高塔半導體 (TSEM) 2019 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

Welcome to the TowerJazz Third Quarter 2019 Results Conference Call.

(Operator Instructions) As a reminder, this conference is being recorded November 13, 2019.

Joining us today are Mr. Russell Ellwanger, TowerJazz's 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 go ahead.

Thank you and welcome to TowerJazz financial results conference call for the third quarter of 2019.

Before we 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 Israeli Securities Authority.

They are also available on our website.

TowerJazz assumes no obligation to update any such forward-looking statements.

Please note that the third quarter of 2019 financial results has been prepared in accordance with U.S. GAAP.

The financial tables and data in today's earnings release and in this earnings call also include certain adjusted financial information that may be considered non-GAAP financial measures under Regulation G and related resulting requirements as established with the Securities and Exchange Commission.

The financial tables include the full explanation of these measures and the reconciliations of these non-GAAP measures to the GAAP financial measures.

Now I'd like to turn the call to our CEO, Mr. Russell Ellwanger.

Russell, please go ahead.

Thank you, Noit.

Welcome to our conference call.

During the third quarter of 2019, our revenues were $312 million, with EBITDA of $75 million and free cash flow of $30 million.

These numbers represent quarter-on-quarter of 4% and year-over-year 11% organic revenue growth, being defined as total revenue excluding revenues from Panasonic in the TPSCo fabs and revenues from Maxim in the San Antonio fab.

In addition, we recorded sequential increase in gross and operating profit, EBITDA, net profit and free cash flow.

This profit and cash flow generation continues to strengthen our balance sheet, and at quarter end, we recorded record shareholder equity of over $1.32 billion as compared to $1.24 billion at year's begin with over $400 million in net cash.

Our fourth quarter guidance is $312 million with an upward or downward range of 5%.

Oren will provide full financial details later in the call.

Now to discuss our businesses.

First, the analog IC business unit.

The third quarter saw solid quarter-over-quarter growth in our RF mobile segment, led by a steep production ramp in 300-millimeter RFSOI, which has now resulted in the highest RFSOI revenue in the company's history.

Looking into 2020, a strong design win pipeline in both 200- and 300-millimeter RFSOI, coupled with the 300-millimeter capacity expansion announced last quarter, positions us for good growth throughout the year.

Growth in this segment for us has been a function of market share gain in both U.S. and China as well as overall market growth with expanded RF front-end content in 5G handsets which are expected to be deployed starting in 2020 and anticipate to continue and accelerate through the next several years.

In addition, we are working on breakthrough RF technologies, such as RF MEMS, being the only foundry to successfully deliver RF MEMS, having delivered into many millions of handsets together with Cavendish Kinetics.

With Qorvo's recent purchase of Cavendish Kinetics, we now work directly with the market leader with this leapfrog technology, rendering higher assurance of strong market adoption over time.

Looking forward to Q4 in the infrastructure segment, our strong SiGe market share in optical fiber transceivers has led to growing orders for 5G infrastructure, where optical fiber connections are being upgraded from a 10 gigabit per second to a 25 gigabit per second throughput.

These orders will help offset any further inventory-related downside from the data center market, stabilizing our silicon germanium revenue.

According to our customers, the excess data center inventory should be consumed within the next 3 to 5 months, hence, we expect to see a rebound as we move into 2020 as well as a resumption of growth in data center silicon germanium shipments.

Longer term, we expect that our silicon photonics platform will have strong growth contribution.

This platform is currently in volume production, expected to reach significant volumes over the next years, being integral to 400 gigabit per second connection speeds.

We are well positioned to demonstrate strong market leadership as the cycle platforms ramp.

Our power management business has seen strong quarter-over-quarter growth, mainly due to a strong ramp in new electrical vehicle battery management products.

Looking further out to 2020, the expanded 300-millimeter capacity will position us for growth in our highly differentiated 65-nanometer BCD technology.

We have a good pipeline of design wins from Tier 1 customers, ranging from data center to mobile applications.

We have also gained multiple wins in our high-voltage technologies, 140-volt RESURF and 200-volt SOI, which expands our available market to a voltage range not previously covered by our technologies, providing further opportunities for market growth.

Looking at our sensor business, we have new projects in the pipeline with large customers on our 65-nanometer global shutter platform that we expect to materialize toward the end of next year, while sales of the existing products are still growing.

Although soft year-to-date, the industrial sensor market is beginning to recover and we foresee double-digit growth through 2020.

Our recent announcement of a wafer stacking backside illumination flow, connecting a top thin BSI imaging wafer to a bottom CMOS wafer on our 65-nanometer 300-millimeter technology, has generated much interest.

We're in various engagements with significant customers for projects using this technology in the time-of-flight market mainly for face recognition.

We also continue our work with multiple customers on optical fingerprint solutions for under OLED and under LCD displays as well as we are seeing good adoption of our stitched single die 300-millimeter substrate X-ray sensors.

Looking at our TOPS business, although the worldwide discrete market remains soft, the automotive portion is growing for our Tier 1 customers and this has translated to an increased rate of new automotive codeveloped product flows.

Aside from discrete, multiple other predominantly customer proprietary flows are moving along, including advanced differentiated micro display technologies, DNA sequencing and others.

Looking at our co-developed Tunnel MagnetoResistance sensor technology.

The overall market for TMR sensors is taking off across many different market segments, including industrial, automotive, consumer and medical, and signals the shift of the demand for this new advanced technology, providing higher performance, greater sensitivity and less power consumption, hence, higher efficiency sensors.

Third quarter 2019 saw the following utilization rates.

In Migdal Haemek, Israel, Fab 1, our 6-inch factory, was down at 60%, with the decrease from the previous quarter driven by the market slowdown in power discrete.

Fab 2, our agent's factory, was 80%, similar to that of the previous quarter.

Newport Beach, California, Fab 3 was at about 50% utilization, predominantly due to the low present demand from data center.

And as mentioned earlier, expecting to ramp a bit in Q4 and then throughout 2020.

Our San Antonio factory Fab 9 was at 55% utilization, an increase from the previous quarter.

Looking at our TPSCo fabs in Japan, starting this quarter, we will report our utilization as a percentage of our allotted foundry capacity allocation according to the new manufacturing agreement.

Our 8-inch foundry business utilization was at about 50% rate, similar to the previous quarter.

Our 12-inch foundry business utilization was 60%, a 20-point solid increase over the previous quarter.

We are now bottleneck constrained, albeit it is non-photolithography bottleneck.

This will be addressed with the new tools for the recently announced capacity expansion program in that facility, which should be fully qualified within the first half of 2020.

This capacity expansion is already fully covered by existing customer demand forecasts.

Important to note, the available 200-millimeter capacity in Tonami and San Antonio as a present base should be well consumed towards our 85% utilization model in 2020 as a result of recent and present activities.

With that, I'd like to turn the call over to our CFO, Oren Shirazi.

Oren, please.

Thank you, Russell, and welcome, everyone.

Thank you for joining us today.

I will start by providing the P&L highlights for the third quarter of 2019 and then discuss our balance sheet.

Revenues for the third quarter were $312 million, representing both year-over-year and quarter-over-quarter organic growth.

We achieved strong results for the third quarter, demonstrating quarter-over-quarter improvement in all our margin and cash flow indicators.

Revenue for the third quarter of 2019 were $312 million as compared to $306 million in the prior quarter.

Compared to 2018, revenues for the first 9 months of 2019 reflect organic growth of 7%, and revenues for the third quarter of 2019 reflect 11% organic growth as compared to the third quarter of 2018.

Note, we define organic revenue as total revenue, excluding revenue from Panasonic in the TPSCo fabs and revenue for Maxim in the San Antonio fab.

This 11% organic year-over-year revenue growth and some efficiency measures enabled us to mitigate a large portion of the Panasonic revenue and margins reduction caused by the March 2019 contract amendment, as announced on March 26, 2019.

Gross profit for the third quarter was $58 million, up $4.9 million when compared to the second quarter of 2019 gross profit, reflecting 80% incremental gross margins as compared to the revenue increase.

Operating profit for the third quarter was $23 million, which is $5 million higher than prior quarter operating profit, reflecting 82% incremental operating margins as compared to the revenue increase.

EBITDA for the third quarter was $75 million, up $5.2 million when compared to the prior quarter, reflecting 86% incremental EBITDA margins as compared to the revenue increase.

Net profit for the quarter was $22.2 million and diluted earnings per share was $0.21, higher than $20.9 million and $0.20, respectively, recorded in the previous quarter.

I will now provide the cash flow highlights for the third quarter and a balance sheet analysis as of September 30, 2019.

During the third quarter of 2019, the company generated $73 million in cash from operations and invested $43 million in fixed assets, net.

This is compared to $72 million cash from operations and $44 million investment in fixed assets, net, in the prior quarter.

Looking at the balance sheet, total short-term debt and long-term debt presented in the balance sheet as of September 30, 2019, increased as compared to December 31, 2018, mainly due to the implementation of accounting standard update ASU 2016-02, Leases, effective January 1, 2019, in relation to lease rights-of-use assets and lease liabilities, which implementation also increased fixed assets balance.

Additional detail regarding ASU 2016-02 are included in Note 2Y to our previously filed 20-F and annual financial statement for the year ended December 31, 2018.

In addition, the first 2 principal payments in the total amount of $37 million, scheduled to be paid in 2020 for the Series G bond, which was issued in 2016, are included as short-term debt as of September 30, 2019.

Shareholders' equity as of September 30, 2019, reached a record of $1.32 billion, an $87 million increase as compared to December 31, 2018, mostly due to the net profits we generated so far in 2019.

Total balance sheet increased from $1.79 billion as of December 31, 2019, to a record $1.90 billion, reflecting shareholders' equity to total assets ratio of 69% and a current assets ratio of 4.5x.

Moving on to elaborate on the tax line in the P&L, I would like to describe our applicable and effective tax rate.

Our U.S. affiliate Jazz Semiconductor and TowerJazz Texas, which own our Newport Beach and San Antonio fabs, respectively, were taxed at 21% rate starting in 2018 following the U.S. tax reform, as compared to 35% prior to that reform.

TPSCo's profits from its Japanese operations are subject to an approximate 32% tax rate.

Our profits in Israel from Fab 1 and Fab 2 operations, while subject to a 7.5% statutory tax rate, are not expected to result in any tax payment in Israel for the foreseeable future since we have more than $1 billion in historical NOLs still to be utilized, which can be carried forward indefinitely.

Considering this and since we have certain tax exemptions, discounts and credits, our all-in worldwide weighted average effective tax rate was 4% for the year ended 2018 and approximately 1% in 2019 to-date.

With regards to hedging activities, in relation to the euro currency, we have almost 0 business in Europe, hence, no exposure to the euro.

In relation to Japanese yen, since the majority portion of TPSCo's revenue is denominated in yen and the vast majority of TPSCo's costs are in yen, we have a natural hedge over most of our Japanese business and operations.

In order to mitigate the remaining yen exposure, we execute 0 cost cylinder hedging transaction.

This 0 cost cylinder transaction, hedge currency fluctuations to be contained in a narrow range as compared to the spot exchange rate.

Hence, while the yen rate against the dollar may fluctuate, the impact on our margins is limited.

In addition, and related to the Japanese yen impact on the balance sheet, we have a natural hedge on cash and loan balances since the loans and the cash are both yen-denominated.

This helps to protect us from potential impact on -- of yen fluctuation.

Lastly, in relation to fluctuations in the Israeli shekel currency, we have no revenues in this currency.

And while less than 10% of our costs are denominated in the Israeli currency, we hedge a large portion of this currency risk using 0 cost cylinder transactions as well.

And a last note on our share count.

As of September 30, 2019, we had 107 million outstanding ordinary shares.

We no longer have any capital notes outstanding since all were converted in the past into equity.

The fully diluted share count is 109 million, the difference between the outstanding and the diluted share count is comprised entirely of ESOP-related options and RSUs.

And now I wish to turn the call to the operator.

(Operator Instructions) The first question is from Cody Acree of Loop Capital.

First off, Russell, could you just talk about puts and takes that went into your December quarter guidance?

I think, given what we've been hearing out of some component vendors, but largely out of the end OEMs in both the data center and in the RF market, what has seemed sort of naturally led to some more growth, if not into September definitely in the December quarter.

Could you speak to that, please?

Certainly.

So as stated, we have not yet seen a rebound in data center.

I am not sure we'd be referring to that as quoting strength within data center from our end customers.

But the strengths that they might be seeing again is off of inventory that they already have in hand.

So as stated on -- in direct communications with lead customers, they believe now that they have somewhere of a quarter to a bit more than a quarter of inventory to burn off.

So our order level in data center has not yet rebounded even close to what it had been in the run-rate of the end of 2018.

I did state that we are seeing some stabilization there due to the fact of 5G infrastructure.

But the amount of parts for 5G infrastructure and the size of the parts is, at this point, not enough to compensate fully for the reduction that we've seen within data center.

But that data center we are very optimistic about from recent dialogues and updates that the orders will be coming back within Q1 and the shipments will be coming and ramping in Q2 throughout the year of 2020.

So that's one area.

And the other area that I had mentioned that we had had very, very nice growth in -- well, okay, so that's a downside.

The other strong area of down for us is within power discretes.

It's been press released who our main customers are within there and -- but it's not just our main customers, the entire power discrete market remains weak.

That will also pop back at some point here, but it is weak at this moment.

So those were 2 of the bigger or maybe the 2 biggest downs that we have against our base revenue that we were having as run rates in 2018, and maybe even at the beginning of this year.

So that's the down part of it.

The up part of it, where I refer to the organic growth and the organic growth then means that you already have to make up for multiple tens of millions of the base having fallen out.

So in organic, that I say a 4% quarter-over-quarter or 11% year-over-year, that's a very, very big growth.

And that came in very, very strongly within the RFSOI as well as within power management.

And I believe I said it during the script, one of the bigger upsides within power management deals with electric vehicle battery management.

So those are quite substantial for us on the upside.

Now if you look at the RFSOI as a whole, I had stated that's one of our big growths there in Q3 was the fact of a ramp in 300-millimeter, where we have just really an outstanding platform and very, very strong customers.

The statement was also made, however, that we are now bottleneck constrained at the 60% utilization level, being a 50% increase over the last quarter on utilization or a 20-point increase of actual from 40 to 60.

But that the capacity expansion that we spoke to last year -- I'm sorry, last quarter of the circa $100 million investment will relieve those bottlenecks, other non-photo bottlenecks.

And that would be being qualified within the first half of the year, planned to be fully qualified by the end of the first half.

And hence, the RFSOI on the 300 millimeter, which is a big portion of the growth, will continue to grow there.

As well as on 200 millimeter, we have very advanced platforms that are being qualified at this time, that should also show very, very strong growth.

And we believe that those numbers would be quite positive.

So if you were to say that the industrial sensor market, which is one that's also decreased and flat, we see that rebounding in 2020.

I can't say at this point how quickly the discretes will rebound.

I really don't know.

But the data center, I believe strongly, will be on a ramp on shipments in Q2, Q3, hitting levels of the past and beyond it in Q4.

So with that base coming back, then the other activities that we have and the increased capacity at 300 millimeter, we foresee 2020 being a very strong growth year, which I think I'd stated in the press release and in all financials.

So hopefully, that's a pretty detailed answer for you, Cody.

Yes, it is, Russell.

Oren, for you, if you would, maybe follow those -- that same walk with gross margins.

If things happen as in the timing expectations that Russell laid out, how you expect the quarterly trends of gross margins to follow?

Yes, I believe our model will sustain the same incremental, about 50%, 55% incremental margin.

If the growth is from 5G and/or Uozu, it should be above 50%, 55%.

And other business like power or discretes, it's a little bit below.

But the average is 50% to 55% incremental gross profit to any incremental revenue.

The next question is from Quang Le of Crédit Suisse.

Russell, you mentioned that your capacity will be installed in the middle of 2020.

In that case, am I right to say that your first half of 2020 will still be rather sluggish in terms of revenue?

I can see that the first quarter normally is down quarter-over-quarter, with Q2 was slightly up.

Is that what we should be expecting in 2020 as well?

So I've not yet guided the first quarter.

And you're correct, the first quarter is typically seasonally down for the entire market.

However, the capacity that you're speaking of, it's not 100% correct.

I said it would be fully qualified by -- within the first half of the year.

There's incremental capacity increases throughout the entirety of (technical difficulty).

So the 300 millimeter shipments will have certain step functions, but I suppose over the -- on a start basis, it will be somewhat monotonic increase throughout January, February and March, da da da, throughout the entire first half.

So no.

Now a sluggish first half, I don't really believe at all.

Meaning there's a good amount of activity, as stated, to drive up utilizations in our 8-inch factories.

And I believe that that will all occur.

So the 60% utilization in Uozu is right now bottlenecked at 60%.

As the year goes on, meaning the 2020 year goes on, those utilization levels will come up against the present photo capacity.

And it won't just be the fact of that that starting Q3 the utilization will go up.

They'll go up throughout the year, meaning throughout the first half.

But there's many, many other activities that are equally accretive to our business.

The big thing that will have a momentous impact is the consumption of the present inventory in data center silicon germanium.

And that has a very good impact in both ways, increases utilization, it's also extremely high-value parts.

So I don't -- hopefully, I answered the question.

But no, I do not expect a sluggish first half, but I have not guided to Q1.

I see.

Got it.

And can I ask for your silicon germanium revenues that you had this quarter?

And what do you expect next quarter?

I really don't want to give specific numbers.

I don't typically do that.

We did, in the Q&A last quarter, state what we thought the SiGe revenue would be, I said $32 million.

And that number was actually exceeded slightly, so it was above the $32 million.

And there will be, against whatever the Q3 number turned out to be, which as I state is -- I did give the $32 million answer in the Q2 Q&A, so it's above the $32 million.

But Q4, at this point, looks like it will be about 10% higher.

The next question is from Raj Gill of Needham & Company.

Congrats on all the good momentum on some of these new products.

The ramp in 300 millimeter for RFSOI, I was wondering if you could kind of describe some of the dynamics there.

Is this being driven by 5G smartphone ramps that you're seeing?

Is it being driven by kind of market share gains?

You had mentioned some market share gains in U.S. and China.

I just wanted to get a little bit more color on the record revenue that you've gotten in smartphone 300 millimeter RFSOI.

And then as we go into next year, there's been a lot of positive commentary about the overall 5G smartphone market in terms of overall units and replacement cycles, et cetera.

And we know that the RF content is going to be going up, so just any kind of qualitative color on how we should be thinking about RFSOI, that the smartphone component, going into next year after a pretty strong year this year?

Two very good questions.

So the 300-millimeter RFSOI ramp is I believe very close to 100% market share gain for us.

The biggest portion of it is dealing with one customer who has gained major market share for themself in areas that we weren't serving previously.

So that, I believe, if it's not 100%, it's very close to that as far as the growth there being market share gain.

Now how much of it is the market growth within the market share gain, it's a little bit hard for me to state.

If I look at 5G as a whole, it would appear that -- and this is from a Yole report that is different from what some large front-end module makers are saying.

But Yole is talking about a 17% increase in RF content on 5G in the -- for the next years.

And I believe it's probably correct because it's not just the fact that 5G is competing against 4G, 5G is actually eating up the previous 2G and 3G market.

And where people have stayed with 2G or 3G, didn't move to 4, the 2G and 3G I think is pretty much dying out.

So the 5G is really being compared not just to 4G content, but predominantly, or at least to a good extent, to 2G and 3G.

So I think the 17% growth in RF units is probably somewhat correct.

And that's a Yole number, it's not my number.

But -- and that being said, I would foresee that as our capacity increases in 300-millimeter, as well as the new 200-millimeter platforms which are now in early qualification stages, as they're qualified, we would continue to see a very strong growth.

Will we have a 17% CAGR?

I don't know.

But I would expect a good double-digit growth next year in our RFSOI.

Okay.

Yes, that's very helpful.

Very interesting in terms of the 2G/3G comp.

And then next question for me in terms of the data center.

So you expect that some of this excess inventory from your data center customers will start to be burned through over the next several months.

And just to understand, you're expecting kind of shipments in SiGe to -- that market to rebound in kind of Q1 of next year, then kind of continue from there in Q2.

So by Q1, we should have the risk of excess inventory being cleared out and then the benefit of restocking plus the deployment of this 25-gig and maybe 100-gig for the optical transport layer?

So you kind of have this double kind of positive impact potentially on Q1 and Q2?

Exactly as you were saying, the -- again, it's not my expectation, it's really based off what our large customers are telling me about their inventory at this point.

So they are saying that it's a 3- to 5-month inventory level that they have.

If that's the case, then orders should be coming in within 2 to 4 months.

And then for the SiGe, it's -- we're running at very good cycle times, although it's many, many layers, but it's within quarter term on the shipment from start to ship.

So if we're starting to increase the data center orders in the beginning of Q1, then we would start to see -- if it's really the very beginning of Q1, you'd start to see some increase at the end of Q1.

But predominantly, the on-ramp and starts would be shipping in Q2.

Okay.

Very good.

And that's the first part.

The second part was also a good statement that you had made.

Certainly, the case that if you have an infrastructure, 5G infrastructure part of 25 gigabit per second, it's a smaller part than a 4 by 25 that you're putting into a 100 gigabit per second data center application.

So the infrastructure portion, it's good to have and it's a very nice business and we like it.

But as far as the amount of wafers for the even equal amount of units, certainly, the 4 by 25, it's larger, right, it's more wafers.

So you're correct, as the data center snaps back and the movement to 100 gigabit per second, you have a double benefit on it.

Right.

Okay.

That's helpful.

And then last question for me.

You had talked about industrial sensors, a little bit slow, primarily because of China, I would presume, and power discretes being a bit weak.

Both those areas you're expecting to kind of bounce back.

Is there a -- and you -- specifically, you mentioned industrial sensors growing double digits throughout 2020.

So some of that kind of positive commentary about a bounce back, is that based on dialogue from your customers where they have -- maybe have cleared out any inventory that they've overbuilt perhaps last year?

Are they starting to see some snapback in China industrial, specifically?

Any kind of commentary on the power discrete, what's giving you a little bit more confidence that that could potentially rebound?

The second part of the question first.

I said I didn't have a specific timing in mind as to when the power discretes will rebound.

That we've not been told.

Customers are still just leery.

And I honestly don't fully understand truly why the power discrete market is down.

I don't, because the power management market is not.

Right.

It's a little contradiction there.

So I don't know.

But it's obvious that it is.

I mean if you look at the top 5 power discrete makers, none of them are talking about gangbuster sales.

So -- but it's -- that is the case.

But now on the second part, for the industrial sensors, yes.

I mean everything that I say really deals off of customer interaction and customer forecasts.

So customers are now talking about coming back with certain orders.

There's obviously, and I said this in the script, we have a variety of new products that are well underway in qualification.

But the old products that are in volume manufacturing now, customers are forecasting an increase.

So if I look at our Q1, Q2, Q3, Q4 forecast in the industrial sensor market, it's very strong.

Now how much can I take that to the bank, again, I'm not giving a target right now for 2020.

I believe 2020 overall is a very, very strong year.

So -- but the customers, as a rule, that we're selling into that market and we have -- it's a big portion of our image sensor business, they're very optimistic of the numbers they're giving us.

The next question is from Joe Flynn of Craig-Hallum.

Congratulations on the strong results.

Thank you.

Just my first question is, I guess, on the incremental gross margin upside, I think around 80% you guys did in the third quarter.

Are you -- to what degree that would be like directly related to the 300-millimeter pickup?

And what kind of levels we should be expecting going into the fourth quarter?

I know you guys just mentioned that you are maintaining your longer-term target of 50% to 55%.

But I was wondering, as RFSOI keeps ramping, if you could possibly see any near-term upside?

Yes.

Obviously, 80% is great that we had the incremental margin this quarter, but we don't assume that it will continue this pace.

And it's attributed both to the 300-millimeter ramp and other -- some mix changes.

But from this baseline, I don't believe you can assume additional 80% incremental; 50% to 55% will be a reasonable conservative assumption.

Okay, great.

And then just one question.

If -- you guys mind walking us through what kind of revenue levels you'd be able to support on the current capacity if, let's say, it was utilized at close to 100%?

And if you mind breaking that up between the total revenue level across all the fabs, and specifically the 300 millimeter?

So I'll let Oren give a swag at it.

But to start with, you have to understand our model: 100% is not doable for us.

There are fabs or companies that will, from time to time, talk about 130% utilization.

That's because they report utilizations with a lot of takeaways in it.

For us, 100% utilization is the absolute maximum that you can load in the factory on the photolithography.

So you can't run at 100% utilization without having a cycle time that just expands and expands.

Our target is somewhere between 85% to 88%.

So that's the number.

When I talked about an 80% utilization, that's 5 points off from the maximum utilization that we can do.

It's not that 80% utilization is 20 points off of the maximum.

And again, I think it's a very important point to make.

So the utilization levels, if I -- again, as I talked about Fab 2, an 80% level is very close to max utilization.

But there are times that we will run at 90% utilization.

But if you were to run at 90% utilization in the way that we calculate utilization for long periods of time, your delivery cycle times increase.

So to maintain a good cycle time, the model that we have is, again, 85% to 88%.

So that's just the first part.

So, Oren?

Yes.

So I believe that at the 88% utilization or 90%, we can reach maximum of 1.6 billion at the current capacity.

I mean, the current, including the new 100.

Maybe 1.65 billion, yes.

1.6 billion to 1.65 billion.

The next question is from Lisa Thompson of Zacks Investment Research.

Oren, I was wondering, given what's going on in Japan, have you given some thought to what next year's tax rate and minority interest might be?

Oh, so with Panasonic, we signed in March 2019 a new contract, which is relevant for the coming 3 years until March 2022, at least.

And so I expect 2020 to be the same run-rate and performance.

As you -- as it is now, I mean, in Q2, Q3 '19.

I don't expect significant changes.

Right.

And if you're investing more, will that mean lower profits for you?

No.

It should bring, of course, higher profits because the time that we will start to depreciate the new tools will be pretty much the time that we will enjoy the fruits of the investments, and this should be quickly showing fruits.

Because like I think Russell mentioned, with all the explanation about the fact that actually Uozu is bottleneck constrained now, so every additional tool and wafer that we can manufacture are really spoken for by customers that really wait for that.

So I believe it will -- the profitability will improve from the revenue improvement that will bring from those new tools.

But I was thinking, your question was about Panasonic demand, which we believe will be -- remain flat like this -- like the current run-rate.

So keep the tax rate at 1%?

The overall tax rate for the company, well, 1% is a very big achievement for us.

I mean, it's not a high number compared to the fact that it should be, theoretically, an average or a weighted average between the 7.5% from the corporate paid in Israel and the rates between 21% and 32% of the affiliates.

So it should be, by model, bigger than 7.5%.

But since we have some credits and deductions and benefits from the specific locations of the factory, maybe you can assume 3% or 4%.

I mean, '18 was 4%.

'19 is 1%.

Everything between 1% to 4% is reasonable.

If you want to be conservative, you can put 4%.

We'll try to achieve 2% or 3%.

The next question from David Duley of Steelhead Securities.

Yes.

Just one clarification.

You mentioned that you thought that your SiGe revenue would exceed its peak late in 2020 of the peak that it achieved late in 2018.

Could you just remind us what the peak revenue was on a quarterly basis in late 2018?

Yes.

We had focused on achieving a $200 million annualized run-rate, so it was at about the $50 million level.

Okay.

And as far as CapEx goes for 2020, could you just remind us what you -- what the total CapEx number is going to be?

I think you're going to spend incremental money there in Japan to increase capacity, could you just remind us what the total should be for 2020?

Yes.

It should be exactly the model -- the current model of $160 million to $170 million a year.

Plus the -- our $100 million that we announced in July for the new CapEx in Japan like you mentioned, so the total is the summary of the 2 numbers.

And as far as the $100 million that you're spending in Japan, could you help us understand the dollar content of capacity that you'll get from that?

Or help us understand the metrics about how much capacity increase you're going to get for this $100 million.

Yes.

It's -- we're talking about -- I mean when fully -- when spoken for, it should be about $70 million annual run rate of revenue.

And the margins from that, you can assume a bit better than the average, which is 50% to 55%.

So you can do the calculation.

Okay.

And as far -- when 5G phones and infrastructure are really in the full ramp, could you take a -- help us understand.

There's a pretty significant content increase of the types of products that you make in both handsets and in the base stations.

Could you help us -- do you have an idea about the percentage increase in content that you might be seeing in -- for 5G, both in phones and infrastructure versus 4G?

I don't know an exact number for the amount within infrastructure of 5G versus 4G.

I really couldn't give an answer on that off the top of my head.

On the mobile platform side, in the front-end module, it's what we had stated, and this is in accordance with Yole, it's about a 17% overall increase, I guess.

But as I stated, that is not just against 4G, that's against 2G and 3G that the 5G will be replacing.

If you were to look at the change between 5G and 4G, there's others that think that that's specifically somewhere about an 8% or a 9% increase in RF content.

This concludes the question-and-answer session.

Mr. Ellwanger, would you like to make your concluding statements?

Certainly.

Thank you.

So again, I thank everybody for their interest, for the very good questions.

Q3, we really did post solid results, we are happy for that.

But as stated in the PR, and not at the moment tangible in dollar signs but believe that it will be over the next period of time, we made very substantial progress, really achieved multiple milestones across all of our business units; had stated that we had major activities, contracts within display, major contracts within BSI; working on extremely exciting things and customer qualification, big end customer qualification with our SiPho platform; continual design wins at 300-millimeter RFSOI; continual design wins at 200-millimeter RFSOI, very advanced RFSOI; and much progress within the power management sector, with the activities on the 65-nanometer BCD platform.

So things are really moving very, very strongly for the company.

We have continued strong activities, pursuing very accretive growth strategies, and we have a very strong outlook, very positive outlook for 2020.

As we approach the year, and I'll talk in a minute about the conferences we'll be presenting at, but typically, at the Needham conference in mid-January in New York, we'll give our target for the next year and I think it will be a very positive one.

So that being said, we're, I believe, in an amazing position right now, activity-based and also financial-based, with the opportunity and capability to do many things that are really being pursued and milestones being achieved with.

In the very short term, in December 4, we'll be attending the Crédit Suisse 23rd Annual Technology Conference in Arizona.

Dr. Racanelli will be there.

It's a whole series of one-on-one meetings that I believe is highly subscribed.

But we'd love to see as many of you as we can.

Needham conference, I mentioned, that's January 14, 15 in New York, a presentation and one-on-one meetings.

Look forward to see whoever would like to meet with us there.

And then in the -- still in the fourth -- in the first quarter, we'll be doing a large event at Nasdaq that we'll be announcing shortly, and we look forward to see everyone at that event.

So many good things happening, and we look forward to updating you on them as they occur.

And thank you very, very much.

Thank you.

This concludes the TowerJazz Third Quarter 2019 Results Conference Call.

Thank you for your participation.

You may go ahead and disconnect.