安森美 (ON) 2006 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the ON Semiconductor third quarter earnings release conference call. [OPERATOR INSTRUCTIONS].

As a reminder, this conference call is being recorded.

I would now turn the conference over to our host, Mr. Ken Rizvi.

Mr. Rizvi, please begin.

Thank you, Christopher.

Good afternoon, and thank you for joining ON Semiconductor's third quarter 2006 conference call.

I'm joined today by Keith Jackson, our CEO, and Donald Colvin, our CFO.

This call is being webcast on the Investor Relations section of our website at www.onsemi.com and will be available for approximately 30 days, along with our earnings release for the third quarter of 2006.

Our earnings release and this presentation include certain non-GAAP financial measures.

Reconciliations of these non-GAAP financial measures to the most direct comparables under GAAP are in our earnings release and posted on our website in the Investor Relations section.

In the upcoming quarter we will present at the CSFB Technology Conference on November 30 and the Lehman Brothers Technology Conference on December 6.

During this conference call, we will make projections or other forward-looking statements regarding future events or the future financial performance of the company.

The words estimate, intend, expect, plan or similar expressions are intended to identify forward-looking statements.

We wish to caution that such statements are subject to risks and uncertainties that could cause actual events or results to differ materially.

Important factors relating to our business, including factors that could cause actual results to differ from our forward-looking statements, are described in our Form 10-K and other filings with the SEC.

The company assumes no obligation to update forward-looking statements to reflect actual results, changed assumptions or other factors.

Now let's here from Donald Colvin, our CFO, who will provide an overview of the third quarter 2006 results.

Thanks to everyone who is joining us today.

ON Semiconductor Corporation today announced that total revenues in the third quarter of 2006 were $420.9 million, an increase of approximately 12% from the second quarter of 2006.

Total revenues during the quarter included approximately $372.2 million of product revenues and approximately $48.7 million of manufacturing services revenue.

During the third quarter of 2006, the company reported net income of $76.8 million, or $0.23 per share on a fully diluted basis.

Third quarter 2006 results included approximately $2.9 million associated with stock-based compensation expense, primarily due to our adoption of FAS 123(R) Share Based Payment.

During the second quarter of 2006, the company reported net income of $67.5 million, or $0.19 per share on a fully diluted basis.

Second quarter 2006 results included approximately $2.4 million associated with the stock-based compensation expense.

On a mix-adjusted basis, average selling prices in the third quarter of 2006 were approximately flat with the second quarter of 2006.

The company’s gross margin in the third quarter was 38.1%, down approximately 270 basis points as compared to the second quarter of 2006, as anticipated, due to an increase in lower margin manufacturing services revenue.

Product margins were approximately 40% during the third quarter compared to 41.6% during the second quarter of 2006.

EBITDA for the third quarter of 2006 was $106.3 million compared to EBITDA for the second quarter of $96.8 million.

During the third quarter, the company paid down approximately $60 million of its senior secured credit facility with cash on hand.

It made its final payment of approximately $15 million to LSI, associated with the purchase of the Gresham Wafer Fab, and paid $9 million to increase our ownership to 70% in our China JV.

The company exited the third quarter with cash, cash equivalents and short term investments of approximately $269.2 million.

At the end of the third quarter, day sales outstanding were reduced to approximately 42 days.

Internal inventories decreased on a days basis by 8 days to approximately 72 days.

Distribution inventories in dollar terms remained relatively flat, as anticipated, at approximately 11 weeks.

Cash capital expenditures during the quarter were approximately $52 million.

Now I would like to turn it over to Keith Jackson, our CEO, for additional comments on the business.

During the third quarter of 2006, our product revenue growth was driven by our consumer, computing and wireless end markets.

Our largest end market was the consumer end market, which represented approximately 25% of third quarter 2006 product sales and grew approximately 7% sequentially.

Computing, our second-largest end market, grew 14% sequentially and represented approximately 23% of our [second] quarter product sales.

Growth in this end market was driven by the account penetration with our new products around Vcore, trench MOFSETs, gate drivers and DDR memory controllers.

The wireless end market represented approximately 18% of product sales during the quarter and grew approximately 9% sequentially, driven by account penetration with our EMI filters and analog switches.

The automotive end market was down seasonally and represented approximately 15% of product sales.

The industrial and networking end markets fell sequentially and represented approximately 12% and 7% of product sales, respectively.

Industrial was down primarily due to slower sales in the automated test equipment market, and networking slowed with lower sales in the wireline equipment market.

Shipments to Motorola represented approximately 6% of product sales in the third quarter.

During the quarter, Motorola completed the sale of its automotive division.

In addition to our sales to Motorola, we have sold approximately 2% of product sales to Continental TEMIC, the acquirer of Motorola's automotive business during the quarter.

Our top five OEM customers on a direct billing basis for the third quarter were Continental TEMIC, Delta, Motorola, Samsung and Siemens.

Our top manufacturing service customer was LSI, as anticipated.

On a geographic basis, our contribution this quarter from product sales in Asia, excluding Japan, increased by 200 basis points to approximately 60% of sales.

Product sales in the Americas decreased by approximately 200 basis points and represented approximately 20% of third quarter sales, primarily due to the slowdown in automotive sales.

Product sales in Europe remained flat at approximately 16% of sales, and sales in Japan were flat at 4% of sales.

Looking across the channels, sales to the distribution channel remained at approximately 48% of product sales.

Direct sales to OEMs decreased by 100 basis points to approximately 37% of product sales, and the EMS channel increased by approximately 100 basis points to 15% of product sales.

We also estimate that approximately 20% of sales through the distribution channel were for third-party logistical services for the EMS channel.

During the third quarter, product revenues broken out by our new divisions were as follows.

The standard products group represented approximately 35% of product sales.

The automotive and power regulation group represented approximately 29% of product sales.

The computing product group represented approximately 25% of product sales.

And the digital and consumer products group represented approximately 11% of product sales.

We will post the quarterly revenue, gross margin and operating margin breakouts on our website when we file our 10-Q, which we expect to file shortly.

Now, I'd like to provide you with some details on the progress we've made.

We continue to see strong growth and traction with our new computing products.

In the third quarter of 2006, we saw 14% sequential growth in our overall computing revenues.

This growth was driven by our leading-edge, dual-core VR controllers and our trench II MOFSETs.

The VR11 controller on the Broadwater platform represents our first major processor controller win.

We are currently ramping our revenues on the Broadwater platform with two major OEMs, and our design efforts through our solution engineering centers are paying off.

We've won additional designs with six major OEMs and ODMs for our VR controllers and their companion MOFSET drivers and MOFSETs that will begin ramping in the second half of 2007.

We will also be releasing VR controllers for both the notebook marketplace, as well as PCs, based on the AMD processor in the upcoming months.

In addition to our VR controllers, we continue to launch products such as our high-side and low-side MOFSETs, our low voltage buck controllers on the memory side, enhancing our overall computing offering.

Year to date, we've grown our power management portfolio to more than 8,000 products.

We continue to work with our customers to help solve their problems through innovative product design solutions.

We are also working with our customers in industry standards organizations to develop new products and solutions like our GreenPoint reference designs to maximize power efficiency and minimize design costs and effort.

In the portable and wireless end markets, we continue to innovate and migrate to smaller packaging solutions providing our customers with the flexibility and design and the ability to optimize power consumption and board space.

We saw 9% sequential growth in the wireless end market due to the increased penetration with major global cell phone manufacturers.

We continue to expand our product portfolio with products like our two-channel charge pump, white LED drivers, and S-Wire link control.

Our new high performance PDM controller for this market is also the industry's smallest solution, integrating a 3 megahertz buck converter and an LDO in a micro-packaged solution, optimizing battery life and extending play time.

As you know, the automotive industry, and, in particular, North American automotive manufacturers have been experiencing a slowdown due to the overall restructuring of their business.

Even though vehicle sales have been relatively flat, the electronic content of each vehicle continues to show healthy growth.

We continue to work with our customers in the U.S. and abroad on next-generation products, enabling automobiles convenience, comfort and safety.

We expect to roll out our first solutions engineering center dedicated to the automotive market in the first quarter of 2007, which will be located in Munich, Germany.

This SEC will help align us with what has become the global center for automotive electronics innovation.

Our automotive SEC will enable us to expand upon our recent successes in the automotive end market.

We have recently won designs with our new BCD smart driver to manage vent and vacuum functions and have recently introduced new low class and current LDOs that is the first in the industry to enable manufacturers to meet the new low class and current requirements being enacted by major automotive OEMs.

We've also a strong presence in the automotive infotainment front, with approximately $1 to $1.50 content per satellite radio console, which provides a platform for growth as satellite radios become a standard feature in automotives.

Our products continue to win awards from publications, our channel partners and our customers.

We recently received a Top-10 New Products Award from Electronics Products China for our VR controller.

Our high side SMART HotPlug IC and our notebook DDR power controller both won innovation awards for 2006 with EDN China.

Now I'd like to turn it back over to Donald for our other forward-looking guidance.

The fourth quarter of 2006 outlook.

ON Semiconductor recognizes revenue through a distribution channel on a sales-through basis.

During the third quarter, we saw a high amount of distribution sale-through revenue at the very end of September, pulling in some revenue into the third quarter that was originally anticipated to ship in the fourth quarter.

Based upon product booking trend, backlog level, anticipated manufacturing service revenue and estimated turns levels, we anticipate that our total revenues will be approximately $390 to $400 million in the fourth quarter of 2006.

We also anticipate that approximately $30 to $35 million of our total revenues will come from manufacturing services revenue during the fourth quarter.

Backlog levels at the beginning of the fourth quarter were down from backlog levels at the beginning of the third quarter of 2006 but still represent over 90% of our anticipated fourth quarter 2006 revenues.

Due to annual contract negotiations, we expect that average selling prices for the fourth quarter will be down approximately 1% sequentially.

We expect our product gross margin will range from 38% to 39% and will be down from the third quarter primarily due to lower manufacturing rates.

Our manufacturing services gross margin is expected to be approximately breakeven in the fourth quarter of 2006.

Starting in May of 2006, we entered into a take or pay contract with LSI associated with a portion of our manufacturing services revenue.

This represents approximately $200 million over two years, averaging $25 million a quarter.

For 2007, we expect to have approximately $100 to $110 million in overall manufacturing services revenue with gross margins similar to our fourth quarter guidance of approximately breakeven.

We are also working on opportunities to partner with other semiconductor manufacturers to gain additional process technology and foundry business for our Gresham facility.

We believe we will be able to harvest these opportunities in the upcoming months.

In the fourth quarter, we believe that distribution inventory will fall on a dollar basis to less than 11 weeks.

For the fourth quarter, we expect cash capital expenditures of approximately $40 million.

For the fourth quarter, we also expect total operating expenses of approximately 18% to 19%, with SG&A expenses of approximately 11% to 12% of sales and R&D expenses at approximately 7% of sales.

We anticipate that net interest expense will be approximately $10 million for the fourth quarter.

We currently expect stock-based compensation expense on a pre- and post-tax basis to be approximately $3 million; and this expense is included in our guidance.

We expect our fully diluted share count to be approximately [$331 million] in the fourth quarter, which includes approximately 325 million of common stock and approximately 6 million shares related to options.

Further details on the share count and EPS calculations are provided regularly in our 10-Qs and -Ks.

With that, I would like to start the Q&A session.

[OPERATOR INSTRUCTIONS].

Our first question or comment comes from Mark Edelstone with Morgan Stanley.

Your line is open, sir.

Nice job on the quarter overall.

A couple questions.

I guess the first one may be for you, Donald.

On the gross margins you saw in the third quarter your own product sales down sequentially.

Can you talk about what the drivers were there?

Was that just all mix related?

I'm guessing maybe the ETE business being weaker drover that because, obviously, pricing was pretty flat.

So, what were the issues there?

I think you've hit the nail right on the head, Mark.

It was specifically due to mix.

We did have a significant amount of-- a lower amount of higher margin high frequency business that goes into both ATX and testers.

That was down sequentially, and that hit us.

But, also, we reduced slightly our manufacturing rates compared to the fourth quarter.

So, that's something, as well, that we had anticipated when we gave guidance.

Indeed, we also anticipated a slightly lower mix of ATX business and high frequency.

So, our guidance-- we came in spot on to our 38% guidance.

Okay.

Then, for you, Keith, if you could.

It sounds like distribution inventories are in good shape.

What's your assessment of OEM inventories at this point?

I think inventories overall are in pretty good shape.

We did see during the third quarter some of the buffer stocks being slightly reduced, and I think some of that momentum will continue into Q4.

But, in general, I think inventories are actually in pretty good shape out there.

We should be seeing something approximating the end markets as we head into Q1.

Lastly, I just wanted to clarify-- Donald, when you talk about the foundry business being breakeven, I assume there you're talking about the bottom line impact.

Right?

Yes.

There's very little operating expense associated with that business, Mark, so it's both part of gross margin and the bottom line.

Got it.

Thank you.

Thank you.

Our next question or comment comes from John Barton with Cowen.

Mr. Barton, your line is open, sir.

Donald, could you elaborate on your comments about the acceleration of [DSB] billings at the end of the quarter and the fact that it moved dollars from Q4 into Q3?

What I'm interested there is kind of magnitude and motive, if you would, please.

We actually had an exceptionally strong finish to the quarter.

Primarily, in Asia, we did see a lot of higher sell through, and we were internally projecting.

That has made, probably, the transition from Q3 to Q4 a little bit sharper than would have been desirable.

So, I think we're talking about something like $4 million or so-- more than we were projecting internally in our numbers.

We'll pay the price for that in Q4 because we've got $4 million less.

So, that is kind of an ironic thing when there's so much nervousness out there that, actually, business comes in so much stronger than we'd thought.

We really finished the quarter with a-- out with the high end of range on sell through from DSB.

And, again, the motive on it?

Do you think customers pulled in requests?

Do you think distribution needed to make up some--?

I'll chime in there.

This is Keith Jackson.

I believe what happened there is we had Golden Week coming up, and we had a number of the distributors there wanting to get that stocking into the system before Golden Week.

I think that was a primary motivator.

Got it.

Thank you.

Then, if you could just comment on the trend line recently in the shape of the game console business and how you think about that as we look into next year.

I think the game consoles are weaker than was anticipated earlier this year due to some delayed launches, etcetera.

But I do expect that as you get into Q1, we should see full ramp effects on the new consoles.

From a sell-through pattern, they seem to be, based on last year's experience, much more stable than they have in past years in the Q4/Q1 transition.

So, my guess is Q1 should be as strong, if not stronger, than Q4.

Thanks, Keith.

Thank you.

Our next question or comment comes from Michael Masdea with Credit Suisse.

Your line is open, sir.

I guess kind of a higher level question.

If you look at the DSCs, they seem pretty lean now across the board.

I guess someone asked earlier, but-- The EMS-- do we need to have a little bit of reduction there on the OEM side in 4Q?

Then, given we've been under growing-- we're starting to under grow here the end markets in the second half.

Do you think that the supply chain is getting too lean again, and we go through the same sort of [inaudible] some point next year?

Keith Jackson again.

I do think we are pretty lean by historical standards in this part of the cycle and season both, both in distribution and EMS.

I think by the end of the quarter, OEM will also be quite lean.

As we look forward on factory utilization, we're not seeing that drop substantially like we do see whenever there's an inventory correction out there.

So, bottom line is;

I think capacities are going to stay relative full.

Distribution channel is relatively lean, and my guess is that we should see some very modest seasonality and then some very strong demand as we get into the second half of next year.

Great.

It's always an opportunity for you guys to look out there for potential M&A.

Given this environment right now and some of the skepticism, is there anything out there that looks interesting to you guys that's worth working on right now?

Or, is it a pretty quiet environment on that front.

Well, there's certainly been some very interesting large deals that have gone down in recent months.

But, I think there's a lot of us out there feeling our equity is undervalued, and there are certainly opportunities.

As a company, of course, we are looking at any opportunity that might make sense for us relative to the analog and power piece of the industry.

I think you would understand, Michael, that we're not going to announce a short list of any targets on the conference call.

You can feel free, if you want.

The last quick one, real quick, is-- Everybody's so focused on inventory right now.

Is there anything from the demand perspective or just cycle or company perspective that we're really just not focused enough on right now, or even for the industry?

Again, I think demand is overall pretty good.

Clearly, the end markets are growing a little less quickly than they were at the beginning of last year.

But, nonetheless, I think the growth is still there, and we expect it to continue into next year.

So, I think my perspective at this point is, as I mentioned earlier, kind of a very mild seasonality with some very strong pick up because we've got lean channels and still relatively full manufacturing in the supply side.

Thanks, guys.

Thank you.

Our next question or comment comes from Chris Caso with Friedman Billings.

Your line is open.

I just wondered, Keith, if you could just add to some of the comments you had for Q1.

Perhaps-- I realize you don't want to provide guidance for Q1 now.

But, maybe talk about some of the visibility that you may have.

I guess, are your comments driven mainly by the inventory levels, or are your customers giving you any visibility with regard to how the year might start?

Relative to what most folks call visibility, which is really translated backlog, as lead times have come in on the industry, I think, actually, you're going to see less backlog positions as we head into the new year across the entire industry.

However, as far as outlook, the customers are actually feeling pretty good.

They see the-- the consumer season lasting into Q1 with kind of a later Chinese New Year this time around.

And, again, I think in our industry, there's a very significant play on capacities relative to that ASP.

So what we're seeing now is relatively full capacities, relatively lean channel.

So, we're looking for very moderate pricing pressure as we go into what is normally the toughest pricing quarter of the year, which is Q1.

That's my basis for much more moderate seasonality than normal.

Okay.

With regard to your production levels, I think I caught from some earlier comments that the intention here was to keep the production levels fairly constant here, at least over the next quarter or so?

We've taken them down a bit to get a little more inventory out of our system, but it's just a very moderate correction.

We ran in the mid 80s capacity for Q3, and I would expect, even with those corrections, to stay above the 80% line as we head into the seasonal slow season.

Okay.

And just one final question with regard to the manufacturing services revenue.

It looked like that was up fairly sharply in the quarter.

I think your comments from last quarter was that it would about double.

It looks like it did more than that.

Maybe you could just talk about that a little bit.

Is this a decision that ON Semi makes in terms of how much manufacturing services revenue to take, or is this kind of being pulled from LSI?

No.

This is a pure pull.

We clearly will service whatever's needed out there.

They just had good strong demand in the marketplace for Q3.

It's Donald, Chris.

We don't-- It's not only LSI; we have other partners in there.

LSI is by far the biggest buyer.

But we are developing other partners and particularly for Gresham.

Right now we have three people in Gresham, including LSI, and probably within the next quarter or two, we will have qualified another, at least, two additional customers.

So, obviously, the game plan as LSI ramps down is to qualify other manufacturing services partners.

Okay.

Great.

Thank you.

Thank you.

Our next question or comment comes from Craig Ellis with Citigroup.

Mr. Ellis, your line is open.

I missed the early part of the call, so excuse me if you've already hit this.

Operating expense is now at a very lean level relative to the last six quarters.

How should I think about operating expense as a percentage of sales as we head into 2007?

I think we gave the guidance.

We said R&D would be about 7%, and SG&A would be 11% to 12%.

So, that's the kind of range we'll be thinking of.

You add the two together, you come to about 19%.

That's a good scene of what you should think of looking into next year.

As you start growing again, Donald, you've got good leverage against those expenses?

Obviously, there's a wee bit of [inaudible].

Our game plan is to try to keep OpEx and make sure it never gets above 20% and to keep it well under that when we have high growth on the top line.

So, yes.

We do have leverage on OpEx, especially we demonstrated that this year.

As Keith mentioned, we are confident that our business will grow substantially next year, especially in the second half.

That's when you'll see the real leverage on operating expense.

Okay.

Outside your foundry partner, any 10% customers in the quarter?

No.

Our customer base has broadened, thanks to Motorola selling its automotive division to Continental TEMIC.

So, we always-- this is the first call, I think, Keith, when we haven't mentioned more than 10%.

Yes.

Because we've sold the auto business to Continental TEMIC.

All right, guys.

Thanks so much.

Thank you.

Our next question or comment comes from Steve Smigie with Raymond James.

Your line is open.

A first question was just on the revenue related to Broadwater.

I think Intel said in their call that they were a little bit slower there than they had anticipated because people were waiting for it, for an integrated graphics version.

I was just curious if that affected you or if you're tied to that.

I was hoping you could talk about that a little bit.

Sure.

Not really.

Basically, again, this was our first big platform.

So, we're really penetrating a market we previously did not participate in.

So, very strong growth going from virtually nothing is what we're really seeing.

So, on an absolute basis, it may have been slower than what Intel was anticipating.

Bur, for us, it was all growth.

Right.

You experienced very nice growth.

I guess maybe one question more is, as the new graphics portion starts to ramp, do you see an even further accelerated ramp?

We see continued growth there, but the new platforms certainly provide new opportunities for us.

Some of the design wins that I mentioned earlier are actually on new platforms as well.

So, we see that that business will continue to accelerate on us through next year.

Thanks.

If you could help me understand the foundry business a little bit better-- The first question on that is-- Obviously, you had a decent amount of revenue here, close to $50 million, I think, and guided to something like $30-some million next quarter.

The total was supposed to be, I think, $200 over two years.

Given the guidance for next year, it seems like you get through it in about 18 months.

I think because of this [inaudible] and business that we are wanting-- we're not in this strategically to-- the product line we're planning to ramp it down, especially dependent upon LSI.

We thought we would break with our normal habit and spell it out for next year to make your life easier.

So, although the contract does last for two years-- it does continue after next year, but we just wanted to give you the color as far as we can see now.

We see between $100 and $110 million of this manufacturing services revenue with a zero contribution so that you can better model our base business and our overall business.

As I also stated, we are very encouraged by the opportunities that we have to make use of that facility in ways that will be complementary to our own technology development efforts.

We hope that we can harvest some of these opportunities and update our guidance.

But, for the time being, we'll stick in the sand now at $100 to $110 million.

If we have success, we will update our guidance going forward.

But there still will be substantial business with LSI after the end of 2007.

Okay.

How long would you think about bringing on other foundry partners?

At what point is it--?

It's not really foundry partner; it's more-- We're thinking more of people who can-- we can share in technology development.

The end game there is that we will run all of our products in our facility.

It will allow us to move up the food chain, higher gross margin products, better functionality for our customers.

We've already started a few hundred wafers a week for some of our products, and the initial results have been exceptionally encouraging, much better yields than we have in our existing facilities.

That's the end game.

We're looking for partners that can help us accelerate the introduction of processes and help us offset the cost of doing that over time.

We do not want to be a foundry.

Okay.

Great.

Thank you very much.

Thank you.

Our next question or comment comes from Chris Danely with JP Morgan.

Your line is open, sir.

Keith, I guess to paraphrase you, it sounds like the issue in Q4 is more, I guess, demand being a little bit lower than people were hoping for?

But, then you expect a better than season Q1.

Is that--?

I guess I'll spell it out a little clearer.

I believe that people were anticipating a little faster growth, particularly in consumer, earlier, if you were going to ask this question a quarter ago, than what we're really going to see.

So, I think, again, the growth rates have moderated, or they're slower than what was expected a quarter ago.

There's still growth.

But, certainly, with the OEM side doing some of their, what I would call, buffer stock-- leaning out with shorter lead times, etcetera, you're seeing some impact, I think, just with a very minor inventory correction going on in Q4 on top of what is growth but not as strong as one would hope in Q4.

Okay.

Then, you talked about your lead times, I think, being in excess of 12 weeks last quarter.

Have those changed, or are you trying to bring those in at all?

We're trying to bring them in a bit.

I think we range from 8 or 9 to 13.

So, we're-- Probably, instead of 12, we're probably closer to 10 or 11 at this point.

But it's now at the point where the orders can still be placed inside the same quarter, and you get more turns.

So, we're going to move to a little higher turns environment.

Sure.

What would be your idea lead time?

Well, if we could manage it, we'd like to be 8 to 10.

That's kind of the sweet spot for us and gets close to our manufacturing cycles.

Great.

Can you talk about your expected utilization rates and gross margin trends for next year, just on a rough basis?

No; we really don't give next year at this point, Chris.

Again, I think we've got a pretty good baseline with the work we've done this year to keep the utilization rates going into Q1 in the 80s and then grow it from there.

You get the appropriate margin fall-through as you would expect.

That makes sense.

Thanks, guys.

Thank you.

Our next question or comment comes from Ramesh Misra with Unterberg, Towbin.

Your line is open.

A question on automotive.

When do you expect that part of your business to start showing some growth again?

Typically, actually, Q1 is when we started seeing the next growth in that business on a seasonal basis.

I would suspect that that would happen again next year.

The Q3 is always the softest quarter, as they do their changeovers for the models.

Then, Q1 and Q2 tend to be the growth quarters heading into the next year.

Okay.

Then, on the foundry side, with the addition of these four other customers that you plan to bring on, I thought that your guidance for the foundry business for next year-- I think you said $100 to $110 million.

That seems-- appears rather low.

Am I missing something?

What I said, Ramesh, was that we had a couple of additional customers already on a small level, but we're planning to bring on several others.

The several others-- the two others are not in [that gated], but we don't have that currently won.

But once we get that, we will update our guidance.

So, I got your-- I think I understood your point about trying to spread some of the technology development costs with these other customers.

But, on a general basis, do you find that the capacity at the Gresham fab is a lot more than you can utilize in the near term, and that's why you're looking also--?

Yes.

That's a good way of looking at it.

We knew when we bought it that LSI was ramping down, and our needs were ramping up.

It's just a matter of the pace at which we can do that.

For us on our own product side, we're putting in new processes there, which takes time.

So, it's that time gap, if you will, that we're trying to work with partners to get full.

Okay.

Then, a final question.

Remind us what your CapEx was in Q2 as well, and where are you allocating most of that CapEx going forward?

Thanks very much.

The cash CapEx-- we said it was $15?

$15.

$15.

I think the way to look at the capital expenditure is this year we're going to be running at just over $100 million, if you exclude the-- I'll put it simple.

I think we're running about-- Our projection this year is something like $115 to $120ish, if you exclude Gresham.

Next year, we should be looking at 7% or 8% of sales, something in the $80 to $100 million.

That's what we've always said is our capital expenditure model.

So, that's not changing.

Relative to how we spend that, I expect more of it will go into assembly test capacity than wafer fabs, as you might expect.

With the addition of Gresham last year, I think most of our needs will be to expand our capacities in assembly and test.

Okay.

Thanks very much, guys.

Thank you.

Our next question or comment comes from Romit Shah with Lehman Brothers.

Your line is open, sir.

A question on pricing.

Traditionally, when unit growth slows, we've seen some of your competitors get more aggressive on pricing, particularly in MOFSETs.

I think in the first half of the year, ASPs were up.

Q3 was flat, and now you guys are saying Q4 pricing should be down 1%.

Would you expect pricing pressure to accelerate beyond Q4?

I really don't think so.

Moderately perhaps in Q1, but I would be surprised if there was much additional pressure at all.

As I mentioned before, the difference this cycle is the amount of capacity put on during the peak was much less than in past cycles.

So, the industry is staying relatively full, and the channels are relatively lean on inventory.

So, the gap, if you will, or the perceived gap on the sell-in basis by some of those people you mentioned is going to be much smaller.

So, we're guessing that it's going to be a much more stable pricing environment in Q1 than traditional.

That's what we see when we look at the backlog.

We have no-- As far as the fourth quarter's concerned-- I'm sure you didn't listen attentively to my outlook-- I gave guidance where the margin depression was primarily associated with inventory run rate reduction because we believe in the fourth quarter our mix will improve, offsetting any price pressure.

So, that's a pretty benign environment.

Can you just remind me again, two years ago, how weak pricing got for you guys?

I think we ought to show a chart.

We showed charts on this in the past.

I think we should look further back than that.

If you go to 2003, we were seeing in the first half of 2003 something like a 10% price reduction.

Last year, or 2005, we could see something like 5%.

So, what we have seen is a lower [inaudible] cycle that pressure on the prices [inaudible].

Clearly, the competitive behavior we see on the market is very encouraging because lots of our competitors that were going after market share are now managing margin.

That has been a major factor.

Plus, our introduction of newer, higher margin products and lower costs have helped us manage our margins.

You don't need me to remind you that some of our competitors have now taken on a ton of debt, and they have to service that debt.

So, that should be another factor helping reduce their appetite to go for blind market share.

I think all that is helping.

We don't really see any major pricing pressures on the horizon from the backlog and market data we see today.

Okay.

Fair enough.

If you remember from last year, I think we got the sense early in Q4 that the first half of '06 was going to be a pretty good cycle for industrial for a few analog companies.

I wanted to get your sense today.

Are you confident or positive on another strong industrial cycle for ON as we head into the first half of '07, or is it a bit early to tell?

It's a little bit early to tell, but I will say that the indicators are positive.

This year, our European business, which is one of the-- Basically, the U.S. and European are the heaviest industrial centers.

Europe held up really well.

There was no drop in Q3, which there normally is.

And, Q4 is holding in very well to boot.

So, bottom line there is we're actually seeing quite a lot of strength on the Europe side in industrial.

And, on the U.S. side, other than the automotive piece, which has been soft, it has also held up well.

So, my guess is we should be looking at a pretty reasonable cycle going into Q1 for the industrial and automotive side.

Okay.

Thank you.

Thank you.

Our next question or comment comes from Quinn Bolton with Needham & Company.

Mr. Bolton, your line is open, sir.

Thank you.

I was wondering if you guys could help us quantify, perhaps, the Broadwater opportunity and maybe just discuss the competitive environment.

Obviously, you're coming in, and it looks like you're taking some good sockets.

I was just wondering if you might be able to comment how you see the competitive landscape there in the VR space.

We think we have very competitive products, giving our customers the opportunity to get the best power performance in their sockets.

We have, indeed, won quite a few.

And, again, it's not totally fair for me to pick on any of the competition because we've gone from, again, zero to starting to make some significant end roads.

The encouraging part for me is that, of course, is most of these designs get locked in six months before to nine months before they actually ramp.

These Broadwater ramps we're talking about in Q3 were kind of locked in at the end of last year.

What we're seeing now is for continued share gain into next year, and we're seeing designs locked up that will be ramping kind of midyear next year, again, accelerating our growth into that space.

Again, I believe it's just we've got a very good technological product that gives them best power performance, and, of course, the company has developed a strong reputation for supply chain support and service.

Okay.

Great.

And then, just one for Don.

On the foundry or manufacturing services, I'm assuming the gross margin in the third quarter here was 0%, but I haven't heard you say that.

I just wanted to ask.

No.

We didn't see that.

You're obviously-- We're careful.

We disclose the maximum amount of information that we can.

We had a positive gross margin in the third quarter.

That business is obviously very dependent upon volumes.

We had significant more revenue in the front quarters.

So, it was a reasonable contribution to our earnings in the third quarter.

Okay.

So, the rate you're guiding of $100 to $110 million or about $25 million or high $20s a quarter-- we should assume it's zero if you have a strong growth in that-- In any given quarter you might get some incremental fall-through at the bottom line.

Exactly.

Let's say more than-- I think that any incremental growth above that would give a reasonable fall-through.

That's why we are working very hard to make that happen.

If we get any update on that, we'll let you know.

But, we are encouraged at the opportunities that have presented themselves.

So, you're right; that's the base level.

We thought it was appropriate to baseline that.

If we can harvest some of these opportunities, we'll be only too excited to share them with you.

You can expect that, as we grow the revenue from that, the margin will improve.

Lastly, was all of the $48-plus million of manufacturing services revenue under the take or pay contract?

Does that subtract from the $200 million total?

Or, are you capped at a certain level per quarter?

There is-- the contract means if you take more, you can get full credit for it.

But not all that $48 was with LSI.

As we mentioned, we have other customers there in the manufacturing services area.

It's not just LSI.

But, not all of our revenue was LSI.

The minimum commitment from them is still pretty substantial for next year.

We're working well with them, and no reasons to think that that's going to change.

Okay.

Great.

Thank you.

Thank you.

Our next question or comment comes from Greg Berger with Wedbush.

Mr. Berger, your line is open.

I wanted to know on your internal inventories-- what's the plan there in terms of working them lower?

How much do you want to work lower, if at all?

It looks like they're up $45 million in the last three quarters or so.

We obviously monitor that.

Our inventories are up, but so is our revenue, which is nice. [inaudible] at the end of the third quarter.

I think the key thing is control that.

We have a lot of finished goods, which are for some of these programs that we're working with servicing our customers-- warehouses and things that Keith mentioned are an integral part of our service, [inaudible] and finished goods.

That will work itself down.

And, also, we have [inaudible] or manufacturing run rates, as Keith mentioned, to manage that and keep it under control.

So, I would expect that that will fall.

Internal inventories will fall in the fourth quarter, not substantially but a little bit.

That gives us-- we also have some buffer inventories there as well, if there's any strong upside in demand to take advantage of it.

So, we are pretty happy with where they are now.

They're more or less in line with everything we projected.

I'll remind you that we stated that our DSC inventory would be between 11 and 11.5 weeks.

Actually, it came in closer to 11 because of that Golden Week spot that took place at the end of Q3.

So, we are generally tracking in line with what we expect on our business and for our inventories.

As we don't see a substantial growth this quarter, we are reducing our manufacturing run rates to accommodate that.

Thanks for the detail on that.

The second part of that question is where are utilization rates now, and where do you think they trough in Q4?

Mid 80s for Q3.

I'd be looking for something in the low 80s for Q4.

Is that the trough, Q4?

It could very well be.

Again, it's a little soon for me to make precise predictions there, but it certainly is feeling that that's going to be, if not the trough, very, very close to it.

Then, real quickly, on the ASPs.

I know we talked about it a little bit already.

But, down 1% sequentially in Q3.

So, we should plan on continued kind of down in the 1% per quarter range?

That's my first question.

Secondly, is that more a part-for-part decline or a mixture of decline?

It was down 1% Q4.

Q3 was actually flat.

That is a part-for-part basis, a non-mix-adjusted type of measurement, as best we can do that.

Again, I would say that I wouldn't expect that to accelerate from the 1%.

It ought to be pretty close to that as you get into Q1.

But I certainly would not use that as a four-quarter estimate.

We do think there's going to be-- with the stable manufacturing utilization and the low inventories, I just don't see that pressure continuing as you get to the end of next year.

That sounds awfully stable, awfully boring.

Do you think as the guys like Fairchild and [ERF] pull out of the low end of the market that competition is diminishing for you guys?

Do you see that happening at all?

I don't know if I'd say competition diminishing.

There's still plenty of competition out there.

But, again, I do see that some of the capital investment rates have moderated to the point that it's a good, healthy competition based on a variety of factors and not just price.

Thanks for the detail.

Good luck.

Thank you.

Our next question or comment comes from Tristan Gerra with Robert Baird.

Your line is open, sir.

It sounds as if implementing the new processes at Gresham is taking a little bit more time than you previously expected.

When is your expectation in terms of migrating outside capacity to your Gresham fab, and what could be the outsourcing as a percentage of your [inaudible] by '07, if you have a new outlook on that?

Tristan, it's really not slower than we anticipated.

We knew this was the scenario when we got it.

We're actually ahead of schedule I ramping our trench products into that factory.

They are already being ramped, and we'd originally talked about that kind of late Q4 time frame.

So, we're doing quite well there.

We do expect that our external foundry purchases will decline as we go through next year.

There's no big abrupt decline, but, certainly they will be coming down.

And, whereas we're kind of low to mid 20% purchases outside today, my guess is that gets down well below 20% before the end of next year.

Great.

And, in terms of utilization rates and the expectation to be in the 80s going forward, obviously, you're expecting the low 80s to be your trough.

If for any reason things were to get weaker than expected, and it doesn't seem that you're seeing that, at least for the near term, where would be the point where you would consider shutting fabs?

From the perspective of deciding to shut fabs, the answer is probably not.

There won't come a point like that.

We continually look at how we can get more efficient and have some long range plans on how we can shift capacity, so we can close entire sites.

But, we don't see that on the imminent horizon, even from a strategic perspective.

We've got good demand across the board, and I think it's going to be some time before we can even effect a change.

But it will be through more effective utilization elsewhere, not low demands.

Okay.

Great.

That sounds good.

The last one.

Any guidance in terms of where we should expect the debt level to be by the end of '07, maybe just directionally?

Well, we paid off a big chunk of bank debt, which was the most expensive debt in Q3, $60 million.

It's our plan to continue to do that.

We will continue to pay down bank debt.

Where the overall debt will be at the end of Q4?

We're just over $1 billion, so maybe I can pay some and get under $1 billion.

Right now, we're looking at ways of getting lower cost debt and swapping our higher cost debt, which will still help.

But, I think you can be sure that going forward-- In our current strategy that we exposed at the analyst day, we are a strong cash generating machine.

We generated well over $100 million EBITDA.

We require probably a quarter of that for our capital expenditures.

So, we throw off a lot of cash.

Our interest expense is now under-- just over 3% of revenue or under that even.

So, we are able to generate a lot of cash and pay down debt.

We've done that in Q3.

We'll continue to do that going forward.

How much we pay off in Q4, we haven't exactly determined that.

But, the pattern will continue throughout next year.

Thanks a lot.

Thank you.

Ladies and gentlemen, this does conclude the conference for today.

We do thank you for your participation.

You may now disconnect.

Good day.