 2008 earnings conference call. Joining me on the call today are Dr. Richard Chang, Chief Executive Officer and President, and Ms. Morning Wu, Acting Chief Financial Officer.

Our call will be approximately 60 minutes in length. The earnings Press Release and presentation are available for download at www.smics.com.

Please note the following Safe Harbor statement. SMIC's statements of its current expectations are forward-looking statements, subject to significant risks and uncertainties. The actual results may differ materially from those contained in the forward-looking statements.

Information as to those factors that could cause actual results to vary can be found in SMIC's Form 20-F, filed with United States Securities and Exchange Commission on June 29, 2007.

For today's agenda, Morning will highlight our first quarter 2008 financial results and second quarter 2008 guidance, with a summary of the cash flow statement and balance sheet in the appendix for your ease of reference. Following that, Richard will provide an update on our business.

I will now turn the call over to Morning.

MORNING WU, ACTING CFO, SMIC CORP.: Thank you, Theresa. I would like to highlight the following items which are all stated in U.S. dollars. Non-DRAM revenue increased by 5.5% quarter on quarter to $318.6m from fourth quarter '07, and increased by 25.6% year on year from first quarter '07.

Total revenue, however, declined to $362.4m in first quarter '08, down 8.3% quarter on quarter from fourth quarter '07 and down 6.7% year on year from first quarter '07, due to lower DRAM shipment quantities.

DRAM as a proportion of total revenue fell to 12.1% in first quarter '08, from 23.6% in fourth quarter '07.

Logic sales from 0.13 micron full-flow and the 90 nanometer technology nodes have increased by 33.5% in first quarter '08, quarter on quarter.

Gross margin was negative 9% in first quarter '08, as compared to positive 8.9% in fourth quarter '07, primarily due to losses from the commodity DRAM business, including a $44.5m additional loss provision taken against the remaining DRAM inventory.

The Company recorded a net loss of $119.1m in first quarter '08, including the reversal of $20.5m deferred income tax benefits recorded in fourth quarter '07, as required under U.S. GAAP, resulting in an adjusted net loss of $19m in 2007.

Fully diluted EPS was negative $0.3205 per ADS in the first quarter of 2008. CapEx of $136m was reported for first quarter '08, which is lower than our guidance, due to slower equipment move-in schedules. Total planned CapEx for the year remains at approximately $700m and will be adjusted based on market conditions.

During the first quarter of 2008, the Company reached an agreement with our customers to exit the commodity DRAM business. The Company considered this an indicator of impairment in regard to the long-lived leased assets of the Company's Beijing fab in accordance with SFAS 144.

As of today, the Company has engaged an external valuer and in the process of evaluating whether or not such assets have been impaired. Any impairment loss, if so determined, would result in an additional non-cash charge to the Company's net income for the first quarter of 2008.

Summaries of our income statement, balance sheet and cash flow statement are available in the appendix.

Our guidance for the second quarter of 2008 is as follows. Due to the stopped delivery of DRAM products, total revenue is expected to decrease 3% to 6% from first quarter '08, while non-DRAM revenue is expected to grow 3% to 6% from first quarter '08.

Operating expenses as a percentage of revenue is expected to be around 20%. Capital expenditure will be approximately $160m to $200m.

I will now turn the call over to Richard for the business update and review.

DR. RICHARD CHANG, CEO, SMIC CORP.: Thank you, Morning. There are a number of items that I would like to touch upon, many of which will have great impact on our business in 2008 and beyond.

As part of our ongoing strategy to enhance our foundry services to our global customers, we are focusing on several key elements; continuing to develop leading-edge technologies, advancing our high-margin Logic business, enhancing our turnkey services, capturing industry migration to China, expanding partnership with local fabless companies in China and terminating our business exposure to DRAM.

We reported a quarterly loss of $119.1m, which includes an additional loss provision for DRAM inventory of about $44.5m, as well as the reversal of $20.5m deferred income tax benefits recorded in fourth quarter '07.

During the first quarter, management reached an agreement with our customers to exit the commodity DRAM business. This reduction of DRAM production and conversion of DRAM capacity into Logic, which will continue throughout 2008, remains key to our strategy.

As a result, SMIC made a concerted effort to reduce its DRAM foundry service by 53% from the previous quarter and 67.5% from the first quarter of 2007. At the same time, we've increased Logic shipments 6.2% quarter on quarter and our 90 nanometer Logic shipments surged 136.8% over the fourth quarter. And as DRAM as a portion of our total revenue fell to 12.1% compared to 23.6% in the first quarter of 2007 our overall Logic revenues gained 6.5% quarter over quarter and 25.6% year over year.

By the end of this year, we only plan to produce a small amount of specialty DRAM, which should account for about 1% to 2% of our business.

The shift from DRAM contributed greatly to the quarterly loss, a temporary but a necessary consequence that will enable SMIC to focus on Logic production which, we believe, will accelerate our path to profitability.

I would like to take this opportunity to thank our DRAM partners for supporting us to stop our commodity DRAM services, starting in Q1. And after Q2 2008, we'll completely stop any new loading.

Our Tianjin and Shanghai fabrication facilities, which have been manufacturing Logic products for some time, are profitable. We expect no different from our Beijing fab, once we've completed its conversion to Logic, which is making strong progress.

We're already seeing positive signs, as our Beijing fab increased Logic shipments by 145.2% quarter on quarter. With our current Logic capacity fully utilized, we are working hard to increase our Logic capacity. And, in 2008, we expect shipments for 0.13 micron 12-inch products to increase at a triple-digital growth rate and the 0.13 micron 8-inch and 90 nanometer 12-inch product shipments to grow by more than 50%.

We will continue to develop our technology in radio frequency, mixed-signal, high-speed, low-leakage, system-on-chip, and embedded Logic products. We forecast persistently strong demand in advanced technology nodes through the remainder of 2008. as we witness tremendous market demand for the devices that consume Logic ICs, such as mobile base band, multimedia processors, PDA, GPS, Flash controller ICs, power management ICs, MP3, MP4, digital TV video processors and mobile TV, and so forth.

As the world's largest market for integrated circuits, China has experienced exceptionally strong market growth since the beginning of the year. Our China revenue increased by 22.6% since the fourth quarter of 2007, and we enjoyed the addition of 15 new domestic customers this quarter. We also reported a 17.1% quarter-on-quarter increase in new products tape outs

We have a number of Chinese fabless customers for whom we are manufacturing at the 90 nanometer technology node, one of which has already attained end-user qualification for its first 90 nanometer product and entered commercial production.

Our focus on China will remain strong as it matures as the industry's driver and, increasingly so, at advanced technology nodes.

Not only have we experienced considerable growth in our China sales, but we also enjoyed a significant boost in our North American sales. Despite the challenging economic situation in the U.S., our sales in North America grew by 10% quarter over quarter, and increased as a portion of our total revenues to 53.6% in the first quarter of 2008, compared to 44.6% in the fourth quarter of 2007.

We are also pleased with the progress we have made in regards to our 45 nanometer licensing agreement with IBM. Currently, there are 10 top-tier fabless and IDM companies expressing interest to work with us. As a result of our milestone agreement, we have also increased our customer base for the 65 nanometer and the 90 nanometer Logic nodes, as customers are confident with SMIC's future technology roadmap.

In addition, we have garnered the interest of customers in China in working with SMIC's 45 nanometer technology solution. We plan to enter process qualification in 2009.

I also have updates on two of SMIC's business ventures. First, TSES SMIC's Joint Venture with Toppan to produce color image sensors became profitable in March, and we expect it to remain so.

I am also pleased to announce that our solar business, SMIC Energy Tech, continues to be successful. The dedicated R&D work of SMIC's engineers has allowed us to achieve energy conversion efficiencies that are among the best in the industry.

Over the coming quarters, we will continue to improve upon the quality and efficiency of SMIC Energy Tech 's products and devote efforts to take the unit to even greater heights.

Our commitment to enhancing shareholder value in our Company remains absolute. To that end, we will continue to execute our plans. which we believe will accelerate growth, serve our customers and boost our bottom line. We are very confident in our strategy and optimistic about upcoming year.

I will now hand the call back to Theresa, who will moderate the Q&A session of this call. Theresa?

THERESA TENG: Thank you, Richard. I would now like to open up the call for Q&A. Please limit your questions. Operator?

Questions and Answers

OPERATOR: (OPERATOR INSTRUCTIONS). Your first question comes from the line of Dan Heyler with Merrill Lynch. Please proceed.

DANIEL HEYLER, ANALYST, MERRILL LYNCH: Good morning, Richard.

DR. RICHARD CHANG: Good morning, Dan.

DANIEL HEYLER: I had a question on the profitability roadmap and revenue contributions from the other core businesses. So it seems as though you have a lot of investments that are taking place here, and the Joint Venture strategy and exiting the DRAM strategy, as well. When would you think you would be at a net break-even level across the businesses and in a sustainable profitability structure?

DR. RICHARD CHANG: Dan, thank you for the good question. We stopped DRAM business on March 31, and we immediately start the conversion of all the DRAM equipment into Logic. So far, based on our information, about 75% of the equipment can be directly used for Logic production. 25% of DRAM equipment needs modification. This takes some times. We estimate it may take us two quarters to completely convert all the DRAM capacity into Logic production.

For Q2, Q3, still will be a little bit tougher for us. But, from our own estimate, starting in Q4 this year, we should be able to be profitable. As I just reported, that our Shanghai fab has been profitable since 2004. Shanghai basically is making a lot of Logic, small amount of DRAM, but now we also only do a small -- very small amount of specialty DRAM in Shanghai. That makes us very profitable.

Tianjin also stopped the DRAM last year. And this year, Tianjin 100% Logic, and we observe Tianjin fab also become profitable. That's why I reported to you that we see no difference that, once Beijing is converted into Logic, it will be profitable also. That will be completed -- the conversion will be completed in Q4. So, starting Q4, I expect the entire Company will be profitable.

And from then, and on, we believe we are going to be profitable.

DANIEL HEYLER: Great. Thank you. And one quick follow up. On the Beijing fab, maybe you could walk us through the products and the technology nodes that you'll be converting to and whether or not that's planning for -- the IBM technology will also be deployed at the Beijing fab.

DR. RICHARD CHANG: Yes. Currently, Beijing, the Logic part, we are manufacturing 90 nanometers That is the main part. We also do Logic -- we are serving customer with 0.13 micron Logic. And we are also in the process of being qualified in second quarter for the 65 nanometer. We already have passed the preliminary qualification by our customers. And in Q2 we are entering the final qualification stage.

So starting the second half this year, Beijing will also start 65 nanometer Logic manufacturing. 45 nanometer IBM technology, we are doing the installation, testing, qualification process together with IBM. IBM did a wonderful job to transfer this technology to us. We expect, starting 2009, 45 nanometer will enter mass production mode as well.

DANIEL HEYLER: Thank you, Richard.

DR. RICHARD CHANG: Thank you, Dan.

OPERATOR: Your next question comes from the line of Steven Pelayo with HSBC. Please proceed.

STEVEN PELAYO, ANALYST, HSBC: Yes. I wonder if you can comment a bit on 90 nanometer ASPs. Your shipments were up 136%. Your revenue was up about 100% or so. So is there some significant 90 nanometer ASP pressure?

And then my second question is also on ASPs. Could you just help us understand the relative premium, I guess, of the 90 nanometer DRAM wafer versus 90 nanometer Logic. Is a 90 nanometer DRAM wafer comparable in pricing to 130 nanometer Logic? I'm trying to understand how the ASPs are going to move as you move out of DRAM.

DR. RICHARD CHANG: 90 nanometer ASP Logic is holding fairly well because not too many foundries provide these kind of services; only a few of the top ones. So we -- competition is less and customer demand is high, so ASP is holding very well.

As far as 90 nanometer DRAM versus 90 nanometer Logic, there's a big, big difference. 90 nanometer DRAM is at a big loss, but 90 nanometer ASP is positive. So, again, a big difference. Sorry, I cannot tell you the exact number, but you can do the calculation. Assuming that a 90 nanometer DRAM, the ASP is, say, $1 per chip and, therefore, the 90 nanometer ASP for Logic usually may be anywhere between $5 to about $10 range.

STEVEN PELAYO: Okay. When I look at the current quarter, you just had memory down about 50% quarter over quarter. 90 nanometer doubled quarter over quarter. Logic up 6%. But I think the blended average ASPs were relatively flat. So it seems kind of odd to me that such a mix improvement didn't result in a significant ASP improvement.

DR. RICHARD CHANG: It's because the DRAM prices dropped -- continued to drop from Q4 last year to Q1. And Logic price is holding, so you see that. But volume -- our volume of 90 nanometer increased. So when you do a calculation, you can find out. Also, I just mentioned to you, that the ASP for DRAM versus Logic you can calculate. It's a big difference.

STEVEN PELAYO: Okay. And now that you'll be exiting the DRAM business, you should have a lot more stability in ASPs. Do you think that we can start talking gross margin guidance, as well?

DR. RICHARD CHANG: We notice that, in this industry, it seems like the top foundry people did not give ASP guidance anymore. But they will report -- we all will report the ASP quarter on quarter to everybody. I think it is fair to our customers not to provide that ASP guidance. But you can (multiple speakers).

STEVEN PELAYO: And just -- I'm sorry. I was actually looking for just gross margin guidance overall for SMIC for the second quarter and beyond, as you get more out of DRAM and so some of the volatility of your margins should decrease.

DR. RICHARD CHANG: The overall margin will increase.

MORNING WU: Yes.

STEVEN PELAYO: Okay, thank you.

DR. RICHARD CHANG: Thank you.

OPERATOR: Your next question comes from the line of Pranab Sarmah with Daiwa Securities. Please proceed.

PRANAB SARMAH, ANALYST, DAIWA SECURITIES: Yes. Thanks for the questions. Richard, I have a couple of questions. First one, could you let me know, are you holding any more DRAM inventory on your inventory at this point? And, if so, is there any more write-off risk on this DRAM inventory?

DR. RICHARD CHANG: In Q1, we basically wrote off all of the inventory. And starting Q2 we did not start any new commodity DRAM at all. So we do not expect any further inventory write-off because already written off this time.

PRANAB SARMAH: Okay. On the Solar business, could you elaborate a bit what type of capacity you have currently and what type of profitability level you are going to? And, obviously, (inaudible) is there in their business, whether you are able to increase your production ramp-up there.

DR. RICHARD CHANG: The solar energy is very interesting. SMIC has been focusing on technology development and not really focusing on the mass -- the -- to increase the volume yet. And because we use the CMOS technology to improve the solar cell efficiency, we have reached wonderful performance.

Our customers tell us that our energy conversion efficiency is among the best in the industry. However, we very cautiously working with our partners whom -- from whom we purchase the raw material from, which is a poly-silicon material. And currently this operation is still in very small quantities. Even in this small quantity, we become profitable. because efficiency is very high.

So I believe we will continue to acquire more feedstock, then we will expand the volume.

By the way, the first gentleman, Dan, asked me a question about us getting into new business. Actually, it's not a new business. Like TSES CMOS image sensor, color filter and micro lense business we already started a little over two years ago. But it took us about -- together with Toppan about -- a little over two years, we made this business profitable, so we started to harvest what we planted two, three years ago.

Solar energy, is same thing. We started this project almost three years ago. And, starting last year, we are basically profitable. And this year remains to be profitable. We are planning with our partner, as soon as we can receive more of the feedstock, that we are going to expand the service. The market is not a problem at all. Profitability is not a problem. It's -- the challenge is in raw material, which is the thing we've started to work on now.

As I mentioned to you previously, we focused on technology development. And the efficiency -- the energy conversion efficiency is excellent. That was our focus and we did a wonderful job. So maybe later part of this year we will report to you more about the solar energy project and that business.

PRANAB SARMAH: Okay. Since your share price is significantly below book value, does it violate any of the loan covenants you have with your bankers?

DR. RICHARD CHANG: So far, no. And we still have a lot of cash on hand. So we did not see this kind of problem. We -- yes, I really feel that our stock is -- it's a personal feeling. It's really way undervalued.

PRANAB SARMAH: Thank you.

DR. RICHARD CHANG: Thank you.

OPERATOR: Your next question comes from the line of Bill Lu with Morgan Stanley. Please proceed.

BILL LU, ANALYST, MORGAN STANLEY: Hello, good morning. I'm hoping we can go back to the profitability question because, if you look at Q1 -- even if you back out the write-offs from DRAM, margins and profitability were still lower than it was in previous quarters at similar revenue levels. Basically, you had said that Logic margin was better than the [corporate] average. I was just wondering if you can help me reconcile that.

DR. RICHARD CHANG: The profit margin, I'd like to ask Gareth or Morning to answer this question, please.

GARETH KUNG, DIRECTOR, TREASURY DIVISION, SMIC CORP.: I think when you back out -- I think you only -- which number did you back out -- I think you only back out the $44m, right? Is that correct?

BILL LU: Yes.

GARETH KUNG: That is only the LCM for the remaining inventories. And we have also the realized loss from DRAM shipment. So what happened is that we did not disclose this number but, actually, our -- if you back out the DRAM related losses, actually, the gross margin is much higher.

BILL LU: Okay. Is there any way you can help me a little bit more on just the difference in margins between Logic and DRAM right now?

GARETH KUNG: As Dr. Chang mentioned earlier, the Logic prices have been holding very well in the last few quarters and DRAM prices had declined substantially. So you can see there's a big difference in terms of the gross margins. And, obviously, Logic is profitable for us and DRAM is very negative for us.

DR. RICHARD CHANG: There's a very simple way you can calculate that. The ASP per wafer, Logic compared to DRAM, Logic is many times higher; many times, not only percentage-wise.

BILL LU: Okay, great, thank you.

My second question is you are converting a lot of your new capacity, or I guess all of it, to Logic, but that's going to take some time. Can you just help me with your overall capacity for the next couple of quarters, because I'm having a hard time trying to figure that out without any guidance?

DR. RICHARD CHANG: Okay. First quarter, our wafer capacity for Logic is about -- for Logic is about 167,000 per month, but our conversion is ongoing. By the end of this year, fourth quarter -- this is including all the fabs, we expect it will be increased. It's going to increase a lot.

But, in other words, the DRAM capacity -- if we don't have to buy any new equipment, direct conversion, we can do almost 50% of the DRAM capacity without any modification to convert to Logic. So you can easily calculate how much capacity we'll be able to gain.

However, if we do some bottleneck management, especially in the back end which is the metal interconnection portion, we can reach much higher -- maybe a conversion better than 70%, maybe 75%. But we are working on this now.

BILL LU: Thanks, that's helpful.

DR. RICHARD CHANG: Thank you.

OPERATOR: Your next question comes from the line of Randy Abrams with Credit Suisse. Please proceed.

RANDY ABRAMS, ANALYST, CREDIT SUISSE: Good morning. I wonder if you can provide an update on Chengdu and Wuhan and then also the Shenzhen fab ramps. And I think last quarter you mentioned by year end about 210,000 wafer capacity, and then 267,000 including the managed fabs. I just want to see, with the conversion now, if you can give an update on those year-end capacity numbers?

DR. RICHARD CHANG: Thank you, and a good question also.

Our Chengdu fab already started production -- pilot production two quarters before. And now the capacity is close to 8,000 wafers per month, but we manage this fab. Also the equipment the Chengdu government purchased from Japan arrived in Q1. It takes maybe three months to install, testing and qualification.

So, basically, starting in second half of this year Chengdu's capacity will increase. The goal is about to maybe increase to 25,000 wafers by the end of this year -- 25,000 wafers per month.

Wuhan project is good news. Wuhan already started pilot production since a week ago. It started with Flash and there are quite a few different kinds of Flash products already planned or already loaded in Wuhan fab. So Wuhan fab is going to ramp-up according to the qualification process.

Because the 12-inch fab takes much longer time to be qualified we expect that by the end of this year Wuhan fab income will not be significant. But, starting in 2009, their contribution will be very significant because eventually, with the current plan, Wuhan can reach capacity of more than 15,000 wafers per month by some time in 2009. It depends on how fast we can qualify the product and we can pull in the equipment.

Shenzhen project, we already started the company establishing, registration, many details. Design already done and to be approved by government then we can start. Our plan is still to start ground breaking in June this year. And it takes maybe 15 months, up to 18 months, to get it to start to work on the pilot production. So we expect that the second half of 2009 Shenzhen fab will start to contribute.

By the way, Shenzhen fab is very unique because it is located in south part of China where many semiconductor users and fabless companies are located. And many customers already booked 100% of the capacity of Shenzhen. We told them that it will be 15 months to 18 months later but they already want to book the capacity for the fab. So Shenzhen's location is very, very interesting.

Chengdu, also because they are located in West China, so the demand also is very strong. Wuhan, again, most of the customers are from overseas. As far as Chengdu and Shenzhen, most of the customers are from mainland China. Chengdu is for specialty products. Thank you.

OPERATOR: (OPERATOR INSTRUCTIONS). The next question comes as a follow up from Dan Heyler with Merrill Lynch. Please proceed.

DANIEL HEYLER: Hello, thanks for that. I just had one question on the IBM agreement. I know that it's a confidential agreement on the process technology, but are you -- Richard, is there any special treatment that SMIC is being treated, given your profitability level is pretty low and there are typically licensing fees associated -- fairly significant ones for acquiring the IBM technology? Is this contingent on you successfully ramping and getting partners? Or do you basically have to pay an upfront fee regardless of your ramp schedule?

DR. RICHARD CHANG: We cannot disclose the details but, as the common rule, the technology transfer fee we pay to our partners consists of two parts. One is the upfront fee. Usually it's the fee to cover the technology transfer detail meetings, run wafers, test wafers, testing programs, many -- and also the onsite support from IBM people, from our technology partners. That's a part.

Another portion is associated with the royalty. This means after we start production to serve customers, then we start to pay that portion. So this is always like that.

However, I really feel that the deal we worked out with IBM is very beneficial for both companies and also benefits a lot of potential customers of ours -- of IBM's and SMIC's.

DANIEL HEYLER: Okay, thanks.

And then, the follow up to that would be on the timing of the technology transfer. Since you're a licensee and I believe you're not participating in the joint development, I'm wondering if there may be a lag in the deployment of 45 relative to, say, the ones that are in the joint development consortium.

DR. RICHARD CHANG: We -- because we enter the 45 a little bit late compared to IBM's other partners, but we are catching up very fast. So I do not know the other companies' production schedule. Maybe we are not far away from because we expect to start production in 2009. And I believe the other companies may be also close to this schedule.

DANIEL HEYLER: Okay, thanks, Richard.

DR. RICHARD CHANG: Thank you, Dan.

OPERATOR: The next question comes from the line of [William Tong] from UBS. Please proceed.

WILLIAM TONG, ANALYST, UBS: Good morning, two questions.

I know that previously there was talk about a strategic investor coming in, and I was wondering could you talk a bit more about the rationale for trying to introduce a strategic investor and also the potential timeline.

DR. RICHARD CHANG: We have been approached by many potential strategic partners -- since about a year ago. And they have different proposals and every one of them are not pure financing investors. They also bring additional value to us. We have been carefully evaluating. But, usually, those investors they are exclusive, mutually exclusive.

So if we accept one, then the other may be not easy for us to work with the other as this kind of strategic investor and partner. So we have been very carefully doing our homework and closely we will work with the potential partners to find out which one is -- really can create the maximum value for both companies, but it takes time. However, we are at the final stage to make the final decision soon and, again, we will make the announcement to the world once it is firm and finalized.

William, does that answer your question?

WILLIAM TONG: Yes, it does.

The second question I had was, for the Shenzhen fab, I think one of them will be a 12-inch fab, as you mentioned before. I was wondering as we exit the DRAM business, initially, what will be the target product for some of these 12 inch capacities?

DR. RICHARD CHANG: For the Shenzhen fab, we are going to build the 8-inch fab first. But, however, when we do the construction we will build the 8-inch shell and the 12-inch fab shell together. Because they are so close to each other we just decided to -- the location is close to each other.

To install equipment, we already purchased equipment, the 8-inch equipment, for Shenzhen. The actual line up, up to 20,000 wafers per month is already secured. So once the building is ready, the facility is ready then the equipment can be moved in very quickly; 8-inch first. 8-inch, definitely, will be basically Logic and also some -- we also focus -- we are going to support a lot of MEMS type of customers from Shenzhen -- Logic and MEMS.

WILLIAM TONG: Okay, thank you.

DR. RICHARD CHANG: Thank you.

OPERATOR: Your next question comes from the line of Eric Chen with BNP. Please proceed.

DR. RICHARD CHANG: Yes, please.

ERIC CHEN, ANALYST, BNP PARIBAS: I noticed your second quarter the CapEx increased. So what's the reason behind for the CapEx increasing in the second quarter? What's it used for? And also, for the year 2008, the depreciation [expense] do you have any idea. could you give us some highlights?

DR. RICHARD CHANG: Thank you. Our CapEx for Q2, many are for the IBM 45 nanometer equipment.

ERIC CHEN: Okay.

DR. RICHARD CHANG: Also our R&D budget increased because we purchased a lot of special equipment, for example the immersion type of scanner in Q2. That's why the Q2 has increased. Because when we do the R&D and the technology transfer we like to pull ahead to do this and our partner is working very smoothly with us. So we found we really can do faster. That's how we pull ahead.

ERIC CHEN: Okay, how about the whole year, the depreciation?

DR. RICHARD CHANG: Whole year, our CapEx is still the same. $700m is our budget -- our plan, budget plan.

ERIC CHEN: Okay. That's for CapEx. How about depreciation expenses? Do you have any idea?

GARETH KUNG: As disclosed in our earning release, right now we are doing an analysis on any potential impairments in terms of our Beijing fab. That would have an impact on our going forward depreciation. So, at this point in time, we cannot provide detailed guidance.

ERIC CHEN: Okay, thank you, anyway.

The second question is regarding Spansion. First, I would like to know do you have any CapEx plan on the new equipment purchase, and how you deal with all the DRAM equipment going forward?

And for Spansion, NOR Flash equipment, suppose you are putting in the fab 8. Is that correct? How about for the original NAND Flash business in the fab 8 in Shanghai fab. Could you give us an idea? Thank you.

DR. RICHARD CHANG: Okay, very good question also.

Because our NAND Flash is based on Saifun NROM technology, Saifun and Spansion basically use the same technology, all NROM technology. So, for our NAND Flash equipment to support Saifun, and also the NOR Flash technology supporting Spansion equipment list is exactly the same. So we don't have to add any additional equipment to support Spansion. So the CapEx is well under control.

ERIC CHEN: I see. But do you have to buy the additional equipment for NOR Flash manufacturing?

DR. RICHARD CHANG: Yes, we do. Based on our agreement with Spansion whenever the orders, the PO from Spansion, increases we will prepare additional. And that additional CapEx is within our budget of $700m. The reason is because we already have the equipment, very good equipment, borrowed and leased from our Saifun project. So, to extend the capacity to serve Spansion is very natural, very smooth.

THERESA TENG: Hello, operator.

OPERATOR: Your next question comes from the line of Donald Lu with Goldman Sachs. Please proceed.

DONALD LU, ANALYST, GOLDMAN SACHS: Good morning, Richard, Morning.

My first is does your potential strategic investors you are considering including investment with goods rather than cash?

DR. RICHARD CHANG: No. The potential investors, there are quite a few of them. Some propose to have cash plus different contributions to us and certain customers their investment will be their IP. Actually, they did not charge us anyway. They will provide IP to us free as a partnership and also all cash. So there are many different possibilities. We are very carefully evaluating and going to finalize this very soon. Donald, did I answer your question?

OPERATOR: Ladies and gentlemen, this concludes this question and answer session and today's call, and I'd now like to turn it back to Dr. Richard Chang for closing remarks.

DR. RICHARD CHANG: Ladies and gentlemen, SMIC previously, when we started in year 2000 in China, because at that time SMIC was still under the stringent export control from some countries outside China. At that time our permission to transfer technology, or even our permission to mass produce advanced technology, were restricted to DRAM.

So we started with DRAM as the technology driver and, whatever we learnt from DRAM, we applied the technology to develop our Logic technology. That has been serving SMIC's function and purpose and serve our customers very nicely for a number of years, until last year with the special support from many Western countries, that restriction basically lifted.

We have been granted the permission to directly transfer Logic, Advanced Logic technology from many countries outside China. And it happened that the DRAM partners also agreed with our request. They agreed that we can stop DRAM. So we thank all the partners and we put DRAM in history. DRAM technology served SMIC nicely before we were granted with the permission to work on Advanced Logic technology.

So, actually, we were granted the Advanced Logic export license in fourth quarter of 2007. So it took us a while. Finally, we are able to move full gear forward with Logic only. So we are very appreciative to all the friends from different countries. We thank the governments of different countries who continue to support us.

Again, I'd like to take this opportunity to thank our DRAM partners. They helped us in the past and we will -- we hope they continue to do well and we will focus on Logic from now on. And we really thank so many friends. God bless you. Goodbye.

OPERATOR: Ladies and gentlemen, thank you for your participation in today's conference. That concludes the presentation. You may now disconnect. Have a good day.

[Thomson Financial reserves the right to make changes to documents, content, or other information on this web site without obligation to notify any person of such changes.

In the conference calls upon which Event Transcripts are based, companies may make projections or other forward-looking statements regarding a variety of items. Such forward-looking statements are based upon current expectations and involve risks and uncertainties. Actual results may differ materially from those stated in any forward-looking statement based on a number of important factors and risks, which are more specifically identified in the companies' most recent SEC filings. Although the companies may indicate and believe that the assumptions underlying the forward-looking statements are reasonable, any of the assumptions could prove inaccurate or incorrect and, therefore, there can be no assurance that the results contemplated in the forward-looking statements will be realized.

THE INFORMATION CONTAINED IN EVENT TRANSCRIPTS IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE CONFERENCE CALLS. IN NO WAY DOES THOMSON FINANCIAL OR THE APPLICABLE COMPANY OR THE APPLICABLE COMPANY ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY EVENT TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S CONFERENCE CALL ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.]