Welcome to the Semiconductor Manufacturing International Corporation's fourth quarter 2016 webcast conference call. Today's conference call is hosted by Dr. Tzu-Yin Chiu, Chief Executive Officer, Dr. Gao Yonggang, Chief Financial Officer, Mr. Gareth Kung, Executive Vice President of Strategic Business Development, Finance, and Company Secretary, and Mr. En-Ling Feng, Vice President of Investor Relations. Today's webcast call will be simultaneously streamed through the internet at SMIC's website. Please be advised that your dial-ins are in listen-only mode. However, at the conclusion of the management presentation, we'll be having a question and answer session, at which time you will see further instructions as to how to participate. The earnings press release is available for download at www.smics.com. Webcast playback will also be available approximately one hour after the event at www.smics.com.
Without further ado, I'd like to introduce to you Mr. En-Ling Feng, Vice President of Investor Relations, for the cautionary statement.
Thank you, operator. Good morning and good evening. Welcome to SMIC's fourth quarter 2016 earnings webcast conference call. For today's call, our CEO, Dr. T.Y. Chiu, will first provide some remarks. Afterwards, our CFO, Dr. Gao Yonggang, will highlight our financial performance and give guidance on the next quarter. Our Executive VP of Strategic Business Development, Finance, and Company Secretary, Mr. Gareth Kung, will give the detailed financial commentary. This will then be followed by our Q&A session. As usual, our call will be approximately 60 minutes in length. The earnings press release and quarterly financial presentation are available for you to download at our website under Investor Relations in the Events and Presentations section. Before I turn the call over to Dr. T.Y. Chiu, let me remind you that the presentation we'll be making today includes forward-looking statements.
These statements and other comments are not guarantees of future performance, they represent the company's estimates and are subject to risk and uncertainty. Our actual results may differ significantly from those projected or suggested in any forward-looking statements. For a more complete discussion of the risks and uncertainties that could impact our future operating results and financial condition, please see our filings and submissions with the U.S. Securities and Exchange Commission and The Stock Exchange of Hong Kong Limited, including our annual report on Form 20-F, filed on April 25th, 2016. During the call, we will make reference to financial measures that do not conform to generally accepted accounting principles, GAAP. These measures may be calculated differently than similar non-GAAP GAAP data presented by other companies. Please refer to the tables in our press release for a reconciliation of GAAP to the non-GAAP numbers we will be discussing.
Please note that all currency figures are in US dollars, unless otherwise stated. Please also note that all 2016 full-year figures are based on the summation of the unaudited quarterly results for the year of 2016. I will now turn the call over to our CEO, Dr. T.Y. Chiu, for the opening remarks.
Thank you, En-Ling. Greetings to everyone. We wish every one of you an exciting, prosperous, and healthy Year of the Rooster. We just finished another record year in 2016 with great performance and significant business growth. We recorded historical high revenue of $2.9 billion and annual revenue growth of 30% over 2015, outpacing the foundry industry average. Operating profit reached an all-time high of $340 million, representing 12% operating margin. Net margin was a high of 11%, and the net profit attributable to SMIC reached a record high of $376 million. EBITDA surpassed $1 billion for the first time, $1.1 billion to be precise, and we achieved an improved annual ROE of 9.6% from 7.6% in the previous year. In terms of significant accomplishment in the last year, we successfully acquired LFoundry in Italy, thus securing a significant foothold in the auto IC market.
I'm also proud of the team's quick ramp-up of Beijing JV Fab as well as the Shenzhen Fab, while maintaining high overall utilization of 97.5% in Q2 2016. Over the past few years, we continue to improve productivity with an increase of 8.9% last year, 8.9% revenue per headcount in 2016. In addition, SMIC engaged several dozen new customers and the number of new products to enter risk production increased 50% compared to 2015. In 2016, we also successfully completed 10-to-one consolidation of our ordinary shares, which we believe has attracted a broader range of investors. We had another solid quarter in Q4 to wrap up the strong 2016 year. We achieved our eighth consecutive quarters of record high revenue, $815 million, representing a growth of 33.5% year-over-year and a 5.2% quarter-over-quarter.
Our Q4 gross margin was 30.2%, and annualized ROE maintained a double digit of 10.1%. 28 nano, 40 nano, and 0.13 micron drove most of the growth in Q4 2016. 28 more than doubled sequentially, contributing 3.5% of our wafer revenue in Q4. 40 nano grew 92% year-over-year and 9.3% quarter-over-quarter. The 0.13 micron grew 103% year-over-year and 21% quarter-over-quarter. From an application perspective, specialty technology, including imec, non-volatile memory and sensor together grew above 26% year-over-year and 7% quarter-over-quarter. We are also pleased to see that smart car business has taken off recently as our smart car related revenue tripled year-over-year and grew 48% quarter-over-quarter. From a regional perspective, all regions experienced healthy growth when comparing Q4 2016 to Q4 2015. China, 42% year-over-year. North America, 36% year-over-year.
Eurasia, 14% year-over-year. We have exited 2016 with strong momentum, while in short term we see the impact of seasonality as reflected in our 2017 Q1 guidance, our team has responded quickly to fill in the gaps, we are targeting still 20% annual revenue growth in 2017. In this year, we continue to focus on careful expansion of our existing facilities in response to customers' needs, while executing our strategy of keeping a balanced focus on both growth and profitability. In 2017, we target to maintain a mid to high 20% growth margin and an EBITDA margin of high 30s. With regard to key growth driver for 2017, 28 nanometer will be one of the primary contributor to growth, we target 28 wafer revenue contributing to reach high single digit contribution by the end of the year on a quarterly basis.
We experienced great demand from 40 nano in 2016. In 2017, we are able to begin to transition some of our 20 and 40 flexible capacity towards 28 nanometer. Other growth driver in 2017 include a more diverse variety of mature technologies. This year, we expect revenue growth will be from various geographic regions and particular strength from North America-based customers. We continue to benefit from our strong position in China, not only the growing domestic fabless industry, but also from international customers with a desire to capture more content share in China. In 2017, we expect to increase absolute dollar of R&D spending to low to mid-teens of revenue, the highest as a percentage of revenue among all major pure-play foundries. We continue to follow our technology strategy of diversification and paced advancement, which feed us continuous growth opportunities.
Our R&D spending covers both advanced and the specialty technology, allowing our customers a platform for a longer-term collaborative roadmap. We are preparing the baseline for diversified technology, we are investing much of our R&D this year on 14 nano FinFET, which is in early stage with process flow and features defined. SMIC is among the world's top five patent filers for FinFET, both domestically and globally. Meanwhile, given 2016's high utilization of 97.5%, to address the need of more capacity, plans to consolidate the foundry CapEx of 2016 is $2.3 billion, of which about $850 million is for the Beijing JV Fab, which will be 49% funded by our joint venture partners. We plan to add an estimated 11% total installed capacities to close out 2017 with 450,000 wafers per month, compared to 406,000 at the end of 2016.
In terms of overall annual effective capacity, the planned capacity growth is approximately 25% in 2017 versus 2016. All capacity addition this year will be to our currently running fabs, in addition to the Shenzhen 12-inch fab, which will install a new line by the year-end. With more than $2.1 billion cash on hand as of December 2016 and increasing cash generation from operations, we are in excellent position to fund our 2017 CapEx plans. We will continue to expand carefully as we gauge our demand and the overall market. I would like to take the time now to welcome our newest board members. We welcome Dr. Chiang Shang-Yi, renowned foundry industry R&D veteran, Dr. Tong Guohua, a distinguished businessman, entrepreneur, chairman, and president of Datang, and Dr. Jason Cong, influential professor and a researcher of advanced computing at the UCLA.
We are very honored to have such prestigious and seasoned experts joining our board to contribute their valued insight to our company's vision and directions. To conclude my remark, SMIC has delivered excellent performance in 2016 and continue to strive to grow profitably and add value. We reiterate our target of 20% compound annual growth from 2016-2019. We have an advantageous position here in China, and we continue to work hard to seize opportunities for the benefit of our stakeholders. In this time of growth, our team strives to perform above par, executing our strategy of differentiation and diversification, serving our customers with excellence, and building value through balancing growth and profitability. We thank you for your continued support and for your time. I will now hand the call over to Yonggang for the financial highlight and 2017 Q1 guidance.
Okay. Thank you, Tzu-Yin Chiu. Greetings to all our listeners. First, I will highlight our 2016 full year unaudited results, which are based on the summation of our unaudited quarterly results for the year of 2016 and our fourth quarter 2016 results, and then we give our first quarter of 2017 guidance. Revenue in 2016 was $2.9 billion, a record high, compared to $2.2 billion in 2015. Gross margin in 2016 was 29.2% compared to 30.5% in 2015. Profit for the period attributable to SMIC of 2016 was $377 million, a record high, compared to $253 million in 2015. Net profit margin was 12.9%, a record high, compared to 11.3% in 2015. ROE reached a record high of 9.6% in 2016, compared to 7.6% in 2015.
EBITDA reached a record high of $1.1 billion in 2016, compared to $0.8 billion in 2015. Now I will highlight our fourth quarter 2016 results.
Our revenue was a record high of $815 million. Gross profit was record high of $246 million. Gross margin was 30.2%. Profit for the period attributable to SMIC was $104 million. Now looking ahead into the first quarter of 2017. Our revenue is expected to decline by 2%-4% quarter-over-quarter. Gross margin is expected to range from 25%-28%. non-GAAP operating expenses are expected to range from $158 million-$164 million. Non-controlling interests of our majority-owned subsidiaries are expected to range from positive $6 million-positive $8 million, which are losses borne by non-controlling interests. The planned 2017 CapEx for foundry operations are approximately $2.3 billion, while the planned 2017 CapEx for non-foundry operations are approximately $70 million. I will now hand the call over to Gareth for a more detailed financial commentary.
Thank you, Gao. Thank you everyone for joining us today. On the income statement, revenue increased by 5.2% Q-on-Q to $815 million, mainly because of an increase in wafer shipments in Q4 2016, excluding LFoundry, and also the revenue contributed from LFoundry. LFoundry only contributed for two months in Q3 2016, whereas it contributed to the full quarter in Q4 2016. Gross margin was 30.2%, above the guided range, mainly due to product mix change. Operating expenses increased to $197 million in Q4 2016. R&D expenses increased by $36 million Q-on-Q to $118 million. The change was mainly due to high level of R&D activities. Funding of R&D contracts from the government was $23 million in Q4 2016. G&A expenses increased by $25 million to $61 million in Q4 2016.
The increase was mainly due to accrued employee bonus. Excluding the effect of employee accrued bonus, government fundings and gain from disposal of living quarters, non-GAAP operating expenses were $193 million in Q4 2016. Profit from operation was $49 million. Profit for the period attributable to SMIC was $104 million, while the non-controlling interests were $46 million of credit to SMIC attributable profit. The change in non-controlling interests was mainly due to the recognition in Q4 2016 of the contribution to SMIC's group's advanced technology R&D expenses incurred in 2015 by the company's majority-owned subsidiary in Beijing. Excluding the impact of the finance cost, depreciation and amortization, and income tax benefits and expenses, our EBITDA margin was 34% in Q4 2016. Moving to the balance sheet. At the end of the fourth quarter 2016, cash and cash equivalents increased to $2.2 billion, if including other financial assets.
At the end of Q4 2016, our gross debt-to-equity ratio was 56%. Our net debt-to-equity ratio was at a healthy level of 16%. In terms of cash flow, we generated $406 million of cash from operations for the quarter. On a full year basis, we generated $977 million of cash from operations in 2016 compared to $669 million in 2015. Cash used in investing activities was $128 million. Cash from financing activities was $231 million. To examine our revenue by applications, the communication and consumer segments contributed 44% and 37% of our revenue, respectively, for the quarter. On a full year basis, the communications and consumer segments contributed 48% and 30% of our revenue, respectively, in 2016. Geographically, revenue from China, North America, and Eurasia contributed 48%, 33% and 19% of total revenue, respectively for the quarter. On a full year basis, revenue from China contributed 50% of total revenue.
North America contributed 29% of total revenue, and Eurasia contributed 21%. In terms of technology, revenue from 28 nanometer contributed 3.5%. Revenue from 40/45 nanometer contributed 23.6%. Revenue from 55 nanometers and 90 nanometers contributed 19.8% and 1.6%, respectively. Meanwhile, 0.13 micron above line width contributed 51.5% wafer revenue for the quarter. On a full year basis, revenue from 45 nanometers and below contributed 24%. In terms of our overall capacity, total monthly capacity at the end of the fourth quarter increased to 406,000 8-inch equivalent wafers. The increase was mainly because of the capacity expansion in our Beijing 300mm fab, as well as our majority-owned fab in Beijing during the quarter. The planned 2017 CapEx for foundry operations is approximately $2.3 billion, of which about $900 million will be spent for the expansion of our capacity in our majority-owned Beijing 300mm fab.
The planned 2017 CapEx for non-foundry operations are around $70 million, mainly for the construction of employees' living quarters. Our planned 2017 depreciation and amortization is around $1.1 billion, an increase of about $380 million year-over-year. I would now hand the call to En-Ling for the Q&A sessions.
Thank you, Gareth. I would now like to open up the call for Q&A. As usual, please be reminded to limit your questions to two per person. Operator, please assist.
Thank you. Ladies and gentlemen, we will now begin our question and answer session. If you wish to ask a question, please press star and 1 on your telephone and wait for your name to be announced. If you wish to cancel your request, please press the pound or hash key. Your first question comes from the line of Randy Abrams of Credit Suisse. Please ask your question.
Okay. Yes. Thank you. Good morning. My first question, wanted to ask just on the change in guidance from the original expectation to grow in first quarter. Could you talk about how broad-based the slowdown you saw in first quarter, or whether it was isolated to a few applications? Then, maybe talk about the rebound, if you see it more a matter of new products being qualified, or it is the same applications coming back. If you could characterize just whether any business was lost to other foundries, given you were tight for several quarters last year.
Okay. Thank you, Randy. Yes. We have seen business loss in a few area. Our customers seems to gone through some of the marketing difficulties. We think that is on a few particular area. We are winning additional opportunity to backfill these capacities that is coming up. Actually, it is a good opportunity for other customers. During last year, we had very constrained supply for them, and now they can freely grow with SMIC. We think that eventually we will be able to backfill completely these downward trend.
Actually, I think the slight correction in Q1 is really in line with what we see in the industry. I won't say it's anything specific about SMIC in this regard.
Okay. Great. Hey, I wanted to ask a follow-up question on the margins. Could you go through the factors, the gross margin dipping to 25%-28%? Maybe how much is the depreciation coming in or ramping through the year? If you could also talk the impact of now moving some business from 40 to 28, if there's any headwind at this stage ramping up 28 relative to the mature 40 nanometer. I guess that margin 25%-28%, if you think that should be reasonable range through the full year, or any headwind from those factors, depreciation or 28 nanometer.
Yes. Well, first of all, regarding the dip in the Q1 gross margin, I think there are two impacts here. First of all, you're right to point out depreciation is a factor because we're going to see about $30 million increase in the depreciation in Q1 compared to Q4. Secondly, as mentioned by TY, we're going to see some slight correction in business in Q1. You're going to see our utilization is going to drop below about 90% in Q1. That has some impact on our gross margin, for sure.
Okay. I guess, Stephane, can you maybe give a look as we go through the year if you improve utilization, but then factoring depreciation in 28, if there's a rough feel for maybe a medium term or through the year, how gross margins may trend?
As I said, let me just clarify. I think we are looking at a utilization of high 80s, close to 90% in Q1. For the gross margin for the rest of the year, as mentioned earlier, that we're going to experience, I would say, an increase in depreciation in 2017 over 2016 to tune of about $380 million. That is in line with the expansion in our fab capacity. We're going to ramp up our fab in Beijing and in Shenzhen. Obviously that will have an impact on our gross margin. We are still guiding, I would say, we are still targeting a mid to high 20s gross margin. As I said, at this point in time, we don't have a clear picture about 2017, but this is our initial feelings.
Okay. Is 28 still dilutive, like as you shift from 40 to 28, where yields are far enough along, that it's less headwind now?
Our yield has actually been along our expectations, and I think that it's still doing steady improvements. I think that it should not be a significant drag to our margin. I think it's just simply the increase in depreciations.
Okay, great. Thanks a lot, TY and Stephane.
Thank you.
Thank you.
Thank you. Your next question comes from the line of Steven Pelayo of HSBC. Please ask your question.
Yeah. A few questions. First of all, I guess 2017 growth, you're talking about 20% year-on-year growth with North America outperforming. That's roughly one-third of revenues growing faster. I guess, does that mean China underperforms? Do you think Eurasia actually can decline? When you think about the other regions for 2017, maybe a little bit of color there. Also, could you comment a little bit on 2017 from a node perspective with 28 nanometer ramping to, I think you said, high single digits by the end of the year. Does that mean 40 nanometer still grows this year? Help us understand the node outlook for 2017 as well, geographically and by nodes.
Hey, Steve. We commented North America is going to grow quite well in 2017. They are for two reasons. First of all, 28 nanometer will be a major growth driver in 2017. As you know, we have major U.S. customers for the 28 nanometers. Secondly, for LFoundry, we only consolidated the company for about five months in 2016. For 2017, it will be a full year consolidation. Their customer base are mostly in the North America. That also contributed to the growth in the U.S. customers. Okay. In terms of the node perspective, I think the major growth, as we can see it right now across the node, there's not enough visibility to say precise at this point in time. Will be mainly from 28 nanometers, 65 nm, and 55 nm, as well as the growth, as I mentioned, the contribution from LFoundry.
40 nanometer node, do you think it actually decline year-on-year?
No. We are seeing the major growth. We still think that the 40 nanometers would remain to be a major node for us.
Okay. Then, we struggle to kind of forecast your model here with your operating expense volatility with R&D credits and some of the other things that go on, property sales and other benefits that you can get through there. Then now bonus accruals as well. Could you talk just a little bit more? I know you like to guide excluding those things, but could you talk a bit about including those things? What kind of expectations you have for those in the first quarter as well as 2017?
What we've guided in Q1 in terms of the OpEx, we don't think there's any extraordinary items in Q1 that you should be concerned about. In terms of the full year OpEx guidance, in terms of the normalized OpEx, we're still looking at a high teens number relative to the revenue. The main increase will be in the R&D area, as mentioned by TY. We will continue to invest heavily in R&D, we will continue maintaining very disciplined spending in terms of G&A expenses and sales expenses.
Let's see. This is the struggle, though, because I think R&D in 2016 was about 13%, I think of revenue, if I exclude the R&D credits, maybe around 11% including. I guess on a comparable basis, you talked about that, I think going to the low teens, if I remember correctly, as R&D. Is that including or excluding R&D credits? Do you expect R&D subsidies to increase in 2017? What kind of numbers should we think about?
Yes. Right now, we are guiding the R&D spending probably in the low teens to mid-teens level. That is excluding R&D funding. In terms of the R&D funding, for 2016, for the whole year, it's about $52 million. We expect this number would go up in 2017, maybe close to $65 million to $70 million. This number, as we mentioned earlier, that there's some uncertainties depending on the completion of these R&D projects as well as the funding availability from the government.
Okay. The last line item that I also struggle on is the non-controlling interest line. I guess I'm trying to understand as this fab ramps up, does it ramp up initially where you have more losses and then there's a greater add back, or does that number then decrease over time? When you think about that line through 2017 or maybe just on a full-year basis, how do you think that will track this year or contribute this year?
Well, first of all, the Q4 number, in terms of the contribution from the non-controlling interest, was high because of this sharing of R&D expenses with our majority-owned subsidiary in Beijing. That part contributes about $29 million for the quarter. Okay. Actually, if you remove that number, actually the number is quite consistent quarter-on-quarter. Right now, we are still looking at similar number in 2017. Yeah.
I'm sorry, similar every quarter, this kind of $7 million-$8 million, I think, is what you guided-
Yes. That's right. Yes.
per quarter.
Yes. That's right.
Okay. Excellent. All right. I'll get back in the queue. Thank you.
Thank you. Your next question comes from the line of Roland Shu of Citigroup. Please ask your question.
Yeah. Thanks for taking my question. First question is, can you repeat the overall capacity increase number this year?
Just give me a second. The increase in the capacity for 2017. Let's see. We are looking at an annualized capacity increase of about 25%. On year-end basis, the increase is about 11%.
Year-end means compared with year-end in last year?
Yeah. That's right. We're talking about year-end over year-end, about 11% increase. On an annualized capacity basis, it's about 25% increase.
May I add.
How about 12-inch and 8-inch increase?
It's combined. Yeah.
Yeah.
How about the increase on the 12-inch? Do you have the breakdown for 12-inch and 8-inch increase?
Obviously, the increase in 12-inch is more than 8-inch. I think the major expansion in the capacity will still be in our 28 nano capacity. Also, there'll be quite a bit of increase in our 65, 55 nano capacity. Yeah.
Okay. Yeah. Okay. Thank you. You talk about first quarter-
Just to add that the 25% annualized effective capacity increase, also a lot of it is because full year's foundry's capacity will come in. Yeah.
Okay. Thank you. Yeah. For your utilization in first quarter, talk about close to 90%. How about the 12-inch and the 8-inch utilization specifically?
Well, we don't break out the utilization for each fabs. I would say it's quite even. Yeah.
Okay. Both the 12-inch and 8-inch are close to 90%.
I would say 8-inch is still better than 12-inch. Yeah.
Okay. Yeah. Thank you. For the whole year, the first quarter revenue will be corrected a little bit, the whole year revenue target still grow by 20%. What's the quarterly revenue linearity in this year? Will it be increase gradually, or will it be more backend loaded?
I think it's too early for us to comment on that. I think the 20% growth is our target. Okay? At this point in time, I think we don't have much visibility beyond the first quarter. We'll update the market as we see more visibility.
Okay, thanks. I think last question is for you. Others revenue by application increased a lot in Q4 last year. What contributed into this category? Was it just a one-off, or will it be recurrent in this year? Thank you.
Yeah. That category, actually, we have put in the revenue from LFoundry. As you know, LFoundry's customers are mostly in the auto industrial sectors.
Okay. We'll expect it will be increased meaningfully this year because we'll have the full year consolidation number from LFoundry this year, right?
Yeah. That's correct. Yes.
Okay. Thank you. Yeah.
Thank you. Your next question comes from the line of Leping Huang of CICC. Please ask your question.
Okay. Thank you. My question. The first question is about your 28 nanometer migration strategy. I remember that you previously was mainly think that the competition in 28 was too intense and the profitability, you would prefer to focus on the 40, 45 nanometer process, which is much better in terms of profitability. What has changed that this time you start to migrate into 20 nanometer? It's more a customer driven, or it's more that you have excess capacity to migrate, or which application you will first migrate? Thank you.
Okay. I think last year, basically, we had a very tight capacity year. A lot of our 40 customers or 40 products are single sourced in the SMIC. Any discontinuity in these product supply will really impact a significantly larger economic range. That is the reason that we had actually. Even at the beginning of the year, I think of 28, the market was softer than we had thought. That's why our capacity has shifted towards meeting the 40 nano demand. It wasn't a straight economic margin consideration. This year, we have always intend, even at last year, to quickly ramp up our advanced capacity so that we can meet both the 40 nano as well as the 28 nano demand.
At the second half of last year, we have increased our capacity sufficiently at this point of time to meet both the 40 as well as 28 nanometer customer demand. That's the reason this year we can see a significantly faster 28 nano ramp up.
Okay. The second question is that we see October last year, you announced that you will build a new factory in Ningbo. Can you elaborate what's the plan in Ningbo? Also, these days we see that the local government do provide a lot of favorable financial condition to attract the foundry to the local city like recently Chengdu has a cooperation with GlobalFoundries. Do you also consider to expand your geographical expansion in China? Or I remember you previously mainly focused on Shanghai, Beijing, because of your constraint on R&D resource. Yeah, thank you.
Certainly, SMIC is exploring all potential opportunities. I think, as consistent with our previous announcements, our main focus will still be in our present production sites. That is Shanghai, Beijing, Tianjin, and Shenzhen, as well as expanding some of our capacity in Italy. That certainly doesn't preclude us to consider other opportunities. I think at this moment, that we are still putting in capacities in our present production site, and that is our main focus.
The Ningbo will not be a factory, or it will be? What Ningbo will be? Yeah.
Ningbo, right now, we are exploring design service centers. We are exploring converting some of the specialty technology into production. There are a number of conditions that needs to be
How do I say? Proven and verified, such as a working specialty technologies, and a very strong customer set base, before we start any fab constructions.
Thank you.
Thank you. Your next question comes from the line of Sebastian Hou of CLSA. Please ask your question.
Hi. Good morning, guys. Thanks for taking my questions. The first question is on 28 nanometers. I wonder what's your strategy on 28 for this year. Is more on poly-SiON or high-k metal gate by the end of the year, given your high single-digit revenue guidance? Another follow on that is, your strategy is more on your existing customers migration need, or are you going to gain some new applications for new customers?
Okay. Our 28 certainly is focused not only on a single customer. We have a number of customers, both globally as well as domestically. Indeed, certainly we are going to first meet our customer demand that is put in front of us right away. In the same time, there are a number of customers still interested in the poly-SiON. Of course, we are also getting NTOs product tape out in our high-k technology. This is also a significant focus for SMIC this year. Yes.
Okay. In terms of your revenue guidance for 28, about high single count or high single digit by the end of this year, can we assume that most of that or nearly all of that will be poly-SiON?
I think a significant portion will be poly-SiON, but we are targeting some high-k revenue.
Okay. In terms of your position, what do you compare yourself in terms of performance, pricing, versus the other foundries who are offering 20 nanometer for years?
Competitive.
Competitive on performance and pricing or performance-
Both
or pricing? Both.
Both.
Okay. Thank you, TY. My second question is on your first quarter guidance. You guided to decline by 2%-4%. I think, TY, you already mentioned about this one specific or some specific customer and some specific area see weakness. Also to comment about that you think that you can backfill the capacity pretty soon. How soon is that? Can we expect that you can backfill that within one to two quarters, given that right now major 8-inch foundry is already in tightness right now?
Yes. I think I've been mentioned that for example, we have a customer in the fingerprint sensor area that have been some correction in business. We also seeing other new customer coming in for the same applications, and which are going to ramp up this year. At the same time, for example, we are seeing very good orders from our PMIC as well as from our smart card business to fill the gap. We are, I would say, cautiously optimistic in terms of our growth this year.
Okay. You mentioned about the Yep, sorry. Please go ahead.
I think that indeed, we will be able to, or at least we target to, backfill within a few quarters time.
A few quarters, which means not one quarter.
A few quarters, meaning one or a few quarters. Right.
What I'm trying to get a sense is that is because you need to backfill that, so your 8-inch capacity came back to 100%. When do you expect your 8-inch capacity utilization rate to return to your 100% level or close to 100% level?
I think.
One quarter or two quarter?
This year, we have targeted 20% growth. To achieve 20% growth, you can project that it still needs to be fairly high utilization.
Okay. Thank you.
Thank you. Your next question comes from the line of Charlie Chan of Morgan Stanley. Please ask your question.
Thanks for taking my question, and Happy Chinese New Year. First question is regarding your gross margin guidance because it seems like your depreciation increased a lot. Just putting into perspective, the depreciation was around 20% of last year's revenue, and this year it will increase to 30% of this year's revenue. It is like 10 percentage points increase. You mentioned that the EBITDA margin will maintain at the high 30%, but your gross margin guidance implies only 5% decline. I'm not sure what is the gap here. Are you going to reduce your OpEx significantly, or is there any big change for cost reduction in your variable cost?
Charlie, I think my calculation is somewhat different from yours. We are guiding a $380 million increase in the depreciations. That could have an impact on the gross margin for sure. At the same time, we also said that we could intend to maintain an OpEx of high teens. With that, I think we're still quite confident about achieving, or at least this is our target, to achieve a mid to high teens gross margin. Yeah.
Okay. Maybe I misunderstood. What was the $1.1 billion depreciation guidance for?
That is including both. The total depreciation, the majority amount goes to the cost of goods sold, but part of it is also relating to the R&D spending as well.
Got it. I'm sorry. Okay. Next question is regarding your free cash flow, because you made around $1 billion EBITDA next year, which is a good number. It seems like your CapEx was above $2 billion this year, continues to be above $2 billion. It seems like there is ongoing cash outflow. How are we going to fund for those cash flow?
Yes. You're right. SMIC is still not in a free cash flow positive situation yet. What we are seeing right now is that we are continuing our CapEx this year. At the same time, we are increasing our total EBITDA generations. That should help to reduce the gap. At the same time, as mentioned by TY, we have more than $2 billion cash holding on hand. We should be able to fund the CapEx without too much of a problem.
Okay. Got it. Lastly, 28 nanometer. When do you think revenue would significantly ramp up this year? I think your guidance implies revenue is going to double, right? You think that 2Q, 3Q are very back-loaded.
Right now, we are targeting to increase gradually over each quarters, maybe we could see a more big increase in the second half.
Okay. Yeah. I think another topic, I'm not sure if management team has discussed about that regarding U.S. protectionism on the semiconductor industry. It seems like you own some customers from North America and Qualcomm. They also work with you on some leading-edge R&D, like 14 nanometer. Do you think there's going to be any impact to SMIC in the long term? Do you expect any friction between China and U.S. in terms of semiconductor cooperation?
I think that indeed, we are cautiously optimistic in terms of the market growth and the overall market trend. This is, of course, disregarding the potential trade wars that may really come around. We think that, and hope that the semiconductor industry is a very global industry. I think the market here in China is a great opportunity for everyone, outside of China as well as in China. I think that anyone that really have a careful study will find that the China market and the arrangement at this point of time is good for everyone. We hope that there will be a good environment for business, for foundry, for fabless, as well as for all other customers in China.
Okay, understood. Thank you very much.
Thank you. Your next question comes from the line of Gokul Hariharan of JP Morgan. Please ask your question.
Yeah, hi. Thanks for taking my call. Just wanted to ask first on the R&D spending increase with the emphasis on 14 nanometer. Is there any change in direction in terms of 14 nanometer development and trying to bring it on production earlier than expected? I think previously, I think it was more like a 2020 target. The second question is, with that increase in R&D spending, how does it affect the operating leverage expectations in terms of operating margins as we target the 20% growth CAGR over the next few years?
Okay. Indeed, the 14 nanometers, we are providing a faster turnaround time, a better R&D support to our team. Definitely, if possible, we will try to bring it up in production earlier. It'd be great if we can bring it before 2020. I think that in general, SMIC's strategy is to continuously have a strong support to our R&D efforts, both in advanced technology as well as in our specialty mature technology as well. Okay. Do you want to comment on R&D?
Right now, as mentioned earlier, we are targeting low teens to mid-teens R&D spending. In terms of the OpEx spending, in terms of normalized OpEx, will be in the high teens level. I think in terms of the impact on the operating margin, I think the major impact will be coming from the gross margin, I think.
Okay, got it. Can I also ask, what is the status on LFoundry in terms of utilization as well as new customer qualifications? When do we expect some of the new customers to start filling up the capacity?
Okay. Let me say that. We already see the LFoundry utilization increased by about 10%-15% after the conclusion of our merger. Certainly, right now we are introducing additional customers to bring them into LFoundry. It would take a few quarters to bring in more than a few customers to fully fill the fab. We think that this year, we should be able to maintain a reasonably high utilization in LFoundry consistently.
I think just to add on to what Chiu said, we're also seeing a pretty strong recovery of orders from the auto customers. I think the major reason for that, of course, is the end market for autos is doing well. Secondly, it's also the increasing confidence of their customers in terms of LFoundry's future, given the fact that they are combined with a much bigger group now as before.
Okay, great. Thanks.
Thank you. Your next question comes from the line of Michael Chao of Deutsche Bank. Please ask your question.
Hi, good morning. Thanks for taking my question. One question is, as you know, the fingerprint sensor will shift to 12-inch in the future. What is your planning for 12-inch fingerprint sensor product roadmap this year and the next year? Or do you expect your fingerprint sensor will be still from 8-inch process going forward?
Okay. At this moment, I think the requests to shift from the 8-inch to 12-inch are still limited, and our traditional customers, as well as the new customers, are still mainly focused in the 8-inch capacities. Certainly, I think that we are preparing additional 12-inch capacities for similar applications if there is such a need. This is already being set up and will come online in Shenzhen at the end of 2017.
Does that mean you will use the 65 nanometer to do fingerprint sensor by the end of this year in Shenzhen fab?
We will withhold a comment at this point of time. Right now, we still have a lot of activities in our traditional technology arenas for different applications.
Yeah. I think we receive most of our fingerprint customers are still focused on the 8-inch application at this point in time.
Do you expect that they will shift to 12-inch next year? Or you think that most of the customers still stay eight-inch even next year?
I think for 2017, we don't see any migration at this point in time.
In terms of yield rate, do you think your 12-inch fingerprint sensor will be quite okay if you want to enter mass production at some point, let's say maybe 2018? It seems that 65 nanometer fingerprint sensor yield rate is very challenging for some time in the early stage. What is your view?
As I said, right now, because most of our customers still require only eight-inch production for this application, so we don't have any experience in the 12-inch yet.
My second question is, you guys saw high single-digit % of sales from 28-nanometer. We remember that in the past you've been quite cautious in guiding 28-nanometer sales portion, but it seems that there's some change, so you cannot meet the guidance. This time, what caused the difference? You think that you have more customers than before, or more product, or you do see some product qualification? It seems that TSMC should finish 28-nanometer introduction in the first half of this year. In theory, 28-nanometer pricing condition should be very severe in the second of this year. If you can see the UMC situation, there will be more questions in 28-nanometer ramp-up in terms of pricing and sales contribution. What's your edge going forward?
It seems that UMC will move 28-nanometer to the Xiamen fab, so they may have some benefit from the joint venture with the Xiamen government. What's your view? What would you think about your 28-nanometer sales contribution in the long term?
Okay. I think again, last year, indeed we did not meet our 28 nano forecast. That was basically because we have extremely tight capacities, and we had made a strategic decision to do the 40 first because the demand come in first and a lot of the product are single source. This year, we are able to ramp 28, at the same time maintaining a good 40 output. The main reason is that we have expanded our capacities to a situation where we can take care of both our 40 as well as 28 customers. As far as the dilemma of coming in with a technology that is slightly behind our competitors, that's a dilemma we have faced all throughout our last 16 years.
We came in with our 40 nano and I guess a lot of our peers would think that 40 nano for SMIC is probably a node that is not worth pursuing, but we had a very successful 2016 ramp-up in the 40 nano. We believe that as long as we can do good quality 28, as long as there is customer demand, I think 28 ramp-up is still a very important node for us.
Okay. One follow-up question?
Thank you. I think we've run out of time here. Michael, I think we have to stop here.
Okay. Thank you.
Thank you. I would now like to hand the call back to CEO, Dr. Chiu, for closing remarks.
In closing, I would like to thank everyone who participated in today's call. Again, thank all of our shareholders, customers, employees, and the suppliers for their trust and support. We'll see you next time. Thank you.
Thank you. This is the end of SMIC's fourth quarter earnings conference call. We thank you for joining us today.