泰瑞達 (TER) 2006 Q4 法說會逐字稿

Good morning.

My name is Nicole, and I will be your conference operator today.

At this time, I would like to welcome everyone to the fourth quarter 2006 earnings conference call. [OPERATOR INSTRUCTIONS] I would now like to turn the conference over to Mr. Greg Beecher, Chief Financial Officer.

Sir, you may begin.

Thank you, operator.

Good morning, everyone, and welcome to our discussion of Teradyne's most recent financial results.

I'm joined this morning by our Chief Executive Officer Mike Bradley.

Following our opening remarks, we'll provide you with details of our performance for the fourth quarter of 2006 and for our 2006 fiscal year, as well as our outlook for the first quarter of 2007.

First, however, I'd like to address some administrative issues.

Teradyne's press release containing our most recent financial results was sent out via Business Wire yesterday evening and it's available on our website or by calling Teradyne's corporate relations office at 978-370-2221.

This call is being simultaneously webcast over our website at www.teradyne.com.

Note that during this call, we are providing some slides on our website that will summarize and reinforce the highlights of our discussion.

This may be helpful to you in following the discussion.

To access them, simply dial into the investor portion of our site and click on live webcast followed by click here for web cast.

In addition, a replay of this call will be provided on our website starting around noon today eastern time.

If it's more convenient, you can also access a replay of the call by dialing 1-800-642-1687 in the U.S. and Canada or 1-706-645-9291 outside the U.S. and Canada. and providing a pass code 6116387.

Replays from both sources, as well as the slides that I mentioned earlier, will be available through the 8th of February.

Investors should accept the contents of this call as the official guidance from the Company for the first quarter of 2007 and beyond.

If at any time we communicate any material changes to this guidance, it is our intent to do so to all investors at the same time to the best of our ability.

Investors should also note that only Mike Bradley, Tom Newman and I are authorized to supply Company guidance.

The matters that we discuss today other than historical information include forward-looking statements related to future financial performance and other performance expectations, statements as to inventory, bookings, backlog, orders, shipments, pricing, design-ends, and demand for our products, capital spending, market share, and other opinions of management.

Investors are cautioned that forward-looking statements are neither promises nor guarantees but involve risks and uncertainties that may cause actual results to differ materially from those projected in the forward-looking statements.

Some of those risks and uncertainties are detailed in our filings with the Securities and Exchange Commission, including but not limited to our Form 10-Q filed on November 13, 2006 and we incorporate here the discussion of those factors.

We caution listeners not to place undue reliance on any forward-looking statements which speak only as of the date they are made.

While Teradyne is under no obligation to update the forward-looking statements made today, any updates that we do make will be broadly disseminated and available over the web.

We want to make clear to investors that our prepared remarks will be presented with requirements of SEC regulation G regarding Generally Accepted Accounting Principles, or GAAP.

Therefore if we use any non-GAAP financial measure during this call, you will find the required presentation of and reconciliation to the most directly comparable GAAP financial measure on the Company's website at www.teradyne.com by clicking on investors and then selecting the GAAP to non-GAAP reconciliation link.

Also and you may want to note that Teradyne will be participating in the following investor events between now and our next conference call.

The Thomas Weisel Partners Technology conference in San Francisco February 5th to 7th, the Bank of America Tech conference in New York on February 21st, the CIBC conference in Vail, Colorado on February 22nd, the Goldman Sachs Technology conference in Las Vegas on February 27th to March 1st.

Now, let's get on with the rest of the agenda.

First, our CEO, Mike Bradley, will review the state of the Company and the industry in the fourth quarter and in 2006 and will provide guidance for the first quarter.

Then I will provide more details on our financial performance for the fourth quarter, for our 2006 fiscal year and our guidance for Q1.

We will then answer your questions.

For scheduling purposes, you should note that we intend to end this call after one hour.

Mike.

Thanks, Greg.

Good morning, everyone.

Sales in the fourth quarter of 263 million were about at the midpoint of our guidance range but, as expected, down about 27% from our Q3 level.

Earnings were higher than projected due to cost reductions we've achieved earlier than originally planned.

Overall bookings were up 20% from the third quarter.

This uplift in new orders, however, was mainly from the normal year-end service contract renewal business and semiconductor tests, while system bookings were down 11% from Q3 levels.

Our Systems Test Group, that is PC board tests, automotive diagnostics, mill arrow, and broad band tests had a strong Q4 with bookings up 50% equalling their level of the second quarter of 2006.

Since these contained long-term program purchases, however, they don't signal a short-term bump in our revenues.

The short story is that we remain in a very cautious period with the majority of our customers.

They're making technology and spot capacity buys.

They're monitoring utilization closely, and they're taking advantage of our short lead times, all signalling a very uncertain market for at least the next quarter.

Therefore, we plan for sales in the first quarter to be between 245 and 260 million with earnings per share between $0.00 and $0.03.

Although our markets ended 2006 on a low note, our results were very solid for the year as a whole.

We logged over $100 million in new socket wins in the year and our semiconductor test revenue grew 34%, while the overall market grew about 20% to roughly 3.4 billion.

Our FLEX testers broke the 1,000 unit install base level during the year and the J750 is now over 2,300 strong.

Essentially, all of our customers are now using the FLEX.

Our catalyst install base stands at about 1600 systems and that represents a small piece of our ongoing revenue.

Record free cash flow and approving systems test business, streamline manufacturing with six to nine week lead times and a 3800 headcount organization, 1,000 of whom are based in Asia all set us up well for the future.

Now I've commented before on our breakeven plans.

We committed to operate at a 250 to $260 million quarterly breakeven level without stock based expenses throughout 2006 and we did better than that every quarter.

Greg will explain that further in a few minutes and he'll outline our plans for 2007.

I'd like to spend a few minutes on our results this year where we gained a little over $100 million in new socket business in semiconductor tests since our share growth in the future is expected to come from more of these socket wins.

While it took scores of socket offenses to reach the $100 million mark, a few of them stand out.

One was in the Japanese microcontroller test market where the J750 was selected for automotive applications based principally on parallelism and throughput.

A second one was a graphics chip test application where the UltraFLEX won due to our high performance serial bus instrumentation.

The third was a major fabless customer who standardizing on the UltraFLEX across a new embedded processor business segment.

And a fourth was a FLEX win, where domestic IDM selected the FLEX for a wide range of linear and power ICs.

Now part of that selection was based on our open FLEX approach which allows integration of third party implementation.

In this case the customer's [technical difficulty] included the ability to embed their own test IP into the open architecture of the FLEX.

There are numerous other stories that all tend to center on some combination of parallelism, instrument performance, platform extendibility, and capital costs.

Now let me turn to our plans for 2007, starting with some underlying market fundamentals.

First, the semiconductor test market will continue to be cyclical.

With 2006 being on the high side of the cycle, we expect 2007 to be on the low side.

However, underlying this cyclicality is an SOC test market that has been growing at a compound rate of about 5% over the last 10 years.

We expect that trend to continue, as well.

For reference, SOC IC units have grown at about a 10% compounded rate over this same period.

And SOC IC revenue at about a 7% compound rate.

Therefore, for all the productivity enhancements being introduced in test equipment and methodology over the years, we still have a growing market.

Over a 2-year cycle, this adds about $300 million to the SOC test market.

While we will get our fair share of this rising tide, our emphasis will continue to be on market share gains to allow us to outgrow the market.

On the cost front, we're going to keep things tight through 2007, as we've made good progress throughout 2006.

And we're accelerating the volume of product as sourced and low-cost regions to keep driving on this during this year ahead.

In summary, the short-term picture is one of caution and digestion in our semiconductor test customer base and lumpy demand in our system test businesses.

This caution in the semiconductor test market could continue for another quarter or two.

Our customers' focus will be on technology buys for new silicon.

Our focus will be on our core SOC test platforms, the FLEX and the J750.

And we'll continue to drive on improving our productivity while maintaining our breakeven discipline.

Let me turn it back over to Greg.

Thanks, Mike.

We close 2006 with fourth quarter sales of $263 million, down 27% from a quarter ago.

Our fourth quarter EPS of $0.06 compares favorably to our guidance, which was for EPS of between $0.01 to $0.05 on sales of 255 to 275 million.

This EPS improvement from our guidance was primarily due to lower operating expenses.

We exited 2006 operating at our 2007 breakeven target, which we announced on the last call.

I'll talk more about this shortly.

First, though, as we conclude 2006, there are four items that are worth highlighting.

First, in 2006, we generated $344 million of free cash flow.

This was 59% higher than our prior record of $216 million, which was achieved back in 1999.

Our full cycle of free cash flow performance is on track to be better then at any time in our past.

Next, in 2006, we had just over $100 million of SOC test market share gains with the FLEX and the J750 lineup across a wide range of device segments.

This equals about three points of SOC test market share gains in 2006.

Third, in our System Test Group, after flat, top-line performance in recent years, we grew year-over-year sales by about 10% while also growing PBIT.

And lastly, our earnings per share from continuing operations of $0.88 on a non-GAAP basis and $1.03 on a GAAP basis are our best EPS results since the bubble of 2000.

Behind this earnings per share improvement is much improved productivity.

Our gross margins reflect an improved manufacturing model with much more outsource to low-cost region suppliers as well as much faster final configuration and test cycle times inside of Teradyne with FLEX and J750.

Our final configuration cycle times are down from five weeks with prior generation products to five days with FLEX.

Our engineering costs are also lower with a single product platform and greater use of low-cost region engineering.

We now have about 25% of our semiconductor test engineering headcount outsource in India.

In summary, the long list of productivity improvements has allowed us to reduce our real estate footprint by about 25% over the past 18 months.

Now, let me turn back to our breakeven level for a moment.

We earlier announced that we would move our targeted quarterly operating breakeven level down from 250 to 260 million a quarter in 2006 to 240 to 250 a quarter in 2007.

You'll recall that this breakeven target excludes stock based compensation and special items such as restructuring costs or real estate gains and it's at the PBIT line.

As stock based comp will now be included in our most recent fiscal year, we will now talk about operating breakeven, including stock based compensation.

Therefore, we are resetting our 2007 quarterly breakeven target to 250 to 260 million to include about $12 million a quarter in breakeven dollars for stock based compensation.

We exited 2006 operating at this target level and expect to continue throughout 2007.

Now, if I match our 2006 exiting breakeven level targets against our average quarterly sales of about $320 million, we are at about a 10% PBIT rate on average.

And again, this includes two points of stock based compensation.

Our plan to get to our model of 15% PBIT over a full cycle is largely tied to three initiatives.

These are growth in existing markets.

We expect the share growth that we had in semiconductor tests in 2006 to continue.

Second, we have investments in adjacent markets that we expect to add to our existing share growth plans.

However, we won't talk further about these adjacencies until we have a product and a first customer.

And lastly, we have a set of productivity improvements that should also contribute to the 15% PBIT model.

These include, for instance, getting more of our PC boards manufactured in low-cost regions.

Now, let me take you through some of the details of the fourth quarter and then our guidance for the first quarter of 2007.

Our fourth quarter gross margin percentage was 46.2% of sales, down 2.6 points from 48.8% in the prior quarter, due primarily to lower sales volume offset to a lesser extent by favorable product mix.

R&D expense was 49.6 million or 18.9% of sales, as compared to 53.3 million or 14.8% of sales in the third quarter.

SG&A expenses were 69.5 million or 26.4% of sale, as compared to 71.8 million or 20% of sales in the third quarter.

The operating expense declines were primarily due to lower year-end compensation and fringe benefit costs.

Our net interest income was 10.3 million, up from the prior quarter, due primarily to the repayment of our convertible debt in October.

Our income tax expense was offset by a one-time tax credit in the fourth quarter.

Our quarter ending headcount was about 3800 employees.

Year-over-year our headcount is down 5% from 4,000 to 3800 people, while annual sales were up 28% from 1.1 billion to 1.4 billion.

In the fourth quarter, semiconductor sales were 75% of the total, assembly test 16% and other test 9%.

On a geographic basis our fourth quarter sales in descending percentage order broke down as follows -- U.S. 30.7%;

Europe 15.3;

Southeast Asia, 13.4;

Japan 12.9;

Taiwan, 12.2;

Singapore 9.0;

Korea 3.6; and rest of the world 2.9%.

We had net bookings of 287.5 million in the quarter.

On a quarter to quarter basis our bookings were up 20%.

Semiconductor test total bookings were up 11%, while semiconductor test product bookings were down 11%.

Assembly test was up 22%, other test was up 126%.

On a year-over-year basis, that is Q4 of 2005 to Q4 of 2006, total bookings were up 11%, semiconductor test was up 15%, assembly test up 9%, and other test down 17%.

Our book-to-bill ratios were 1.09 for the overall Company, 1.03 for semiconductor test, 1.22 for assembly test and 1.37 for other test.

At the end of the quarter, our backlog stood at 339 million, of which 73% is scheduled to ship within the next six months.

This compares to 315 million at the end of the third quarter, of which 84% was scheduled to ship within the next six months.

On a geographic basis, our bookings for the quarter, again in descending percentage order, were distributed as follows -- U.S. 37.0%;

Europe 18.0;

Japan 12.2;

Singapore 10.0;

Southeast Asia 9.2;

Taiwan 8.5;

Korea 4.5; and rest of the world 0.6.

Now moving to the balance sheet.

We ended the third quarter with cash and marketable securities of 944.6 million, down 201.6 million from 1.146 billion as of the end of the third quarter.

We used cash to pay down the remaining balance of our convertible debt of 261 million and repurchased 2.1 million shares at an average price of 13.52, totaling $28 million in the fourth quarter.

To date we have repurchased 10.6 million of our shares at an average price of $12.96, totaling $138 million.

This leaves us with the remaining authorization for our share buyback of $262 million coming into the first quarter of 2007.

In the fourth quarter, capital additions, net of sales related capital equipment, were 17.4 million, depreciation and amortization for the fourth quarter was 23.7 million.

This includes about 6.2 million for stock based comp.

Accounts receivable stood at 159 million or 55 days sales outstanding, down from 59 days in the third quarter.

We ended the quarter with product inventory of 93.1 million, down 3 million from the end of the third quarter.

In the first quarter of 2007, as Mike mentioned, we expect sales to be between 245 and 260 million, with net income per diluted share between $0.00 and $0.03 per share.

We expect gross margins to be 44% plus or minus a half a point.

R&D should run between 19 and 20%.

SG&A should run between 25 and 26%.

Expect to be about flat in inventory dollars and accounts receivable to increase about 20 million.

In addition, we expect to spend 20 million or less on capital.

In the first quarter our depreciation and amortization should be around 25 million, including $7.7 million for stock based compensation.

Our tax rate for 2007 is expected to be about 17%.

Looking ahead, we plan on holding breakeven level about flat and continuing to make progress against our growth plan and operating model.

Now I will open the call for questions.

Operator?

[OPERATOR INSTRUCTIONS] Your first question comes from Chris Blansett with JP Morgan.

Hi, guys, thanks for taking my call.

Had a quick question on the composition of the equipment for semi ordered during the quarter versus what you revenued during the quarter.

Kind of see what the mix change looked like?

Product mix, Chris?

Yes.

Greg, do we have that?

Yes, in bookings we had for semi test the products were 133, the service was 70 million.

So there's surge of service bookings in the fourth quarter of 2006.

I was kind of trying to understand if the composition of what you booked as far as equipment has changed dramatically based on the, maybe weak consumer for Q1?

Let's see, Chris, the -- I think you're -- are you asking about segments, device segments or are you asking about product?

Kind of like the product mix, FLEX versus J750.

Okay.

Good.

Sorry.

The mix has shifted through the year so that our legacy product line, catalyst and tiger products, are now in single digit percentages of our total bookings and shipments.

So we're into the full FLEX and J750 era.

We don't breakdown what our bookings were by product below that, but obviously both of the products, the 750 and the FLEX products, are the mainstream products now.

During the course of '06, I don't have the numbers in front of me, but I think we were pretty close to 50/50 in terms of unit sales.

In other words, FLEXs and J750s were almost exactly in balance.

And then quick, kind of some details on what test utilization rates were for the fourth quarter and maybe what you think they're going to be for the first quarter?

Yes.

Utilization has been trending down, as you know, through the second half of this year.

The outside agencies that track this have a downward slope as we do in utilization.

It's a pretty gradual slope in the second half of this year.

They're projecting the utilization bottom sometime in this quarter, but those are projections and those are obviously outsiders.

We've seen utilization declining less at IDMs now than at subcon.

The IDMs are essentially flattening out, a little bit of erosion, but it's really very low single-digits quarter to quarter.

And during the course of the year, it's been about -- it's between five and ten points in total from the beginning of last year to the end of this year.

A little bit steeper on the subcon side and that's still going down.

But again, these are in single-digits quarter to quarter.

So you're seeing a much more gradual decline in utilization rates.

Thanks, guys, I appreciate it.

Yes.

Your next question comes from Dave Duley with Merriman Curhan and Ford.

Yes, congratulations on the continued market share gains.

And on that subject, I was just wondering if you kind of implied that you would continue these market share gains.

I was wondering what application [technical difficulties] you felt that you could continue the momentum in?

Or is it more of a case of the big sockets that you won last year are ramping into production?

And then real quickly as a follow-up, could you just talk about what you expect the cash flow from operations to be for March?

And would we expect the R&D budget to continue to be as high as a percentage of revenues?

Thank you.

Okay.

Dave, on the market share end, obviously we're hoping for continued gains.

The granularity on share gains is very, very fine.

In other words, there are no cases where socket gives you a point.

Right, sockets are all fractional share point.

So it took scores of socket wins this year to accumulate to this $100 million plus net improvement in our in our design-end rate.

There are between 20 and 25 different segments that we attack in the market from RF to microcontrollers, image sensors, base band, power management, et cetera.

So it's a real spread out offensive.

And we've got about 500 field -- apps engineers who are involved in that process.

So there's no -- it's a relatively dispersed and flat parado of opportunities.

And we don't see any single segment as being the target area for gains.

It's going to be pretty broad as it was in '06, we expect it to be in pretty broad in '07.

Okay.

And the cash flow and R&D stuff?

I'll take the -- the cash flow for Q1, we expect cash flow to be negative about 25 million in the first quarter.

The reason it goes negative in the first quarter is we settle up our employee liabilities for variable compensation, profit sharing and so forth and that's about $45 million.

And the receivable balance also was quite low this quarter.

Our collection efforts were better than you'd normally expect.

So with paying employees liabilities and receivables moving a bit back normal, we expect free cash flow to be negative by about 25 million.

In terms of your question on R&D, the R&D level, we're actually going to put a little bit more in R&D in the upcoming year.

And as a percent of revenue, it certainly appears high now because we're in a low level of sales.

But over a full cycle, our model is to get to 15%.

But we see some opportunities that are very interesting to us in 2007 for R&D that we're going to pursue.

Would that be in those 4 and $5 million high speed DRAM testers?

Well, we're not identifying where the R&D is going, but we obviously have a broad range of instrumentation programs that are underway.

Thanks.

Yes.

Your next question comes from Mehdi Hosseini with FBR.

Yes, thank you.

I wanted to go back to the services part of the semi test business.

Of the $70 million, what's the length of the contract and has there been any change in the contract here?

Because it seems like it has ticked up significantly in 2006.

The services in semi tests have, in fact, been growing, Mehdi.

We've been able to grow our application engineering business to get more direct business from the customers as our engineers are quite productive in writing test programs quite efficiently.

That business has been growing.

In total, though, it ranges from about, in terms of total sales, around 18% of total sales, could be 16 one quarter, but it's in the upper teens as a percent against total revenue.

Is that a one-year or two-year contract?

Well, the contracts tend to be -- many are one-year, some can be longer.

But then there's also contracts, Mehdi, for application engineers, which are shorter contracts to fulfil a fixed price device program or something to that effect.

So they range.

But generally the surge you saw would be annual contracts.

Now, if I were to look at the average length of the contract in '06, has there been any change compared to average in '05?

No.

No.

Okay.

Thank you.

Your next question comes from Timothy Arcuri with Citigroup.

Hi, guys.

Actually had two things.

Number one, you guys are saying that you gained about three points a share across the year.

If I take the worldwide sems data and I -- of course it's not, it's not out yet for the end of the year, but if I look at how that trended through the year, it looks like you gained share in the first half, but you lost share, actually quite a bit of share in the second half of the year.

So I'm wondering A, if you agree with that, and B, where that share's going?

Well, I don't think there's anything significant about the inter-quarterly share trends.

Overall, the, as we said earlier, the market has grown to about 3.3, 3.4 billion in the year, about a 20, 21% growth in the market.

And we're projecting.

That's our projection about the size of the market.

That doesn't shake out until actually March-April time frame, when all the numbers are in from all of the players in the segment.

But our revenue gains here have been in north of 30%.

We don't think there's any question about having made some ground up on the share side.

And as we look at it from a design-end piece by piece basis, that tracks.

So I don't think there's anything significant in the timing during the course of the year because, if you will, the quarterly share shifts are much more a function of buying patterns of customers than they are of significant shifts of accounts inside each quarter.

Okay, Mike, that makes sense.

Maybe just a strategic question, then.

If I look at this last cycle and I look at what you earned across the cycle, so I kind of normalize your earnings across the cycle, you earned roughly less than $0.70, something in the $0.65 range averaged across the cycle.

That was in a period where you had a big successful restructuring, we had record IC unit growth, you also gained market share with FLEX, and you earned less than $0.70.

So I wonder, I certainly understand that you're going to expand into some adjacent test markets, maybe DRAM and maybe some other markets.

But I wonder if there are other secular pressures here, such as maybe advance probe cards or something like that, that maybe strategically the industry should be looking outside of test for M&A.

So I guess the question is, are you more likely to stay within test or do you agree that there are broader secular pressures that maybe will force the test industry to look outside of test to try to remedy the lack of earnings.

Tim, a lot in that question.

But let me try to cover the main things that you're asking about.

We think there's really the three-pronged effort that Greg was talking about.

Actually our first focus and our short-term focus is around the core SOC share growth.

Secondly, we see that we've -- we have a need to expand our TAM and there are segments near in adjacencies, high leverage adjacencies, that over the next few years we're going to be able to exploit.

So there's no question in our view that this 5% growth in the trend line here, the compound annual growth rate in SOC devices doesn't support the kind of growth that we need to have.

So we've got to expand our share in that segment and then we want to expand in some adjacencies out into the future.

I do want to comment on your question about -- your question's really about is there going to be compression in this market that is caused by technologies that allow more efficient testing, more productive testing.

Your question was around advance probe cards.

In a nutshell, the parallelism that's potential in probe, to higher density probe technology, has a dramatic effect on probers.

So the CapEx on probers would be affected pretty dramatically as that advanced.

It would have some effect on testers because, obviously, testers become more efficient, at the same time the testers have to be geared up to handle higher parallelism.

So there's broader instrumentation sets in those.

And then third, for us the more parallel applications there are actually placed towards us, because we've got a good parallel test architecture, so the more sockets that are evaluated on the basis of overall cost to test, we think is some advantage.

But net/net, I don't disagree with you that we are looking at how we have to grow this business.

And we think there is actually expanded TAM opportunities in and around test.

So I don't think you'll see us outside the test arena.

Okay.

That makes sense, thanks.

Yes.

Your next question comes from Gary Hsueh with CIBC.

Hey, guys, thanks for taking my question.

By the way, good job here on managing the current inventory correction, especially in terms of profitability.

It'd be nice if this is the first cycle where we don't see you lose any money in any one quarter in the cycle.

Basically my question is back to service.

If I look at service again over the last two years, you've basically gone from roughly $50 million to now you're in the mid $60 million level and margins and service have expanded from basically the mid teens to now the high 30% range.

So tremendous job there.

I'm just wondering, based on your service bookings seasonally going up in Q4, what is that number up year-over-year relative to Q4 of '05?

And what does that mean when you look at your backlog in terms of what your service quarterly revenue run rate could get to in '07?

The increase is about 20%, I think was your first question.

And was your next question what's the run rate of service revenue?

Yes, 20%, if I just apply that simply over your quarterly run rate in '06, that gets you to something like a mid $70 million kind of range in '07.

I'm sure that math is wrong, but try and help me out here with the math on what service could be on a quarterly run rate in '07.

I'd expect it to be in the 60s.

And it can fluctuate based upon the level of test buying.

But I think the low 60s is probably where you'd see it early in the year.

And there is a plan to continue to grow it, so we just need to see how that plays out.

Okay.

Okay.

And what about margins there?

Is there any leverage or incremental leverage in service margins, particularly gross margins, once you start to kind of potentially move the needle beyond the low $60 million in the back half of '07?

There should be with greater scale.

We have done a lot already, though, to improve service margins.

Service has multiple businesses, one is board repair and 90% plus of our boards are repaired in low-cost regions.

We've done some work on pricing, as well.

It's a set of things that could provide additional leverage as you get more scale, you should also do better.

So I think you could -- you could expect some improvement in margin as we grow the business.

Okay.

I'm got another question here.

Separately on OpEx on 2007.

You're basically bringing OpEx down in the March quarter in absolute dollars terms, I guess that's no rent-backs with the Boston building and the IT outsourcing initiative and India kicking in.

But it sounds like basically you'll be keeping SG&A kind of at that mid $60 million kind of run rate throughout '07 and then maybe in the back half or middle of '07, you start picking up the R&D spend.

Is that the right way to kind of model '07 OpEx?

I think that's close.

I think the R&D will go up just a tiny bit in Q1, but that's right there, R&D likely go up a little bit more towards the go into the year and SG&A should stay in about that range.

Okay.

Perfect, thanks, guys.

Your next question comes from Mark Fitzgerald with Banc of America.

Just on the outlook for growth here.

If you looked at your numbers post 2000, looks like the business on an industry wide basis and your own basis has actually shrunk given what happened through most of the second half of the 1990s and with the era of the kind of high performance tester.

So I guess my basic question here is do you really have to rely on going outside your traditional [ASO] site and see tester space to get this back to being a growth story because taking market share in a shrinking market really doesn't get you the kind of top-line growth story.

Mark, we don't think it's a shrinking market, but we do think that the growth rate in the market won't be sufficient to get the kind of growth that we need.

So the answer's a yes that we do need to go outside the market.

And rather than debating is it up 3%, 4%, 5%, 6% per year, any of those numbers are not sufficient to what we want to do over the next few years.

So expanding it to some adjacencies and really expanding not with a long shot, meaning where we have to build something that is totally new, but rather expanding where we can get some very, very high leverage out of this platform, FLEX hardware software platform, and the 750, that's the way we're going to do it.

Just to follow-up on that .

If you looked, historically up through 2000 you readily did quarters where you did 4 or 500 million in a quarter in semiconductor test, and post 2000 you haven't come anywhere close to that.

So I don't get the kind of it hasn't shrunk scenario.

I think if you looked at it on a trendline, Mark, you might get to a different conclusion versus the bubble of 2000.

If you went back 10 years and looked at a trendline, I think you'd see that there is a growth rate of about 5%.

But if you start at 2000 and not further back, you'd reach the conclusion you reached.

Our ultimates work where you start your trendline.

That's right.

The 4.8 billion that was in 2000, the market is nowhere close to that.

The last three years, though, have been the three highest years, except for 2000.

So we all wish we'd get back to2000, but that's not realistic.

There is agreement that we need to get into adjacent markets as well as grow in SOC tests.

So, I think we agree with your principal point.

Okay.

Thank you.

Your next question comes from Edward White with Lehman Brothers.

Thanks, I was wondering if you could talk about recent market share dynamics and market share outlook in systems test?

Okay.

The question on systems test is a multi-faceted one because there is multiple businesses there.

I'll comment on two areas that are important for our midterm growth plan.

The mill arrow business we have is a program, long-term program business that results in revenue streams that go out over multiple years, five, six, seven years.

And our mill arrow business has been very successful this year in the program wins that we've had.

In addition, we have expanded into the commercial airlines segment of that business.

So there's an adjacency inside mill arrow that's a growth engine for us out over the next few years.

And we've made penetrations into that segment during the course of 2006.

The second area of growth potential for us has been in the diagnostics systems, automotive diagnostics business.

And we've had a record year this year in diagnostics systems.

Highest revenue and profitability in the history of that division.

And as you know, they provide test solutions in the service bay as well as in the manufacturing lines.

And our growth has been very strong on the service bay end as our customers there are trying to work down their warranty related costs.

We've got a hardware and software platform solution there that is expanding, and a product called VMM, vehicle measurement module, has had record levels of revenue in 2006.

Now both mill arrow and diagnostic solutions is -- it's an up and down program business.

So our objective there is to be able to get those businesses to model profitability, which we've done in 2006.

And then as the programs over a trendline grow, those would be good profit engines for us.

Okay.

Second question is, Mike, you mentioned that -- you talked about the fact that the near-term environment with your customers is uncertain.

What sort of indicators would begin to tell you that you've hit the bottom of that?

In other words what do you look at?

Do you think it'll come from the subcons or from your big IDM customers, or what kinds of things would you expect to see that will give you that, the hint that it's bottoming, that it's bottomed out.

There are not too many early indicators because our customers get surprised and then they turn back to us.

Obviously, the -- if you said what would I look for if I had one thing to look for.

I'd look for some shift in the quote cycles, in the quotation levels for our subcons.

Because that's really more the canary in the mine shaft in terms of the sensitivity to shifts that may occur.

Because it would signal that the IDMs and the fabless companies are moving back in, going towards the subcons rather than pulling back in-house in the case of the IDMs.

That's the one thing I'd look for.

Okay.

Great.

Thank you.

Yes.

Your next question comes from Pat Ho with Stifel Nicolaus.

Thanks a lot.

Just looking at the gross margins going forward.

It looks like it held pretty well in Q4, but just given a little bit of a revenue decline in Q1, it looks like a steeper drop in gross margins.

What can that be attributed to, is it pricing or is it just product mix?

It's mix.

We actually had favorable mix in the fourth quarter that brought our margins up a little bit better than what you might have expected given the revenue decline.

And then Q1 is going the opposite way.

So therefore, when you go from Q4 to Q1, it actually appears to be a more significant drop.

But in terms of where you might expect it to be, Q1's a little bit off and it's due to mix.

So there isn't any additional pricing pressures that are hitting you in Q1?

No, the pricing pressures we have in this industry have been continuing for a long period of time.

And we do a lot of work on the other side to offset those through low-cost region sourcing for material and other actions to ensure that the margins stay about at the right level.

Right, and I just wanted to get some clarification on, I think, Mike, you made the statement in terms of expanding into adjacent markets.

Any, I guess, initiatives that you take on that front are going to be leveraged off of your current platforms.

You're not going to create something entirely new, is that a correct way to look at it?

Yes.

Great.

Thank you very much.

Your next question comes from Steven Pelayo with HSBC.

Yes, guys.

First of all just a quick clarification.

On the utilization rate number you talked about declining in the single-digits, I'm sorry, what absolute level are we at here?

Are we at a very low level that we need to see significant rise before we start buying testers again?

So if you could just clarify absolute level?

And then I had one more question.

Okay.

The absolute levels that are -- that are reported in the world are between the 85 -- in the 80s. 80-90%, obviously, moving down from the top end of that range, but down into the 80-85 band.

And then if you could clarify a little bit for me, I'm sorry I'm away from my model right now.

I'm trying to understand your service bookings seasonality and the lumpiness of your systems business.

We're trying to model, I guess, here your order outlook here.

Obviously, just trying to look only at those two parts.

What's a normal level?

Do you need to go below a normal level in the March quarter then to account for the lumpiness.

Help me to try to quantify at least those two pieces, even though they are the smaller pieces of the total pie.

This is service bookings?

Service bookings and then the lumpiness of the system bookings and what does that likely have to mean?

Then what's a normal level going into the March quarter?

Does it have to go underneath the normal level because you've had such a big lumpy chunk in the fourth quarter?

The service bookings in the first quarter would be certainly less than the fourth quarter.

If I take you back to the third quarter service bookings, they were about 44 million.

So I think it would revert back to a number in that neighborhood, whether it's a little bit less that's possible.

On the system side, where would you think that would?

We're not going to comment on system product bookings.

The revenue level in the non-semiconducted system test business is $60-ish million a quarter. 60-70 million.

Mike, if I could just ask you also, we just heard from TSMC an hour or so ago.

They are talking about industry growth for the semiconductor industry more in the mid single-digits.

Could you just walk me through a little bit about your assumptions for the market growth this year?

And maybe if you could also talk a little bit about some of the secular pressures that are facing the industry.

I think we all worry about the large number of competitors in this space and some of the pricing pressures that we think are going on.

Can you help us just try to understand what growth assumption you have for '07?

Is that a function?

Is some of that been kind of mitigated, if you will, because of pricing pressures in the industry?

Can you just try to help us understand your outlook for '07 and what's behind it?

Sure.

Well, we don't project in a year what the size of the market's going to be.

We do over a cycle and look at it and if you looked at the buy rates, you're asking around buy rates over the last few years, that has orbited around just under 2% in SOC test.

So 1.9 is a number that -- that's been represented over the last few years. 2006 actually will -- we had a 3.3 or $3.4 billion level.

That would be about the right level in 2006.

So while it has felt like an up cycle from a buy rate standpoint, it hasn't been way outside the normal band.

We think that's a -- that's one way of triangulating and we've tended to model and say that it can operate between 1.5 and 2.5% with an average of around somewhat lower than this 2% level.

Now, the long-term trends, the device unit trends are the big driver and that has grown in double digits compound annual growth rate.

The ASP compression you have in devices cuts that not quite in half in terms of the semiconductor device rate.

And then productivity of testers cuts that about in half again.

So if you had a 10% growth rate in device units, you might get something on the 3 or 4% level.

So we look at it over a multi-year period.

And we expect that if you took this buy rate, number one, and you took the growth rates in units and device dollars you could triangulate to a market that right now has been about 3 billion -- is around 3 billion, $3.4 billion level.

And we think there's a story that that moves up, as we said before, low single-digit 3%, 4%, 5% growth rates per year.

Now, the pressures, obviously, are what you described.

They're pressures around multiple competitors and more importantly the pressure on productivity of increasing throughput on the testers is the main thing that's -- that keeps the compression on the buy rate and on the CapEx rate.

But over time, we think that grows.

If you're at $3 billion that could grow 100 to $150 million a year over the next few years.

Have you seen any acceleration.

I'm just curious on the pricing pressure side.

Obviously, the productivity and the throughput you guys are giving, but just everybody moving to single platforms, not so much big iron anymore, the efficiencies of subcontractors, has there actually been -- and just all the players that are out there, has there be an acceleration on the pricing pressure side of the secular worry that we're facing?

There's definitely a -- all of us have developed testers that are lower cost, lower capital costs.

Before I was saying we've got 2300 J750 testers.

Those are significantly lower capital costs than a catalyst, than a predecessor.

So there's definitely CapEx efficiency here in the design of smaller testers.

But I think the world will consume both general purpose testers that have stretch capability across multiple of these 20 plus segments and they'll buy testers that are focused inside one or a few of those segments.

And the era of CMOS technology has been applied to these testers over the last -- last number of years, which has helped on the overall cost, capital costs of the testers.

Okay.

Fair enough, thank you, Mike.

Your next question comes from Raj Seth with Cowen & Company.

Hi, thank you.

Can you tell me in Q1, roughly what would we expect to come from backlog and how much of that revenue number is turns revenue?

We'll get that for you, Raj.

And I guess as a follow-up, does that change?

Does that kind of backlog coverage is it likely to change dramatically across the Europe?

We assumed that bookings stay roughly around the same level that they currently are.

Your first question, we'd expect half of the Q1 shipments to come from backlog.

Okay.

And if I was just looking at the ATE, the core business, is that number meaningfully different?

No, we don't --

I don't think so.

Okay.

And will that coverage -- will that number change quarter to quarter or does it tend to be about 50% every quarter?

It can change at this time period.

It gets about this number, but at other time periods it can vary.

I think there's a lot less, though, Raj, in terms of how the turns business goes.

Going forward there'll be more turns business than there has been in past years and that's just the simple function of five, six, seven, eight, nine week lead times.

We're able to respond much more quickly now.

So the backlog build, even if we get a surge in demand, doesn't go up anywhere near what it has done in past cycles.

Sure.

But then that corresponding means your visibility isn't as good, as well.

Right?

That's correct.

Correct.

Usually you'd be shipping more from backlog, but at this time in the cycle, you actually change the percent and you get more from current quarter bookings than you might otherwise, as a percent that is, not so much by dollars.

I'm sorry, why don't we take one more question?

Your next question comes from Tom Diffely with Merrill Lynch.

Yes, hi.

Just one more clarification on the market share.

What do you think your SOC market share was in 2006?

And is your goal still to get to the 40% range?

We think we're going to end up, once all the numbers are in, 32, 33% in 2006.

And second question was, I'm sorry?

At one point you stated a goal of getting to 40% of the market, is that still internal thinking?

Yes, we don't need to get that high in order to get to our model profitability, but that's our objective.

Okay.

Great, thanks.

Thanks, everyone, we'll talk to you next quarter same time.

Thank you.

Thank you for participating in today's teleconference.

You may now disconnect.