高塔半導體 (TSEM) 2016 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the TowerJazz third-quarter 2016 results conference call. (Operator Instructions). As a reminder, this conference is being recorded November 15, 2016.

Joining us today are Mr. Russell Ellwanger, TowerJazz' CEO, and Mr. Oren Shirazi, CFO. I would now like to turn the call over to Ms. Noit Levi, Vice President of Investor Relations and Corporate Communications. Ms. Levi, please begin.

Thank you and welcome to TowerJazz financial results conference call for the third quarter of 2016.

Before I begin, I would like to remind you that some statements made during this call may be forward looking and are subject to uncertainties and risk factors that could cause actual results to be different from those currently expected. These uncertainties and risk factors are fully disclosed in our forms 20-F, F-4, F-3, and 6-K filed with the Securities and Exchange Commission, as well as filings with the Israel Securities Authority. They are also available on our website. TowerJazz assumes no obligation to update any such forward-looking statements.

Now, I would like to turn the call to our CEO, Mr. Russell Ellwanger. Russell, please go ahead.

Thank you, Noit. I welcome all of you to our conference call discussing our third-quarter 2016 results. Thank you for joining us, as well as sincere thanks for your ongoing support of our Company.

As is clear, we are very pleased with the results of this quarter, a great credit to our employees worldwide, manufacturing, business units, sales, logistics, for close to pristine execution. This was another record quarter for us of ongoing revenue growth and substantial improvements in our profitability.

This is the 11th quarter of year-over-year revenue growth, revenues being at an all-time high, combined with record gross and operating profit, leading to record EBITDA of a close to $400 million annualized run rate, a strong net profit of about $200 million annualized run rate, with corresponding growth in free cash flow, demonstrating that our business model, including acquisitions that provide immediate ROI with long-term guarantees and growth potential, is working and is being well executed.

We continue to see strong demand in our factories, which has driven high utilization in our core fabs, which was foreseen and forecasted and the driving reason for the creation of TPSCo and buying the San Antonio facility from Maxim. Both Fab 2 in Migdal Haemek and Fab 3 in Newport Beach had increases in capacity over the quarter and both fabs are running at high utilization levels of about 90%. In addition, we're adding additional capacity of about 5% to Fab 2, which will be completed during the first quarter of 2017. Utilization in Fab 1 in Migdal Haemek remained at about 70% in the quarter.

At TowerJazz Panasonic Semiconductor, all three fabs are shipping increased diversified third-party products of analog and analog power devices, in addition to a strongly diversified set of applications to Panasonic [self].

Specific to our third-party business, at Tonami we are shipping power management, analog CMOS, and power MOS. At Arai 8-inch 110-nanometer, 130-nanometer factory, we are shipping RFSOI, CMOS image sensor, including the smallest available pixel for global shutter industrial cameras, and we are in advanced qualification of a very high-volume potential embedded non-volatile memory for an existing high-volume customer.

This is a very interesting activity to note in and of itself. This customer is a long-term customer with which we have developed a strong relationship. In one of our regular roadmap reviews, bringing up certain capabilities of our Arai factory, this NVM project was triggered, fitting the needs of a business that they had recently acquired.

Consolidation in our industry allows us to explore opportunities in many areas with customers that we did not envision at the outset. To date, we have benefited from most all, if not all, of the announced IDM/IDM or IDM fabless consolidations.

In Uozu, the 12-inch 45-nanometer, 65-nanometer factory, we are shipping advanced CIS products and prototyping RF-SOI.

Utilization at TPSCo was 50% in the quarter, a continued improvement of 8 points, or 20%, versus the 42% utilization that we reported last quarter, with all fabs showing utilization increases.

At the beginning of 2015, we gave a target that we would hit a $100 million annualized third-party revenue run rate by the end of 2016 in TPSCo. Being now in the fourth quarter, we are pleased to note that we are on track to hit this target.

With regard to San Antonio, our 8-inch factory which we acquired from Maxim and began operating earlier this year, we continue to qualify multiple RF and power discrete flows and we taped out a growing number of new customer products in this past quarter.

As you saw in today's press release, our fourth-quarter midrange guidance is $340 million. This represents continued strong sequential growth, with 12 consecutive quarters of year-over-year growth.

As we draw near to the end of 2016, all indications are for our strength to continue into 2017 and beyond. Our ongoing performance is based on continued strong customer relationships and demand. This is a verification of the value of our long-term customer partnerships, and as well our effectiveness in realizing customer projects within our customer funnel with newer growing customers, multiple of which have grown from single-digit million revenues in 2015 to double-digit in 2016, of which several will break $20 million in 2017, or customers that are single-digit millions in 2016 and forecasted to break double-digit revenue in 2017.

I now want to spend a few moments talking about the performance and trends we are seeing at our various business units. In each business unit, our broad and advanced technology offering provide differentiation in each of the end markets that we are participating in.

Our RF high-precision analog business unit realized multiple significant technical milestones and customer roadmap activities during the third quarter. We ramped to production volumes our [QT8] flow with multiple customers in Migdal Haemek Fab 2. QT8 is our latest 200-millimeter RF-SOI technology with our best production Ron-Coff figure of merit of 124 femtosecond, which provides our customers a roadmap to lower insertion loss products.

We released design kits for our next-generation RF-SOI process, [QT9], to lead customers. QT9 has an Ron-Coff of about 100 femtosecond, providing an incremental 20% lower insertion loss than the QT8 production technology.

We delivered silicon prototypes using our [CS] QT8 technology to multiple lead customers. CS QT8 combines our production best-in-class QT8 switch with a state-of-the-art low noise amplifier for products that will combine these two functions on a single die.

We released a process design kit for new RF-SOI technology for our 300-millimeter facility in TPSCo that adds a best-in-class low noise amplifier to our previously announced 90 femtosecond Ron-Coff switch technology, for which we have already shipped customer samples.

In addition, multiple customers are planning a Q1 2017 tapeout into this new version of the process, where we combined this 90 femtosecond switch and an LNA into a single die, greatly reducing their cost and increasing our margins.

We continue to progress a roadmap with next-generation silicon germanium technology for the fiber-optic market, for which we have about 60% overall market share. We released design kits for our S4 process, a 300 gigahertz speed, which we press announced last quarter, and are now working with lead customers on our next-generation S5 process.

Part of our CapEx investments this year was to bring up a copper back end of line in Newport Beach specifically to accelerate the RF silicon germanium roadmap, as well as for cost qualification of our RF-SOI, initially having been transferred and qualified from Newport Beach to Migdal Haemek, and now taking the advanced flows, which the RF business unit developed in Migdal Haemek with copper back end, and qualifying them to Newport Beach, enabling our customers and ourselves manufacturing flexibility in order to maximize output capacity.

Lastly, we accepted additional tapeins of new RF-SOI parts into our San Antonio facility, in addition to continuing the qualification of the existing RF-SOI parts that had been previously transferred and taped out in that factory.

We continue to see a very high demand in our CMOS image sensor business unit and expect continuous increase next quarter and throughout 2017. Excluding Panasonic, CIS is expected to exceed the corporate CAGR with approximately 40% year-over-year growth. This is, as reported last quarter, a continuing increase in our customers' market share and introductions of new products to the market. This is true in all of our supported market segments, but especially true in the machine vision and the medical x-ray market segments.

The machine vision segment is growing very fast since it is a diversified market from an application point of view, including traffic control systems, barcode reading, also then extending to food control automation. Every factory, especially electronic ones, are becoming equipped with a very high amount of smart cameras that automatically control production quality. There is a continuous increase in the amount of cameras, but also in the resolution and sensor size.

We continue to enjoy the fruits of our excellent global shutter technology developed in our 0.18 micron technology node and just recently released the availability of our next-generation global shutter pixels on our 110-nanometer technology node. This is the smallest pixel in the world for this industrial vision technology, and we have several products designed by numerous customers using it. We expect this technology to gradually replace our existing one and with a substantial increase in volumes.

In the medical market segment, we have continued to see very nice revenue growth and, as well, a substantial amount of new projects. Understand, a new project in this area historically has a minimum eight-year lifetime, but can extend to 15 years and beyond. The sensors that have been developed by our customers to serve the medical/dental markets are now being requested by a different customer set as an off-the-shelf part for industrial markets, namely for [cellular] diagnostics and prevention.

In our 12-inch factory, we are ramping to production several products now to the high-end photography, as well as to the security and automotive camera markets.

In a separate note, camera sensors manufactured by us have helped our end customers to win multiple Oscar awards for cinematography. That being stated, this past quarter we won a very, very large project in the area of high-end photography. We continue to see a nice stream of design wins to all our CIS fabs, namely Fab 2, Migdal Haemek, Israel; Fab 6 and 7 and -- I'm sorry, and the 12-inch factory in Arai and Uozu; and aerospace and defense and commercial infrared cameras in Newport Beach.

Power continues to gain momentum with new platforms. We have talked in previous quarters about our Gen 4 flows ramping to high volume of production with industry best-in-class RDS(on) figure of merit. We have additionally driven a lower cost platform where we have reduced the number of layers, allowing the customers and ourselves to share increased margin without any degradation in performance.

For our TOPS business, we continue to grow with customers such as Infineon; Fairchild, now ON Semi, also a long-term customer itself; and Vishay, among others, including the Maxim activities.

Additionally, we are driving a new and exciting project within this group. One of them is the gallium nitride project, which we expect you will hear more about during the next few quarters, a novel and strongly differentiated approach to gallium nitride.

In summary, we are very pleased with our strong and record results in this third quarter. The execution of our analog specialty business model remains solid, resulting in revenues at all-time high, with all business units preventing -- presenting strong growth of double digits, ranging from 15% up to 40%. Our revenues, combined with the gross and operating margins increase, led to record EBITDA and very strong net profit, with corresponding growth in free cash flow.

We completed this year's TowerJazz Technical Global Symposiums, having held symposiums in China, the United States, and Japan. The reception was excellent. In China, we had over 140 attendees, representing approximately 70 existing and potential customers. In the US, we had approximately 150 attendees, representing over 60 existing and potential customers, and 135 people attending in Japan, representing 75 existing and potential customers. The feedback was excellent for all of these forums.

With that, I would like to turn the call to our CFO, Mr. Oren Shirazi. Oren, please.

Thank you, Russell, and welcome, everyone. Thank you for joining us today.

We are very pleased with the outstanding results we released for the third quarter of 2016, achieving a record EBITDA of $97 million in the quarter, a record cash flow from operations of $86 million, as well as $51 million in net profit, which resulted in $0.58 per share on a basic and $0.52 diluted earnings per share. All these contributed to our very strong balance sheet, with shareholders' equity reaching a record level of $636 million.

I will start by providing our high-level P&L analysis for the third quarter and nine months of 2016, and then discuss our balance sheet and cash flow. To validate the viability of our growth model and demonstrate how a large portion of incremental revenue goes to the bottom line, I would like to present an analysis of the incremental EBITDA and net profit margins increase in the past quarters and years as resulted from the corresponding revenue increase.

Comparing our third-quarter 2016 results to the same quarter in 2015, we see $34 million and $38 million of incremental EBIT and net profit, respectively, as compared to $[82] million incremental revenue, reflecting a 41% and 46% incremental EBITDA and net profit margin, respectively.

Comparing our (technical difficulty) quarter 2016 results to the third quarter in 2014 (sic - see Press Release - "third quarter of 2015"), we see a similar picture. $60 million and $71 million incremental EBITDA and net profit, respectively, as compared to $100 million in incremental revenue, reflecting a 60% and 71% incremental EBITDA and net profit margins, respectively.

Comparing our year-to-date 2016 results to the same period in 2015, we see $89 million and $207 million (technical difficulty) EBITDA and net profit, respectively, as compared to $203 million incremental revenue, reflecting 44% and 102% incremental EBITDA and net profit margins, respectively.

And lastly, comparing our nine months year-to-date 2016 results to the same period in 2014 (sic - see Press Release - "first nine months of 2015"), we see $164 million and $[152] million incremental EBITDA and net profit, respectively, as compared to $317 million incremental revenue, reflecting 52% and 48% incremental EBITDA and net profit margins, respectively.

This analysis demonstrates that consistently a large portion of incremental revenue goes to the bottom line and validates the viability of our growth model.

And now, I'll focus more on Q3 P&L details. Revenues for the quarter were a record of $326 million, compared to $244 million in the third quarter of 2015, an increase of 34% year over year and 7% as compared with $305 million in the immediately preceding quarter.

Gross profit for the third quarter of 2016 was $81 million, reflecting an increase of 47%, as compared to $55 million in the third quarter of 2015 and an increase of 12% as compared to $73 million in the immediately preceding quarter.

Operating profit was $49 million for the third quarter of 2016, with a 104% increase, as compared to $24 million in the third quarter of 2015 and a 22% increase as compared to $40 million in the immediately preceding quarter.

Net profit for the third quarter of 2016 was $51 million or $0.58 of basic earnings per share, demonstrating increased sustainable net (technical difficulty) as compared to $14 million or $0.18 per basic earnings per share in the third quarter of 2015 or $38 million and $0.45 of basic earnings per share in the immediately preceding quarter.

EBITDA for the quarter was $97 million, reflecting a 54% increase as compared to $63 million in the third quarter of 2015 and an 11% and sequential increase as compared to $87 million in the immediately preceding quarter.

The results for the first nine months of 2016 also demonstrate the strong revenue growth, profitability, margins, and cash flow. Revenues for the 2016 first nine months were a record of $909 million, reflecting 29% growth as compared to $706 million in the first nine months of 2015.

Gross profit for the first nine months of 2016 was $215 million, reflecting a 53% increase as compared to $141 million in the first nine months of 2015.

This improvement in margins demonstrate a successful business and operational activities performed by the Company, including improving the revenue mix by the cross-qualification of [difficult] technology platforms in our worldwide fabs, followed by mass manufacture of specific process flows and product. This enables operational flexibility and efficiency in fab utilization, allowing better operational performance and increased average selling prices and increased margins.

Operating profit for the first nine months was $120 million, reflecting a 152% increase, as compared to $48 million in the first nine months of 2015.

EBITDA for the first nine months totaled $261 million, representing a 51% increase, as compared to $173 million in the first nine months of 2015.

Net profit for the first nine months of 2016 was $156 million or $1.81 basic earnings per share and $1.61 diluted earnings per share and included $51 million gain from the San Antonio fab acquisition.

I will now go over the balance-sheet analysis as of the end of this quarter. Shareholders' equity as of September 30, 2016, reached a record of $636 million, an increase of 65% as compared to $386 million as of December 31, 2015, and a 14% increase as compared to $559 million as of June 30, 2016. Current ratio increased to 2.32X, as compared to 2.12X as of December 31, 2015.

Share count as of September 30, 2016, included 90 million outstanding shares. Fully diluted share count as of September 30 (technical difficulty) included an additional 17 million maximum possible shares, comprised as follows: [3] million from possible exercise of warrant Series 9; 5 million ESOP-related options and RSUs; [$]6 million underlying convertible bonds, and [$]3 million underlying capital note.

The net debt as of September 30, 2016, amounted to $16 million, as compared to $51 million as of June 30, 2016, and as (technical difficulty) $105 million at the end of 2015.

With regards to our cash flow report, cash and short-term deposits on September 30, 2016, were $363 million, as compared to $311 million as of June 30, 2016. The main cash activities during the quarter were comprised of the following. $86 million cash generated from operations; $22 million received from exercise of warrant options; $9 million debt received net of debt repayment; and $55 million invested in fixed asset CapEx (inaudible).

Free cash flow for the third quarter of 2016 was $31 million positive, as compared with $27 million positive in the second quarter of 2016 and $10 million positive in the third quarter of 2015.

And now, I wish to turn the call to Noit Levi. Noit, please go ahead.

Thank you, Oren.

Before I will open up the call to the Q&A session, I would like now to add a general and legal statement to our results in regards to statements made and to be made during this call.

Please note that the third quarter of 2016 financial results have been prepared in accordance with US GAAP and the financial tables in today earnings release include financial information that may be considered adjusted financial measures and non-GAAP financial measures under Regulation G and related reporting requirements as established with the Securities and Exchange Commission that apply to our Company. Namely, this release also presented financial data, which is reconciled as indicated in the tables or in the call on an adjusted basis after deducting, one, amortization and (inaudible) intangible assets; two, [due] to compensation expenses in respect with equity grants with directors, officers, and the employees; three, gain from acquisition, net; four, non-cash balancing expenses related to bank [log] early repayment; and, five, other nonrecurring items, such as income tax benefit.

Adjusted financial measures and non-GAAP financial measures should be evaluated in conjunction with and are not substitutes for GAAP financial measures. The tables in the earnings release also contain the comparable GAAP financial measures to the adjusted financial measures, as well as there is a reconciliation between the adjusted financial measures and the most comparable GAAP financial measure.

EBITDA is reconciled in the tables from GAAP operating profit. EBITDA is not a required GAAP financial measure and may not be comparable to a similarly titled measure employed by other companies. EBITDA, the adjusted financial measures, and the non-GAAP financial information presented herein should not be considered in isolation or as a substitute for operating income; net income or loss; [capital] provided by operating, investing, and financing activities; per share (inaudible) data or other income of cash flow statements that are prepared in accordance with GAAP and is not necessarily calculated or presented on a basis consistent with the same or similar data presented in previous communications.

And now, we will open up the call for Q&A. Operator?

(Operator Instructions). Cody Acree, Drexel Hamilton.

Thanks for taking my questions and congratulations on the steady progress. Maybe, Russell, if you could, could you just talk about your degree of visibility as we head into the December quarter and then maybe how that visibility extends into 2017?

December quarter, you mean the Q4?

Yes, sorry.

So we guided, I think, a nice increase for the quarter, going up to $340 million, a 1.36 annualized revenue. Our visibility there, I think we are very predictable at hitting our guidance, so we gave a, I think, nice growth guidance that we feel confident that we will be able to hit.

We had presented, I think, last quarter the fact that we have at the $1.2 billion run rate, that about two-thirds of that $1.2 billion was under special models, be it the Panasonic or the San Antonio long-term contracts, be it pre-payments from customers that is then amortized over time, be it the TOPS businesses that are multiple take-or-pay arrangements. So, I think our business models give us maybe a bit better visibility mid and long term than other foundries might have, who really work off of quarter-by-quarter purchase orders only.

So going into 2017, I did state that all that we see would indicate continued strength. We haven't given a target for 2017 and typically we wouldn't refer to that until we release Q4, but from everything that we see right now, 2017 would look a continuation of strength for the Company, with multiple new activities coming into play that we have not discussed yet as well and a continuation of the strong growth that we've had to date.

Our CAGR over the past years has been varying between 30% and 35%, so very, very strong growth rate, and obviously that shows activities that if they continue, which we think that they should continue, feed stronger midterm and long-term growth. So I think our visibility is good by virtue of our business models and what we hear from customers. We see continued strength. Protracted answer, maybe, but hopefully it answered your question.

It did. Thank you, Russell. Last year, going to the March quarter, you were able to offset normal seasonality with the strength of your underlying businesses or your share gains. Does your visibility extend in the March period and, I guess, how are you thinking about normal seasonality into March, or is there such a thing any longer?

We haven't guided for Q1 and we won't guide Q1 until Q4.

I did state that 2017 looks strong. You don't have the type CAGR that we have been having without having growth in market share. The only thing that can stem seasonality is an increase in market share, so obviously our market share is growing and that is the only way you can stem seasonality, providing that Q1 will be down in revenue, which it typically is versus Q4. But I think our market-share growth would give us the drive and capabilities to stem the seasonality should it be a seasonal quarter, but more than that we haven't guided yet and I don't want to speculate on Q1 at this point.

That's great. Thank you. And then, I guess, maybe over the next couple quarters or maybe even just into the December period, any delineation as to the driver of that $340 million or the growth that you are seeing sequentially, if you can help us, whether that be CIS or RF or power management?

I stated in the script that year to date and nominally the numbers I gave would be our estimation of the year that every business unit that we have has seen at least a 15% year-over-year growth, with some of them increasing up to 40%.

Now you could even say higher than 40% is the business that we have grown through extending our relationship with Maxim. We include that really into our TOPS business, but I did not include that into the numbers of the -- of organic growth and a business unit of 40%.

But the numbers I gave are -- the 30% to 35% CAGR, depending on where we lined up in the Q4 revenue, it is pretty well distributed among all of our businesses. The biggest as far as the amount of dollars of growth was within the RF space on absolute dollars. Probably the biggest as far as percentage growth was in CIS.

And then last for me, just from a capacity growth standpoint as you look beyond your needs for 2017, do you believe that there will be sufficient opportunities for you to replicate the current model where your customers are covering your fixed costs with your long-term -- with these long-term supply agreements? Do you believe that there are enough opportunities in the market for this to be a continuing ongoing business model?

If you are talking about an operational acquisition where we would buy something with the idea of having free capacity on top of it -- on top of the acquisition, where the acquisition itself covers the fixed cost, I believe that those opportunities exist, definitely.

There is additional models that we've talked about separate from that, which is getting involved in enabling a greenfield site, for example possibly in China, to where we would be giving our operational capability towards the establishment of a facility, maybe acting as general contractor for the facility, bringing up technologies, operating that, and hence being able to be part of greenfield that typically is $1 billion plus investments to where from day one it would be accretive to us, not a downside on our cash, but giving us operational capability at cost, cost minus, or a slight cost plus, be it what it may. So I think that those opportunities exist as well.

So there is two different models. One is, again, the ability to take on a going concern where maybe the customers' utilization levels are going down over time, and I think that they exist and readily exist. For us, it was always a question not necessarily -- I can't say always, but the past years it has been a question more of digesting what we have and showing that the models work than taking advantage of all the opportunities that are on our plate. So I can say that over the past years we have had more opportunities in that regard than we have wanted to absorb.

Now I think we are very smart in what we did absorb. The Panasonic agreement was a very, very large agreement for us, and without trying to sound too self-aggrandizing, but within two quarters of that arrangement having occurred or three quarters of it to give a target of hitting $100 million annualized run rate of third-party revenue by the end of 2016 and to be on track to that, having stated so seven, eight quarters ago, I think we have executed well.

And that type of a very, very high-volume capability that you're bringing in a lot of capacity, over 600,000 wafer a year of unused capacity, to be able to grow already and have, I think, a reasonable understanding that we can bring that up to a $200 million annualized run rate in 2017, that's a very, very strong verification of where we are at and what we are doing.

Having taken on such a big activity as that and executing on it I think gives everyone, our Board of Directors, as well as another large company that might want to enter into an agreement with us, as well as ourselves, a model and confidence that we can make such a big acquisition work and at the same time very much serving the partner that we took the majority ownership from, in this case having been Panasonic.

So, there are certainly other opportunities like that, and early on we walked away from one that was offered us because it would've been too big for us at that time to want to take on two big operations of that sort. Now we have a proof model. We have that going, and then what we did with the -- with a very strong and good customer, Maxim, with San Antonio, it has been seamless.

So there are other opportunities like that in that area and, as well, a model to go into a greenfield where we are taking advantage of our capabilities so that we are not investing the $1 billion plus that is certainly a negative on a P&L as you are absorbing the cost and trying to drive utilization levels, but to be a partner in that where we are not an equity owner or at least an equity owner that has to consolidate anything, but has the benefit to get paid for our capabilities from the earliest stages and then have a capacity available to us in a region we might want to play in at a cost minus, cost, or cost plus, I think that those models are certainly available as well. Again, a long answer, but I hopefully believe it is -- well, I do believe it is a very complete one.

Very good. Thank you and congratulations.

Rajvindra Gill, Needham & Company.

Thanks and congratulations on those great metrics. On the guidance for revenue and then also on the gross profit fall-through, in the past you had talked about a 50% incremental gross profit fall-through. Should we see that also in Q4?

For the gross margin?

Yes, yes, long term we talked about between 50% to 60%, so average of 55%. Should continue to see that, yes. This is why I started my reading with explaining -- showing how in the past we demonstrated actually it was between (inaudible) 70% incremental margin, 2015 against -- 2016 against 2015, 2016 against 2014, all those metrics were between 41% to 70%, so we should expect for the future 50% to 60%, yes.

Okay, so then, roughly, that would imply that gross margins would improve maybe around 100 to 130 basis points quarter over quarter [to] something like 27%?

Whatever you calculate based on the 50% to 60%.

Okay, so in terms of the revenue growth that you are seeing, I think last time you break out [potential] of sales by RF, power management, image sensor, and other. I was wondering if you have those metrics as well and if we could talk about specifically what you're seeing in the RF business. You mentioned the new products with a low noise amplifier and switch on a single die, but I was wondering if you could talk a little more about some of the secular trends you are seeing in the RF business, as well as maybe give those metrics on a percentage basis, if you have those available.

We certainly have them, but what we did state is that each year that we would present on the Q4 the breakdown of revenue as a total of the business units, and we plan on doing that in Q4. We are not doing it on a quarterly basis, nor is it necessarily as interesting on a quarterly basis.

So with the Q4 release, as we did last year, we will give a breakdown of revenue per business unit, what that business unit did in growth, and what we'd expect we would have in growth in the next year. As far as the specific trends within the RF, what is your question there?

So you're clearly gaining some share in the RF business with your new products, but beyond the market-share gains in terms of the market demand for higher frequency bands for things like carrier aggregation that are pushing more frequency bands in the market, I was wondering if you could characterize the demand landscape in the RF power management -- power amplifier market.

So the front-end module market is made up of multiple chips. You have the switch, you have the power amplifiers, you have the power amplifier control, you have the low noise amplifiers, et cetera.

We have done very nicely this year with a silicon germanium based power amplifier specifically for Wi-Fi, and that has been a very, very big growth engine for us, so that was a new application we entered into that I think we benefited from right now.

The ability to work off of a silicon germanium RF CMOS platform and combine multiple functionalities into a single chip certainly has [a bit] for our customers as it reduces the packaging cost of the individual chips that otherwise have to be put into a module, and I think that trend continues. So we're working very closely. What we announced on the combined switch LNA is one such avenue that customers are pulling for strongly and, again, it reduces their ultimate cost.

But in that area, we see great strength. Certainly the amount of front-end modules has grown very, very strongly with going from one or two to right now, I think, at present generation, we're at some nine front-end modules for mobile platform and maybe growing beyond that. So if that continues in that trend, I think that that is one of the beauties of that area right now is that even if the smartphone volumes would somewhat stabilize, the content is continually going up.

Okay, got it. And on the power management side, wonder if you could maybe talk about what you are seeing on the market trends there, specifically the automotive or the industrial end markets?

As noteworthy based on your --

Anything that is noteworthy based on your customers and the business trend this quarter.

So overall, power management has, I think, increased demand and interest from everyone. Every system that there is has power management chips.

For IoT to really be facilitated, it definitely necessitates low-power consumption either from the net or from batteries, and we see that certainly in the area of white goods. The necessity to reduce and as well put a sticker on every white good that is sold as far as the power consumption continues, so the combination of discretes and power ICs and the ability to make them at the lowest RDS(on) figures continues.

And we have roadmaps with all of our customers within the discretes that drive -- with most all of our customers, at any rate -- that drives a continuation of a lower RDS(on) the discrete side, and we drive ourselves the RDS(on) on the power IC, having data that we have this [on] Gen 4 flow of the 10 [milliomes] square millimeter and going to a -- our Gen 5 of the 8.5 that I believe is seriously the very best figure of merit not just of foundries, but of everybody in the industry.

So, that continues. If you ask about automotive, we have had very strong demand. Unfortunately, the customer that we are dealing with this on, for a variety of reasons, is not press releasing presently, but believe that we will press release with them, and that's in the area of electric cars. And the ability for power management within electric cars is very substantial, so that's a big growth driver for us within the power management of very high-end capabilities.

So if you ask specific on automotive, that's where we see a big demand right now. Our larger automotive demand, though, right now that we know about into automotive -- sorry, let me just add one other thing. We make a lot of power management devices in Panasonic itself that is sold very strongly into automotive. That is our capability, what we are selling into Panasonic and Panasonic's diversified end products. And we don't give the specific numbers of what that is, but it's (technical difficulty) a lot of the power management that we do there is in automotive.

But from non-Panasonic, a lot of our automotive growth right now and demand is really in the area of doing driver notification, driving into autonomous vehicles, and that deals with millimeterwave, so radar for collision avoidance and as well as cameras. And so, that's right now what we see very, very strong demand from high-end integrators or the end customer itself deals with the, at this time, driver notification, but really driving -- driving in automotive sounds odd.

But right now, it is driver notification and then migrating in the next-generation autonomous vehicle, and that would be the area using sensors, be it in the area of RF or be it in the area of cameras, but then really using sensors to drive an autonomous vehicle ecosystem.

And your manufacturing process can support all different types of sensors, whether it is camera based, radar based, or LiDAR based?

All three, correct.

Okay, got it. And just last question on the -- from a follow-up. In terms of now that the balance sheet is looking good and the business model has played out, is playing out well, are there specific end markets where you would try to target and employ a similar business model that you did with Panasonic and Maxim where you're getting incremental capacity at limited CapEx, basically buying underutilized fabs? Is there a particular market or a part of the world where you would pursue that strategy going forward or do you think you have enough capacity right now just to do that?

I didn't quite 100% understand the question. If you just reword it, and I will --

What I'm basically saying is the business model that employed has been working with Panasonic and Maxim. And my question is, now that the balance sheet is in a good condition, are there other end markets, are there other places in the world where you want to employ a similar business model where you get the capacity at (technical difficulty) CapEx, very limited CapEx, and you can grow your capacity and grow your revenue base? Is there other areas, other markets that you're going to pursue a similar strategy or are we just -- we have enough capacity at this point?

We have enough capacity at this point for this point, be it through a greenfield type of capability, as I had mentioned, or be it through another (technical difficulty) deal, as we did with Maxim or with Panasonic.

The 30% CAGR, $1.6 billion is what we said is our present pure operational capacity, that is not going with NRE, it is not dealing with activities that we do in TOPS where we sell capability elsewhere, but the $1.6 billion, providing that we maintain present growth rates, that's good through 2018, possibly.

Right.

So, certainly we are looking and we're actively looking at pursuing other such deals for capacity expansion.

Now if you ask about -- and that's just in general about capacity expansion. What are the areas that we are very, very interested in? We're certainly extremely interested in the sensor area, and there are certain types of sensors -- if you look, for example, at MEMS, to be a major player in MEMS -- I am not talking about having a contract with one person that you think will be big or so, but really to be a major player in MEMS, you have to vet multiple, multiple customers.

Different with MEMS than with other specific flows. If you have a power management flow, we have PDKs. Our RF flows have PDKs, process design kits. And our power management is a modular platform, so it is -- I'm not saying it is easy for the factory, but it is not difficult for the factory to run prototypes because there is not tremendous customization.

MEMS is not that way. MEMS requires tremendous customization. So it is very possible that we would go after a partnership with somebody that maybe already has a MEMS fab, or to go after a possible acquisition in the MEMS area in order to vet many, many customers that allow us to take many, many (inaudible) customers, and those that then start growing as they get to 500 or more to transfer certain capabilities to higher volume factories that allows you then to benefit from the volume that you have without diverting the capability of the factories to the vetting of many customers that need extremely customized process flows.

Hopefully that answers your question, but very short answer, the specific area that we would focus on is the area of sensors, and there is all types of sensors that can be made. We are very, very (technical difficulty) right now in magnetic sensors, an activity that we had done with Crocus that has been press released multiple times, but we have good capability in magnetic sensors, but there is a variety of other capabilities that we think are critical and very advanced that would require MEMS capabilities. And to really grow within the MEMS front, as well as maybe some other sensors, you would need some very specific manufacturing capability to vet many customers.

Great. That's very helpful. Thank you.

Richard Shannon, Craig-Hallum.

Congratulations on the excellent execution here. Keep up the good work. I apologize, I am in an airport, so it might be a little bit loud here and I might have missed a few comments here as I've been traipsing through the airport. But I will go ahead with the questions here.

I wanted to follow up on one of the last, most recent questions here. I think, Russell, you mentioned a maximum revenue capability of $1.6 billion annually. Did that include the capacity additions you are making in San Antonio?

No, that would be no additions. San Antonio could increase in capacity, but we have not talked about that. We have not released that, nor have we actually began to execute a capacity increase in San Antonio.

I think what you are referring to, in San Antonio there is a separate building that now has a few implanters in it that could really be built out. It is a separate [cleanroom], as well as in the main facility. It would not be difficult to extend present gray area into white area.

But that does not -- that is not included in what we are talking about now. What I am talking about now is the present not just facilitized, but equipped, white-area capability.

Got it, okay, that's helpful. Thank you.

The second question is, Russell, in response to one of the earlier questions, you made an interesting comment. I would love for you to expand on that across all of your relevant business areas, but you talked about your overall very strong growth being driven not only by market growth that you are in, but also market-share gains.

Wondering if you can go through the major segments that you have (technical difficulty) where, to the extent to which you believe you have gained share, and how does the ability look forward to gaining more share? I know in some cases you are getting to a higher share. It may be more difficult to gain even more, but it would be great to hear about your thoughts going forward there on ability to gain more share in your target markets?

Okay, so twofold. I had stated that we have announced on an annual basis, meaning every fourth quarter, that we will talk about the specific volumes and growth within each of our business units as a consolidated company. So we will get into that in Q4, but just in short to answer your question and not be avoidal and I stated in the call.

We had very big market-share increase within CMOS image sensor. We stated [$0.40]. This is of the organic capabilities, excluding the Panasonic, and we expect to continue a very, very high market-share growth in CMOS image sensor.

The biggest dollar growth has been within the RF space. It is not the highest percentage growth in market share, but it is nicely -- a nice double-digit number. Will that continue in these rates? We would expect that we will continue to see very strong growth within the overall RF space.

We have multiple very strong platforms within the infrastructure. Had mentioned this S4, working on S5, and that we presently have a 60% market share. The amount of data traffic is increasing vastly. If we just were able to maintain, not even grow, market share, we will benefit from that very strongly, just as a function of better traffic.

And then, you have the whole area of being able to integrate more devices, front-end module devices, onto single chips, which is moving into the SiGe type of a platform to have certain power amplifier capabilities. But I think that we have very nice opportunities to grow there as well.

Within our TOPS business, I had mentioned very briefly, and I think it is the first time we mentioned it, that we have taken on a very big project with a partner with gallium nitride that has within it a very, very novel construction that we think will be a best-of-class foundry offering. That has very, very long-term growth capabilities within power, not just in the area -- I mean, really a big movement from planar and trench MOSFETs into the capabilities of the gallium nitride platform.

And the potential there, it is really non-forecastable, meaning it can be extremely large, but it will take a bit of time to unseed a very, very strong rooting presently of MOSFETs and moving that into discrete [gallium] units, but I think it's very possible that will happen. And our platform there is extremely strong.

So within our TOPS business unit, both in the development of working on very unique processes with certain customers or with certain partners, as well as just the TOPS business itself, it continues to grow very, very strongly.

What else? The power management, we would see that the power management mixed-signal probably will have from 2016 to 2017 at least a 50%, 60% revenue growth. And that's a triple digit revenue as it presently sits.

So, if that gives you a flavor of the different activities we are doing. Power management itself, power ICs, very strong growth there. Power discretes is growing very nicely, and if you add to that a GaN platform, we will grow even more nicely. I can't state that's a 2017 growth. In fact, I don't think it is, but it is a platform that I believe will be the best platform of all foundry offerings, and as that gains traction, I think we will be in a very unique position to benefit from very strong market-share growth.

Now if you look at the fact that it is zero revenue right now, it is an infinite growth, but I don't -- on the dollars, it becomes a different story. But I believe that it will probably start gaining traction in the 2018 time frame, and then we will get more and more updates as the market evolves and develops as to what that turns into.

I am trying to think if there is anything else (multiple speakers) I have forgotten. I don't think so.

I think you have covered everything, at least that I could think of, Russell, so thanks for that great detail. My last quick question is a follow-up on one of those comments on the GaN project you are on here. This sounds like it is related to the power market and not the RF for wireless market. Is that correct?

Presently, we are focused on the power market. It doesn't mean that it couldn't be brought into the RF market, but it's not our present focus.

Okay, great. I will look forward to hearing about that more in the next few quarters. And that's all my questions, guys. Thank you.

Lisa Thompson, Zacks Investment Research.

Just two quick things. One was on the gallium nitride project, could you give us a hint as to what kind of end users might be interested in that sort of product?

All of the big power players could be interested in a project like that, without overstating. Look, you know who the power players are. They have -- our customers, they exist right now, is a good base of power players. You have within that that are non-semi. You have a Maxim, you have Vishay, you have Infineon, et cetera, et cetera, et cetera.

Now Infineon itself has very good GaN flows and two different types of GaN flows, but the whole power area would be very interested, as well as an enabler for fabless companies that maybe will be coming up and would wish to have something that differentiates them as an entry into the market. But I believe that the entire power market is interested in gallium nitride.

And is there any particular end-user product that it is going to be particularly useful for or better for?

It is across the board. GaN itself promises -- it is a much smaller device as far as the capabilities it has to withstand higher heats and be able to produce the same output parameters.

Okay, and then just quickly on your (technical difficulty)

I'm sorry, please? Hello? Lisa, you got cut off. We couldn't hear your question. Operator, is everything okay?

Cody Acree, Drexel Hamilton.

Just a quick follow-up. On CapEx, you mentioned a 5% increase in Fab 2. Was that correct in your comments?

I didn't say anything about CapEx. I talked about capacity.

Right, I guess I am just questioning how that would flow into your CapEx expectations for Q4, with that being -- you had talked about previously a baseline in the $40 million-ish per quarter. Does that 5% bump that number up somewhat in Q4? What are your expectations, I guess, and then into 2017, how long before we would get back to that $40 million-ish baseline CapEx (technical difficulty) number?

There is no change in any of the forecasts we gave. The capacity increase of 5%, as Russell mentioned, was already in the model. It was already in the CapEx number that we paid and they ordered previously, so already, for example, Q2, Q3 we had payments towards that capacity increase. Until the tools are arrived and installed, like Russell mentioned, Q1 2017, it takes time, but the payments are already in the model, so there is no any additional CapEx from this discussion.

And you are correct that we forecasted that from Q4 2016 and beyond that for the quarters of 2017, we will go down from the current levels of $55 million to about $40 million-ish, like you mentioned, so $40 million to $45 million a quarter, certainly not more than that from this quarter already we will achieve it.

Perfect, and then, lastly, Russell, you alluded to this a couple of times in some of your comments about maximizing -- optimizing your manufacturing, balancing your loads, I guess, among your different facilities. With you just ramping into San Antonio, I think you had expected first revenue in Q4, freeing up maybe some of that -- the capacity in Newport Beach maybe for silicon germanium and some of the CIS capacity in Israel.

As we look into 2017, is that really when we start to see the impact financially of some of all of this transition work that you have done through 2016 or is that already starting to flow through the model?

I think that you would see it in 2017. It's not that it wouldn't have some small impact, but probably really not see it in Q4. But 2017 should start to see the benefits of the increased usage of capacity at San Antonio and hence incremental growth in the Company by using that capacity and being able to free up other capacities elsewhere for maybe things that are higher margin.

All right, very good. Thanks [to all].

Thank you, Cody. Very good questions.

Graham Tanaka, Tanaka Capital Management.

Congratulations, nice quarter. Just wanted to understand if there is any change in average prices. I am not sure how you price your -- per wafer or whatever, but is there any kind of inflationary or deflationary trend and is that going to change in the future? Thanks. Is (multiple speakers) it's getting richer or not?

Basically, we see a very modest increase in average selling price, but it is mainly not because of any market trends. It is because of our business focus that we are -- the cost qualification we do. So, we are -- we, as we said, cross-qualified the EPS (inaudible) fab and also -- mainly Tonami fab and also San Antonio is now in cross-qualification, and also Fab 2 was cross-qualified to the Fab 3 Newport Beach product, so this is resulting in a fact that we are able to accept more higher-margin business to fabs that were previously overutilized, and this is increasing the average selling price. Of course, improve the margins and profitability. But it is modest. It is not like 20% increase. It's very modest, but it is increasing; it is not going down.

It certainly stems the tide of year-over-year ASP reduction. And that's one of the big areas and focuses that we had about being able to cross-qualify in order to have -- and I think Cody referred to it, in order to really optimize the mix at any given factory.

The worst thing that you can have is to gate a high-margin product because of a demand and a customer relationship at a lower margin product. So the manufacturing, the worldwide manufacturing flexibility becomes very critical in order to maximize at any given quarter a local mix.

Just wondering what the cost save is for some of your customers in that higher growth area where you are combining functionality onto one chip? What is their cost save, and how much of that are you -- how much are you able to capture as a lift to ASPs or margins (inaudible)

I honestly think that's a question that you should ask the specific end customers that we have, what they would be saving on it. I don't have a specific number to start with, and [generally] I don't, but I don't think it's really for me to talk about a consolidation of their cost and what it would mean as they put LNAs together with switches, et cetera. So even if I did have the number, I'm not sure that it would be something I should talk to as to how they would be benefiting.

In our case, I can certainly say that as we increase functionality, we always reset the price points, so a new platform, there is -- will always be a year-over-year price reduction, and maybe not every year over year, but on average there will be.

The way that we stem that is by putting out new platforms with either a better figure of merit or with greater functionality. And so that always for us resets the price point and brings us back to a price of where we had been or better from several years ago, because there is going to be a price -- a year-over-year price reduction of an existing flow.

That's terrific. Now on just the small market automobile market where EVs, electric vehicles, will become an increasingly larger percent of the mix globally, just witness what Tesla is doing, what content in a car might Tower be addressing at the current time? What is the potential per car?

As far as the amount of chips or what we are focused on? The major focus is battery management. As far as how many chips per car that is, I would have to speculate. It is very, very many. I don't have for any given car a specific number for you. But it all really, for the most part, deals with battery management.

Thank you. Good luck. Great quarter. Thanks.

There are no further questions at this time. Mr. Ellwanger, would you like to make your concluding statement?

Certainly. Again, it was a good quarter. It has been a good trajectory in the Company. Many of you, both analysts, as well as investors, have been long term with the Company and we really do appreciate your faith in us and the motivation that we have from those relationships.

We are very excited. Tomorrow morning, we ring the opening bell at NASDAQ, followed by an analyst day beginning at 10 AM at the NASDAQ building. Those of you who have registered for the day, we very much look forward to seeing you. For those of you who did not register, the presentation will be available on our website before the event begins, and we will have an on-demand webcast following the event.

Again, we look very forward to those of you who have signed up and invite anyone who didn't sign up to, after the event, look at the webcast. I think there will be very exciting things discussed. It will be -- our Chairman will give introductory remarks and I think that will be very interesting, to hear his perspective and view on the industry and on us. It will be followed by myself. Then the President of the Company will say some words. Each business unit GM will speak, and then we will actually get into a little bit of some philosophy of the Company in that the head of human resources worldwide will talk about the employee base, what motivates them, how and why we've been able to be successful with the acquisitions that we've done, and we will have a good summary of our operational capabilities.

The ending will be our head of worldwide sales talking about the worldwide customer base, why it is so strong and what enables the stickiness we have with our customers. And then, I will give a little closing remark about a recent epiphany that I had had. So, hopefully, it will be enjoyable for everyone, uplifting and valuable, and again invite -- we welcome everyone that had signed up and really invite everyone that didn't to view the webcast.

With that, thank you very, very much.

Thank you. This concludes the TowerJazz third-quarter 2016 results conference call. Thank you for your participation. You may go ahead and disconnect.