The conference is now being recorded. Good morning, good afternoon, ladies and gentlemen, welcome to the BE Semiconductor's quarterly conference call and audio webcast to discuss the company's 2016 third quarter results. The audio webcast is available on BESI's website, www.besi.com. Joining us today are Mr. Richard Blickman, Chief Executive Officer, and Mr. Cor te Hennepe, Senior Vice President, Finance. At this time, all participants are in a listen-only mode. Later, we will conduct a question and answer session, instructions will follow at that time. As a reminder, ladies and gentlemen, this conference is being recorded and cannot be reproduced in whole or in part without written permission from the company. I would now like to turn the call over to Mr. Richard Blickman. Go ahead, please.
Thank you. Thank you all for joining us today. I will begin by making a few comments in connection with the press release we issued earlier today, then we will take your questions. I would like to remind you that some of the comments made during this call, some of the answers in response to your questions by management may contain forward-looking statements. Such statements may involve uncertainties and risks as described in the earnings release, other reports filed with the AFM. For today's call, we'd like to review the key highlights for our third quarter and nine months ended September 30 this year, also spend some time updating you on the market, strategic initiatives, the outlook. First, some overall thoughts on the past quarter and year-to-date results.
BESI reported another solid quarter with Q3 2016 revenue and profit exceeding expectations, strong cash flow generation. Revenue increased by 30.7% versus the third quarter last year, primarily due to increased investments by customers in mobile and automotive advanced packaging capacity in a market more favorable than a year ago. Similarly, BESI's seasonal 13.5% quarterly revenue decrease versus Q2 2016 improved significantly from last year's 30.9% sequential quarterly revenue decrease. Net income of EUR 16.6 million increased by 163.5% versus the same quarter last year as revenue expanded and we realized increased efficiencies from our business model due to BESI's strong market position and ongoing strategic initiatives. In addition, net margins more than doubled to reach 17.6% versus the third quarter last year. Further, net cash grew to EUR 131.9 million in the third quarter due to strong profit generation and enhanced working capital management.
BESI's solid profit and cash flow generation have enabled us to significantly enhance shareholder value in recent years. In October 2016, we completed a 1 million share buyback aggregating EUR 22.5 million. Since 2012, BESI has spent EUR 157 million on dividends and share repurchases using excess cash resources. Given our outlook and prospects, we have initiated a new share buyback program up to a maximum of 1 million shares through October 2017. BESI's nine-month results also demonstrate solid financial and strategic progress, the operating leverage in our business model. While revenue grew by 4%, net income of EUR 48.6 million grew by 23.7% and was roughly equal to our net income for the whole year 2015.
Revenue growth has benefited from increased investments by Chinese and Taiwanese subcontractors to expand advanced packaging capacity, the favorable influence of a new technology cycle to further shrink device geometries, and the expansion of BESI's market position in the major supply chains. Net income development has benefited from top-line growth, higher gross margins from increased labor and material cost efficiencies, and the execution of strategic initiatives to control operating expenses. With that, I'll turn the presentation over to Cor te Hennepe, our SVP of Finance.
Thank you, Richard. BESI had a good quarter on both a top and bottom-line basis. The Q3 2016 sequential revenue decrease was primarily due to lower die attach demand by Asian subcontractors for mobile applications after a large first-half capacity build and typical seasonal factors. The partial offset was ongoing strength in the automotive applications. Q3 2016 revenue was better than guidance and increased significantly versus the third quarter of last year. Growth was primarily due to significantly higher demand for die attach systems by Taiwanese and Chinese subcontractors for mobile applications. We also saw strength in sales of epoxy die bonding systems for fingerprint sensor applications. Orders decreased by 22.3% versus Q2 2016, primarily as a result of lower customer demand for mobile, automotive, and high-end server applications by both subcontractors and OEMs and typical seasonal influences.
Orders increased by 4.2% versus Q3 2015, reflecting general strength in both die attach and packaging systems, although order growth varied per individual products. Year-to-date, BESI's revenue and orders increased by 4% and 4.2% respectively versus year-to-date 2015, primarily due to increased demand by Asian customers for BESI's range of high-end and mainstream assembly solutions and more favorable industry conditions. In particular, we saw strength in our epoxy, wafer level, and multi-module bonders and ultra-thin molding systems. We continue to operate at the high end of our target gross margin range this year. In Q3 2016, BESI's gross margin decreased sequentially by 0.4%, primarily as a result of higher labor and freight costs, partially offset by increased material cost efficiencies.
Compared to Q3 2015, 1.8% point gross margin increase was primarily due to labor and material cost efficiencies, as well as foreign exchange benefits from the decrease of the Malaysian ringgit versus the euro. The benefit of BESI's cost control and personnel initiatives were evident in our Q3 and nine-month financial results. Revenue expanded versus both Q3 2015 and the nine-month period, expense growth was contained, thus aiding the growth of our bottom line. In addition, baseline OpEx continues to remain roughly within a range of EUR 22 million-EUR 25 million per quarter. Headcount trends continue favorably as we increase lower-cost fixed personnel in Asia while reducing higher-cost European employees and vary our temporary personnel per quarter, depending on shipment forecasts. BESI's favorable net income development this year reflects both the progress of our product portfolio as well as increased efficiency of our business model.
With 30.7% revenue growth, higher gross margins, and cost controls, net margins approximately doubled to reach 17.6% in Q3 2016 versus the third quarter of last year. Similarly, net margins grew to 17.2% year-to-date versus adjusted net margins of 13.3% in the prior period of 2015, which excludes restructuring benefits of EUR 3.3 million. BESI's effective tax rate of 11.1% for Q3 2016 and 9.8% year-to-date are at the low end of our target range of 10%-15%. Quarterly tax rates can vary based on the relative profit mix of our subsidiaries and periodic changes in valuation of deferred tax assets in recent years. BESI continues to generate strong cash flow from operations as a result of higher levels of profitability and improved working capital management, particularly in the area of supply chain management and inventory control.
At the end of Q3 2016, BESI's cash and cash equivalents increased by EUR 21.2 million versus the second quarter of this year to reach EUR 153.3 million, and the net cash increased by a like amount. Year-to-date, BESI's generated cash flow from operations of over EUR 65 million, of which EUR 45.4 million was utilized to pay cash dividends and EUR 17.5 million was utilized for share repurchases. During the quarter, BESI's repurchased 231,000 of its ordinary shares for EUR 6.3 million and completed the 2015 repurchase program post quarter end. With that, I'll turn the presentation back over to Richard.
Thanks, Cor. I'd like to update you on the market and our guidance for Q4 2016. BESI's strategy focuses on technological leadership in those markets with the greatest long-term potential. Of equal importance, we seek to reduce costs to enhance our competitive position and increase profitability. On the product side, we refresh systems every one to two years to meet ever-demanding industry specs for speed, accuracy, and miniaturization. In addition, we are at the forefront of technological processes such as TCB and wafer level processing, which have favorably influenced revenue development in recent years. In 2016, customers added to their advanced packaging capacity, particularly Chinese and Taiwanese subcontractors, in response to customer roadmaps for under 60 nanometer applications. As such, we've seen a significant order increase for BESI's eWLB systems, a market in which we have the leading market share.
The TCB market has been relatively quiet in 2016 versus strong growth in 2015 as memory manufacturers digest capacity purchases and determine future demand. We have not seen meaningful application expansion beyond the memory market to date. BESI is also seeking additional opportunities in the solar and battery plating markets after last year's success and to incorporate ever greater commonality in the parts and modules used in our platforms. From an operating perspective, BESI's major focus has been the successful expansion of our Singapore die bonding functional capabilities. In addition, we've greatly increased our Chinese die bonding production for the local market, which adds in the average 30% to BESI's total production capacity. Chinese production greatly expands BESI's revenue potential and adds another source of lower-cost, flexible capacity with minimal capital investments.
In addition, Besi completed the transfer of die sorting production from Austria to Malaysia on time and as per plan in the third quarter of this year. A couple of words about the market and our Q4 guidance. The industry outlook for the second half and full year 2016 has improved significantly. VLSI has again upped its 2016 assembly equipment growth estimate, which now stands at 10%. This reflects both increased demand and a downward revision to 2015's market size. They now forecast continued growth of 9.6% in 2017 and 3.9% for 2018. Since the increased utilization of semiconductors in retail applications such as smartphones and autos, our business has become more seasonal. Orders ramp in the first half of a year and then decline in the second half as capacity is digested and products are introduced.
You can see in the chart that on average, BESI's quarterly revenue declines sequentially by roughly 16% and 12% in the third and fourth quarter, respectively. Q3's decline was slightly better than average, and our 12.5% Q4 2016 decrease at the midpoint of guidance is right in line with historical averages. To complete Q4 guidance, we expect gross margins to remain in the range between 49% and 51%, based on BESI's current product mix, and that OpEx will increase between 0% and 5%, mostly due to one-time higher advisory and consultancy cost. Full year 2016 revenue and operating income are expected to increase by 4.5% and 6.9% over 2015 levels, assuming the midpoint of the fourth quarter guidance. We also expect to further expand our net cash position by year-end from current levels. That ends my prepared remarks. I would like to open the call for some questions.
Operator.
Thank you, sir. Ladies and gentlemen, we will start the question and answer session now. To be registered for the question and answer queue, please press star one. Your questions will be answered in the order that they are received. If you have a question, please press star one. The first question is from Mr. Nigel van Putten, ING. Go ahead, please, sir.
Thanks. Good afternoon, guys. We ended your prepared remarks on the outlook according to VLSI. Can I just get some more color on what you think will be driving 2017, your relative positioning? Do you think you can outgrow the market? Also, if I remember correctly, previously, 2017 was meant to be a strong year because of the 10-nanometer ramp. Do you still have that view? Then for Cor, I was a little bit surprised by the uptick in operating costs. Can you provide a bit of guidance now you've completed most of your move of personnel towards Asia, what we should expect sort of as a run rate for next year in terms of operating expenses? Thanks.
Okay, let me answer your first question. 2017, yes, a further rollout of smaller design geometries, 10 nanometer to some extent, but most of course, still in the 14, 16, and 18. In addition, there's a clear picture developing that we will have growth across the board after 2 years of very careful, you could say, recession, especially 2015. 2016, as mentioned, has come out of that recession somewhat earlier than expected. Besi's products position in all of its areas has improved significantly year by year. That bodes well for a growth which may come our way if VLSI is right for 2017.
Yes, thanks. Maybe if they're not right, would you still expect you to outperform the market? I mean, not focusing on the VLSI number, I'm just trying to see if you can outgrow the market, whatever that number may be.
Well, our focus, as you may know, is in the first place on sustained profitability. Market share as such is interesting from an end calculation, Our focus is day by day, quarter by quarter, on finding the right spots in the mainstream of this market where we can generate above-average margins and also net profits and cash. If the growth does not happen, and we have demonstrated that in the past years, I think we're now at quarter 24, profitable seven years in a row. That's Besi's strategy. Clearly in the higher segments of the market, both in the three pillars. The computing world, the high-end processors for the server market, we have the communication market, automotive. They're all impacted by general economic situations. If the expectation for next year becomes more moderate, our product positions won't change. Does that answer your question?
Yes, definitely clear. Thanks. Maybe the question on OpEx to Cor.
Yeah. You asked for OpEx 2017. Of course, as you know, we don't guide for a year ahead. In general, what we say is that we keep on working on our costs. The increase in Q4, because that's what you say, you see an uptick in Q4. In the conference call, we just said there's some one-time higher consultancy costs, what also plays a bit of role is an IFRS thing, that's the provision for holidays. As you know, Q3 everybody goes on holiday. You have a reduction of that provision, and then you start to build up again. That also added a bit to the uptick as we guided. For next year, we indicated clearly that we do everything to keep costs under control and to reduce our expenses. It's a bit difficult to give a guidance for next year.
As we've demonstrated also this year, we don't let our cost go up dramatically. You could expect costs at an equal or somewhat lower level. This is certainly not the start of significantly increase of expenses.
One of the elements, Cor, you should add is building up Singapore.
Yeah.
As you may know, we have expanded our process and service support, especially in the software arena and process of the various die bonding areas in Singapore. We've added significant headcount, and by training the people at some point that will enable us to reduce the headcount in the respective European facilities. You can expect that in the first quarter, we will have some additional releases in Europe, and that will bring down the average cost of the OpEx, again, closer to the lines as Cor just explained. On an ongoing basis, we have moves to Asia, and always for some period, there is sort of a double cost involved, and that's at this moment impacting Q4 as well.
Clear. Thanks for taking my questions.
Okay.
Next question is from Mr. Peter Olofsen, Kepler Cheuvreux. Go ahead, please, sir.
Yes, good afternoon. I wanted to come back on the industry outlook. More specifically for 2016, where the VLSI Research forecast now go for 10% growth. If I take your Q4 guidance, I get to something like 4% or 5% growth, basically for the full year. Based on your comments on your strategy, I understand you're not going for market share. Could you maybe explain why you might grow a bit less than the overall industry this year? Is there a currency effect? Are there some fast-growing segments where you're less exposed? Any color there would be helpful. One follow-up on the costs and the OpEx. Could you maybe quantify these one-time costs in Q4, maybe to get a feel for what the underlying cost run rate is? Final question on TCB.
I understand that thus far you have not seen a broadening of applications beyond the memory cubes. Any idea when we might see that happening, that broadening of applications? Is that something that might happen in 2017, or is that something further out into the future?
Okay. We're happy to answer. Thanks. First of all, what we hear is that also at the lower end of the market or lower end, let's say LED market, there's a seasonal or you could say cyclical recovery happening at this moment because that's taking place especially in China, so the lower cost part of this industry, and we're not that much active in LED. Also one should note that VLSI Research numbers are pretty flexible in terms of going up and down. We still will have to see. One of the reasons mentioned in the script as well is that 2015 has been reduced, and only by that reason, the growth for 2016 is higher. If you take BESI's over average growth in 2015 compared to the contraction in 2015, you would have to take that into account as well.
My first remark is that the, you could say recovery in the lower and medium end of the market is driving this. OpEx. We mentioned some of the impacts. Consultancy cost also has to do with the preparation of a stronger ramp in China. We mentioned that building up systems, assembly, and test capabilities in our own facility. The steep ramp we currently have enjoyed and will enjoy even more so next year, requires a lot of support, and that is temporary support. We can also give you the bridge. It's also in the presentation from our average levels in SG&A and to build the bridge to our guidance. Cor, that is on slide-
If you look at the baseline operating trends, you see some of the costs that are fluctuating. We take those out because they are fluctuating, to come to a baseline OpEx. These costs that are utilized for the ramp in China are then in the baseline. If you look at the baseline Q1, Q2, you see already some increase for those kind of expenses to Q3. We can say that the significant part of the increased sort of uptick, as called early in Q4, has to do with those expenses.
Okay. That's helpful.
On the third question, TCB. Very important to understand at this moment, TCB is substrate-related assembly. Again, it's a step further up the ladder compared to flip chip. With TSV and stacking memory onto logic. There's a very strong development currently also in the wafer level arena, and that is to somewhat competing with the single die TCB on substrate. There are more battles going on than only the cost reduction in TCB and improvement of the yield. Will next year be a strong year? It's hard to forecast that. It proves once again that these new technologies, same goes for eWLB, similar to when flip chip was introduced. The start is very slow. Focus is more to stretch the current technologies because of cost. Similar to the wire bond process is still used very broadly.
It could very well be that next year we would see a ramp. That all depends on the yields we achieve for the major volume customers so far. That is what will be more clear in the first and second quarter next year. So far, so good. We're doing very well on both fronts. TCB, we have a solid installed base and development with several customers. On eWLB, as we mentioned, we've made a major expansion in that developing market, in the second half of this year in particular. Besi is very well-positioned.
Okay. That's helpful. One final question on plating, especially on solar. Could you maybe shed some light on the progress there, and whether that could be an area with decent growth in 2017?
Solar is also on its way to complete a year of major proof of concept in volume of solar using copper as opposed to silver. It looks very promising, and it could well be that next year for expansion of solar capacity worldwide, we will see more investments in our plating technology. Also here, you must realize there's a huge installed base today using silver and the solar world being far more competitive than the semiconductor world. One cannot imagine replacements. It's more the expansion of the capacity where our chances are growing. Okay, thank you.
Ladies and gentlemen, if there are any additional questions, please press star one. For any additional questions, please press star one. There is a question from Mr. Philip Scholte, Kempen. Go ahead, please.
Yes. Good afternoon. I have a question regarding your new share buyback and the reasoning behind it. Did you also think about maybe paying a special dividend or doing something on your dividend policy? I can imagine that at today's valuation, I am not sure if a share buyback is really the most efficient way to return cash. I was wondering if you had any thoughts on that.
Yes, we certainly have thoughts on that. First of all, as a baseline for independent growth and strategical challenges going forward, about a quarter of our revenue in net cash is, let's say, what we adopt as being very important. Everything on top of that is available to return to shareholders, either through a dividend or share buybacks. If you have made the math in the last 12 months, end of September, we have generated close to EUR 90 million net cash. By paying out the dividend over 2015 and the buyback program, as mentioned in the script, our net cash at the end of September at close to EUR 132 million. The guidance for Q4, we will be looking at year-end again at a cash surplus in that calculation, which should be somewhere above EUR 70 million. Our dividend policy is well-known.
Between 40% and 80% of net income is available for dividends. Last year, we did a special dividend, which then lifted the payout to 94% of net income. Again, it is too early to come to conclusions on dividend proposals. We do that when we have the year-end numbers audited, so that will be in February. If you take the overall cash situation, it certainly presents possibilities along both ways to return cash to shareholders. On your remark of the current share price, whether share buybacks are advisable, it is all a matter of if you look at future perspectives and the possibilities of BESI's strategy. We want to stay away from those considerations as a company because the share buyback program, as we have carried that out in the last year, we are every day in the market buying shares.
Whether the share price is at any level is not important. It is how we return excess cash to shareholders. In other words, we are not an active participant in the market. That is what I can say. A few benchmarks again in that sense. One quarter of revenue as a baseline for independent future execution of our strategy in every direction, and everything on top of that is available for dividends and share buyback. Does that answer your question?
Yes, it does. Thank you very much.
The next question is from Mr. Nigel van Putten, ING. Go ahead, please.
Hello again. Just a follow-up, more technical and of a background nature. We've discussing flip chip and TCB and eWLB, but another trend, at least I'm reading a lot about, is towards system in package as the shrink on the system on chip just becomes too expensive. Do you see that as an underlying sort of driver of the advanced packaging market? Maybe besides this trend, what are exciting prospects or things that are currently happening that you expect to drive the market longer term?
Well, system in package is not new, is very, I would say, old.
We are since many years, very active in that area. One of the systems, the Evo, is a multi-chip platform, up to six and now also up to eight different components it can place. If you also look at camera lenses for that matter, they're also multi-component assemblies, and that's what is referred to then as a system in a package. There are, of course, views that that can go much further. For instance, imagine a smartphone not consisting anymore of 100-plus components, which used to be more than 500. But if all those could go into one package, you would have one system in a package, which is very attractive from the overall functionality of a circuit. That direction is a very important direction in the same way as the ones you mentioned before.
At the very forefront of technology, testing, what can we eliminate? I haven't said that yet. If you imagine that you place multiple components without a carrier, a substrate, or a lead frame for that matter, into one package, that is something which reduces the cost, also brings the circuitry down to much smaller geometry, less energy used to drive the whole system. That's, once again, a very important technology field for many years and going forward. The iWatch, for instance, or a smartwatch for that matter, is also assembled in one package. That's a system in one package.
Yeah, exactly. I mean, towards the IoT, I was actually hinting on the smartphone, not smartphone, but the smartwatch. As I heard, the new release by Apple also contained multiple system on chip, no, system in package this time around. I know you guys are exposed through multiple machines. I was just therefore asking the questions on the SiP.
Yeah. Very important.
Clear. Thanks.
Thanks.
The next question is from Mr. Peter Olofsen, Kepler Cheuvreux. Go ahead, please.
Yes, I had a follow-up on eWLB and fan-out wafer level packaging. You mentioned eWLB earlier in your remarks, and that it had seen a nice increase this year. Were you then referring to bonding tools, or is it also relating to molding, for instance, and singulation?
It is both bonding, so placing the chips in a larger spacing because that's what eWLB does, then molding the fan-out wafer into a carrier, and that is done by a molding process. Singulation is a different tool at this moment, which we don't supply in that dimension yet. That may come in the future. The two horses for us in that race are very accurate chip placement below five micron at a very high speed, higher than any comparable in the industry, and the molding.
Okay. Of course, there has been a lot of press reports about Apple TSMC, using what TSMC calls InFO. Can you confirm that the interest and the traction that you have is broader than that? It's more than just one company? You see multiple clients working on this?
Yes. Also with multiple subcontractors. It's not only TSMC, but ASE, Amkor, STATS ChipPAC. The big ones are all going into that direction for principal customers.
Yeah. Okay. Thank you.
Ladies and gentlemen, if there are any additional questions, please press star one. For any additional questions, please press star one. There are no further questions at the moment.
I thank everyone for participating in the call, in case you have any further questions, don't hesitate to contact us. Thank you all. Bye-bye.
Ladies and gentlemen, this concludes the BESI event call. You may now disconnect your line. Have a nice day.