高塔半導體 (TSEM) 2004 Q3 法說會逐字稿

Good day everyone and welcome to the Tower Semiconductor's third quarter earning 2004 conference call. (OPERATOR INSTRUCTIONS).

I would like to turn the call over to Carmel Vernia.

Please go ahead.

Hello everyone and thank you for joining our call today.

As can been seen from the press release, we have a couple of challenging quarters ahead of us.

But beyond them we are very optimistic about our position in the market place.

Before going into a more detailed discussion, I would like to turn the call to Amir Harel, our CFO, to discuss the highlights of our financial results for Q3 '04.

Amir please.

Hi everyone.

Before we begin, I would like to remind you that some statements made during this call maybe forward looking and are subject to certain risks and uncertainties that could cause actual results to differ materially from those currently expected.

These risks and uncertainties are more fully discussed in the Company's 20-F and 6-K

filed with the Securities and Exchange Commission and the Israeli Securities Authority and are available through the Company's Website.

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

And now I would like to share with you the financial highlights of Q3 of '04.

Revenues were $35.1 million in Q3, 4 percent higher than $33.7 in Q2 and 118 percent higher than $16.1 million in Q3 of '03.

The net loss in Q3 was $39.4 million versus $36.5 million in Q2 versus $37.1 million Q3 of '03.

The increase in net loss versus Q2 '04 and Q3 '03 is primarily attributed to fact to depreciation.

Depreciation and amortization was $29.7 million in Q3 and approximately $83 million year-to-date '04 and is expected to further increase in future quarters and significantly impact our net loss.

The cash closing balance as of September 30, was approximately $53 million.

CapEx in Q3 was approximately $51 million and $132 million in the first 9 months.

The total balance sheet as of September 30, was approximately $857 million.

The total shareholder's equity as of September 30, was $191 million.

As of September, the Company had 55.6 million outstanding shares and 18 million options, warrants and convertible debentures.

As to our guidance, the Company expects Q4 revenue to be in the range of $30 and $33 million.

Depreciation for that quarter is expected to be in the range of $32 and $34 million.

We expect to meet the goal of 14K wafer per month capacity in

December of '04.

According to the updated revenue forecast, we currently expect to reach positive EBITDA in the second half of '05.

And with that I will return the call back to Carmel.

And to a more detailed update on the Company's status.

As I am sure you are aware, market conditions are soft at the moment.

This is true industry-wide and it is true for Tower as well.

We expect the next 2 quarters to be lower in revenues than Q3 of '04.

Q4 revenues will be approximately 10 percent below Q3 revenues.

Our growth plan is set back by 2 to 3 quarters.

Consequently, we now expect to reach positive EBITDA as Amir said, in the second half of '05.

We are also revising our end supply

plans to synchronize them with market condition.

In the longer term, we are very optimistic.

We have a strong sales panel with a substantial lease of new customers who are committed to Tower and expected to reach production in the second quarter of '05 and onwards.

Also, we continue the strong partnership with our existing customers.

The new generation products are also scheduled to run production in the second half of '05.

In Israel alone we had approximately ten customers committed to Tower.

We expect to gain more rapid acceptance now and own significant share of the Israeli market, which is one of the largest semiconductor clusters in the world, as we move forward.

Our capacity ramp is substantially on track and we will have tool capacities of approximately 14,000 wafers per month by year end in fab tools according to our plan.

We are currently reviewing our CapEx plan for '05 in light of current investor condition.

Although, plans are not final for '05, we expect to remain flat at 14,000 wafers per month during the first half of '05.

This is due to the fact that during the next two quarters we expect to have capacity in excess of customer demand and resume growth and expansion in the second quarter of '05.

I'm sure we'll be able to further update on this subject during our next conference call.

On the technology front, all programs are substantially on track.

The first point 13 telecom is running the front end processing of the fab according to plan.

We have expected to be processed at free scale our point 13 technology licensor.

The first point I think embedded fresh customer is running the fab with expected production starting in the first half of the '05.

We now have 7

potential customers committed to Tower, some of them expected to generate revenue for us in '05.

We are on track with our joint development of RSID devices in conjunction with the leading RFID supplier.

Finally, we are currently offering a 10 percent optical shrink to our high volume point 18 customers.

This is a specialized technology note for a specific marketplace that requires minimal R&D investment by the customer.

On the financial front, we had to assume that we will have one or more financing transactions over the course of the next few months to enable us continue ramp up of Fab 2.

I would like to open the floor for questions you may have for Amir, myself, or for Doron Simon President, Tower USA is also on the line with us today.

(OPERATOR INSTRUCTIONS).

Ali Irani with CIBC World Markets.

Given the softness of market itself spoke about

some market share initiatives and you have been on a very strong track for the last year.

Can you give us a sense of the customers especially in the top tier semiconductor manufacturers of IBMs.

What your progress has been -- who you are signing -- what your pipeline looks like and you mentioned specifically ( Indiscernible) it was starting to ramp in the first half of next year.

Could you give some more color on that please?

Thank you.

I will do my best Ali.

This is not studied and please follow up with questions.

Say we now have two customers, which are above 10 percent of our production and another 5 or so, which are between 5 and 10 percent and number of other small customers.

This is how we are split.

The two largest customers I can say so far have been SanDisc and Veron who are in the digital media consumer products going forward in the final.

I don't --- I am not free to mention names, but we have 2 to 3 very substantial customers, which represent volume potential in the same order of magnitude as Veron and SanDisc one of them is in IBM --a Fab like IBM and the other 2 are very large Fab-let companies.

We are seeing in the last couple of weeks Carmel who has been a number of IDMS actually focusing through this downturn on outsourcing more of their business.

Is this outsourcing for you taking place at the. 18 micron node or will it be at the. 13 micro node.

I am hoping you could give us a sense of the mix migrations next year and how much of your capacity in a flat environment 14k through the second half of '05, would you see transition to zero 130?

Hard for me to answer this only but, I don't think we will switch from any of the 14,000 from 0.18 to 0.13, but any increase in capacity will have to reconsider how much of it will be 0.18, and how much 0.13 and that will be treated by customer demand that we will see in the next few months.

We will have both processes ready and we will have to judge in the next few months what capacity to rent out faster.

Great.

I will come back for some more questions after some other people.

Thank you.

Jessie

, Piper Jaffray.

Great.

Good evening gentlemen, thank you for that detailed update.

Back into your guidance there for Q4 can that too still grow sequentially?

Can you ask the question again Jessie, I am not sure I understand it.

We were giving guidance that this will have lower revenues in Q4, and that we will have excess capacities that is for the first time in Q4, since we started ramping fab 2, we will have capacity that is not utilized.

So, --

I am just kind of wondering which fab comes down more sequentially?

In Q4?

Yes.

I don't think we can give you any strong indication.

I think well, there are 2 parts for the answer.

I will give you qualitative answer.

Fab 1 is an older fab and we have many customers there still.

I think we're seeing there a -- some of the decline, which is caused by market condition.

Everyone is reducing their orders.

They want to use their inventories.

So we see fewer orders.

In Fab 2 we are dependent on a more number of large customers and every small move that they are making influences us to a larger extend than the market as a whole.

So, I think what you will see in the next 2 quarters is softness on both fabs but resulting from different reason.

That's helpful.

Amir, could you update us on cell phone and can you remind us what your book value is on that investment?

About the book value, it is easy.

It is written in the books as 6 million, which is the value of the historical investment in our books.

As to if they fund self, since I notice -- even if I notice that I am not in a position to update on their process.

In the media, they say that they are in the process of an IPO, which is probably true.

But, that's as far as I can go in this one.

Fair enough.

And, given the weakness there from what we are hearing in the drilling-edge technology, could you tell us about the -- is there any change in the plan to ramp of Vishay for Fab 1?

No, there is no change.

I didn't mention it because everything runs smoothly.

We expect to start growing production in the second quarter of '05.

The process transfer from a Siliconix to Tower is very smoothly, and we don't see any change in it.

By the way, we order from Vishay for the first 3 years of production starting Q2 of '05 and onwards for 3 years and it's fixed as a

paid.

I am sorry, as what?

paid.

It's a

paid deal for the first 3 years.

So, we are not going to see any change in that.

Darice Liu, C.E.

Unterberg Towbin

In your press release you mentioned declining revenues in those Q4 and Q1 of '05.

Can you further elaborate what you are seeing for the beginning of next year?

As you know, we can only give guidance.

I mean we are giving guidance only 1 quarter ahead.

Our visibility is not as strong as -- the second quarter ahead that is Q1 of '05.

We know enough to say that it will be weaker than Q3 of '04 but, other than this qualitative measure we cannot say anything

Okay and then going into the Q4 quarter can you elaborate in terms of the revenue declining?

Is that more of a function of lower ASPs or function of demand?

It's probably a little bit of both, I would say more of it is a declining demand in Fab 1.

In Fab 2, its specific situations with some of our high-running products where every minor change by one of our customer is affecting us very strongly.

The ASPs are declining, but not in a different way than in previous quarters.

So, the magnitude in terms of the ASP declines are not much different than prior?

That's correct.

Okay.

Thank you very much.

Jessie

, Piper Jaffray.

My question has been answered, but I guess I will throw in.

Could you just discuss your competitive positioning within both the image sensor and the embedded flash markets and has there been any change and if you could just tell us about what you are hearing in those industries?

I would say in short, in CMOS image sensor we see a very heavy competition, new entrants from IBM is entering market and we see everyone is smelling a big market here and everyone is jumping on the bandwagon.

We still think that we have a technology competitive position here that we can maintain, but definitely there are more players out there.

In the embedded flash arena, we are still pretty much alone as far as I know only us and TSMC are offering the kind of line of product that this market requires.

I don't see any changes in this market.

That is helpful.

Thank you very much.

Robert Cash,

.

Hi.

I have a question about how the end markets breakout in your customer, for your revenues, and how many of your customers are in image sensors or how revenues are breakout that way?

Above the break -- Hi it is Amir.

Can you just repeat them?

You asked --

Sure.

I want to know from -- in terms of your customers like what percentage of revenues come from image sensors versus embedded memory or controllers or things like that?

Yes, roughly speaking, I mean in Fab 2, as you might know we are substantially all in core CMOS at this point that we would sell the special technologies from '05 has been increasing quarter to quarter and they are the fixed and owned.

In Fab 1, as usual it's roughly three-fourth with the specialized technologies and one-fourth is with the core CMOS and between the special technologies, it is roughly split evenly between the

and the image sensor and the mixed signal.

Ali Irani, CIBC World Market.

A couple of follow-ups for you.

I am hoping you could give us a sense of utilization as it is running now in Fab 2 and Fab 1, and you mentioned that the Siliconix deal was very much on plan and almost guaranteed.

Can you give us a sense of what that would do to Fab 1 utilization in middle and end of next year?

And then I have some follow-ups on that as well.

Thank you.

Well, I would not say a

utilization fortunately.

So I can't help much on that.

I can say that the share Siliconix as Caramel noted is on track and it will start generating revenue in Q2 of '05 and out of the 10 years or so, or the 7 to 10 years, so the 200 million which is the full deal, 33 million are committed for the first 3 years.

The first 3 years starting from Q2 of '05.

Now, it won't be spread linearly on those 3 years because we would just stop doing.

But you can know that 15 million, again heavier on the second and third year are about still a total of 33 million over those 3 years, so that and it will definitely increase utilization but I can't get into specific numbers.

Okay.

When you look at your current trends in terms of wafer starts, clearly the regular seasonal pickup in wafer fab orders was very soft in September and it seems that it has continued to decelerate in October.

When you talked to your customers about their inventory situation and inventory correction, we see the inventory data coming down over the last 6 weeks.

What sense do they give you in terms of when the monthly production trough will be in terms of their demand?

Do you follow my question?

I think our indications are that, except for the next 2 quarters, Q4 and Q1, which are going to be weak, we expect to see a pickup in the second quarter of '05.

We get those indications from a number of sources and we think they are solid.

But I want to go back to your previous question because I would like to add to what Amir said about utilization.

First of all, the Vishay Siliconix deal, by the time we are entering to full-volume production, it will consume roughly 50 percent of Fab 1 capacity.

So the rest we are going to have sell to other customers.

That's one indication.

But then regarding Fab 3 utilization, we have not been giving figures about it because we were in very fast trends this year and any figure we would have given might have misled you guys.

I think that next year, since our growth rate will be lower relative to our absolute capacity, we will probably be able to give you this indication, I know it's important for you guys to model us.

If you look at the fabs, current equipment delivery schedule and the 14,000 at year end, Carmel, what level of capacity utilization do you think is the minimum that you would have to target as a company in terms of seeking business and bringing business in to weather this downturn and position yourself for a rebound?

Are you targeting a 70-percent utilization level, are you looking at -- the bandwidth so far has had a very good ramp at least keeping up with the average of the foundry industry after all?

A difficult question for me to answer.

I read that the leading foundries are going to be in the range of 80 to 90 percent utilization over the next 2 or 3 quarters.

I think over the next 2 quarters, we will be probably below that.

I think that -- I think in Q2 and certainly during the second half of '05, we project much higher utilization.

Okay. so if the trough month for utilization at the Taiwanese foundries in January is somewhere in the mid 70's range, it sounds like your utilization is, at least on a monthly basis, falling in the same direction.

Yes, I think so.

Okay.

I am wondering if -- the spread you mentioned the Fab 2's volume drop, what's coming down.

But is it specific to 1 or 2 customers, we're all aware, obviously, of the pre-announcement at some of your largest customers, is it a concentrated issue at this point or is it a little bit of each customer?

Is it broadly based?

No, we only have 2 very large customers, everyone else is small.

So everyone has their own issues, which I am not free to speak about and they are both affecting us.

But we definitely see it as a temporary, a local issue and like I said we are very confident about Q2 and onward --

Great.

-- we can be at this point in time.

And just a final question for Amir.

Do you have idea at present of your CapEx in the fourth quarter, and have you started the planning process for 2005 and could you give a sense of at least a range-wise where your CapEx might fall next year?

Let me -- this is Carmel -- let me answer the second part of your question while Amir is looking for the numbers for the first half.

Originally, we planned to invest between $150 and $200 million every year in '04, '05 and '06.

Like I said earlier in opening statement, we are going to revise this plan over the next couple of months taking into consideration the general softness in the industry and for sure during the first half of '05, we will reduce or maybe not even make any investment.

But I don't want to say anything about the full plan of '05.

We will update you again probably during the next conference call.

Okay, fair enough.

And for the fourth quarter ?

For the fourth quarter, it's probably going to be in the range of 50 to 60 million.

Great.

So it sounds like the tools that you were waiting for are finally being delivered and you are signing off on those.

Yes.

It's in line with the statement, we have a goal of 14,000 capacity by year end is still valid.

It relates to those sensors.

(OPERATOR INSTRUCTIONS)

I have -- most of my questions have been answered, but there is one question -- may be Carmel there you can help out.

With regard to I guess your new plan and I guess your CapEx spending next year, can you a bit elaborate on -- you talked about your financing transactions and your plan in next few months a bit what your capital need should be -- what you are planning to do and how you think that's going to impact your balance sheet?

Well, Daniel at this point in time I can only make presidential statement, which means that we'll extend our capacity according to customer demand on the one hand and according to the availability of funds on the other hand.

I don't know what's more specific I can see here in this conference, unless maybe Amir wants to add something to it.

As Carmel, I think, mentioned in his opening statement since everyone including

some I would say awaiting positions right now.

Again we do complete the 14K versus

0.18.

We do complete the fab tool setting to 0.13.

But other than this, everyone we are in waiting positions.

What that means is that probably for the sales half of '05, the capacity would stay at that level.

But we will keep evaluating, and in the few quarters, we would meet to decide if you'll reorder some additional tools, at which point -- in which case we can go up to 20, 21K level or keep waiting because of the demand or financing.

That's the range we are looking at.

Robert Cash,

What is the current burn rate for the Company?

Burn rate is a question when you have a steady stake environment, and this is very far from our situation.

So, it's just the changing because we are in the middle of fairly steep ranks.

So, I can't really relate to the burn rate.

Given your expectations for I guess wafers out and revenues for the next 2 quarters.

How much money we can burn through on an operating basis?

Well, we don't give a specific focus of cash flow, but so I guess I am not sure to how to address this one.

But, we aren't also breakeven yet.

That means cash from operation is very similar to the EBITDA.

It is comparable and we gave our guidance on this one.

(OPERATIOR INSTRUCTIONS) Over to Carmel Vernia for any additional comments.

I want to thank you very much for attending and I'll be talking to you in a quarter or so.

Thank you very much.

And that concludes today's conference call.

We thank you for your participation and have a nice day.