Power Integrations Inc (POWI) 2006 Q1 法說會逐字稿

Good afternoon. My name is Brooke, and I will be your conference operator today. At this time I would like to welcome everyone to the Power Integrations conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. [OPERATOR INSTRUCTIONS] I would now like to introduce Mr. Joe Shiffler, Director of Investor Relations. Thank you, Mr. Shiffler, you may begin your conference.

Thank you and good afternoon. I'm Joe Shiffler, director of IR and Corporate Communication for Power Integrations. With me on the call today are Balu Balakrishnan, President and CEO of Power Integrations, and Rafael Torres, who joined us on July 19th as our Chief Financial Officer.

Earlier this afternoon we issued two press releases -- one addressing the status of our listing on NASDAQ and the other outlining our preliminary second quarter financial results and our outlook for the third quarter. These press releases are available on our website www.powerint.com, and have been e-mailed to those of you on our distribution list.

Before I turn the call over to Balu, I'd like to note that our discussion today including the Q&A session will include forward-looking statements reflecting management's current forecast of certain aspects of the company's future business. Forward-looking statements are denoted by such words as "will," "would," "believe," "should," "expect," "outlook," "estimate," "anticipate," and similar expressions that look toward future events or performance. Forward-looking statements are based on current information that is, by its nature, dynamic and subject to rapid and even abrupt changes. Our forward-looking statements are subject to risks and uncertainties, which may cause actual results to differ materially from those projected or implied in our statements. Such risks and uncertainties are discussed in our most recent reports on forms 10-K and 10-Q filed with the SEC. With that I'll turn the call over to Balu.

Thanks, Joe, and good afternoon. We have unfortunate news to report today, which is that we will not be able to meet NASDAQ's August 2nd deadline for filing of our 2005 10-K and of our first quarter 10-Q. Under a relatively new NASDAQ marketplace rule, the listing qualifications panel is prohibited from extending our listing more than 90 days beyond their original decision to grant an extension. That 90-day period has expired and as a result NASDAQ must automatically delist our shares.

The decision to delist us is not discretionary on the part of NASDAQ, and it should not be seen as an indication of anything other than our inability to meet the deadline. As I will explain in greater detail in a moment, our filings have been delayed due to new developments that are largely beyond our control.

We expect the delisting to occur as of the opening of the market on August 2nd. Two market makers that are already in place to quote our shares in the pink sheets and others can begin entering quotes immediately. We therefore expect a seamless transition in the trading of our shares. Obviously, we intend to apply for relisting on NASDAQ at the earliest possible time.

While it is difficult to predict exactly when that will be, our best estimate is that the restatement and filing process could take another six to 12 weeks.

On a more positive note, we had an excellent second quarter. Revenue was $41.5 million including a net $2.7 million benefit from the settlement of ship and debit claims with two of our distributors. Excluding this one-time benefit, we came in near the high end of our expectations and delivered the highest quarterly revenue in our history.

We also had an exceptionally strong quarter in terms of design wins. The expected annual dollar value of designs won in the second quarter was more than 20% above our previous high in the second quarter of 2005.

Also setting aside stock option expenses, our gross margin has increased sequentially in each of the past two quarters giving us a strong two-year trend of rising gross margins. This has been the result of our deliberate efforts to reduce product cost and shift our mix toward smaller customers combined with the benefit of a relatively stable pricing environment in the recent quarters.

Our business is performing extremely well. I am proud of the fact that our employees have remained focused on the job at hand and have not allowed the challenges of the past several months to become a distraction. I am confident that this will continue to be the case going forward, and I am very optimistic about our prospects for the third quarter and beyond.

Before we go into the second quarter results in more detail, I would like to return to the subject of our listing status and our efforts to regain compliance with NASDAQ rules. Obviously, delisting is a disappointing turn of events, and we have made every conceivable effort to avoid it. I believe that this outcome is largely a function of circumstance driven by the fact that we were one of the first companies to investigate our stock option practices, a process that is now underway at dozens of other companies.

As a result of being first, we were one of the first to miss a filing deadline and one of the first to bump up against NASDAQ's 90-day limit on listing extensions. By virtue of being first, we have also had to break new ground in terms of process for revaluing stock option trends for [inaudible] measurement base were deemed incorrect. Despite complexities of the restatement, we were very far along in the process when we were made aware of new information clarifying the accounting treatment for stock options whose brand base may differ from the original stated measurement dates.

We have also been advised very recently to submit our analysis of the accounting treatment to the SEC for the review prior to filing upon our restated financials. Given the large number of companies going through this process, these developments are understandable. However, they have come at a very inopportune time for Power Integrations.

While we are asking for additional time, NASDAQ is bound by its own rules and does not have discretion to grant us a continued listing. Naturally, we will continue to do everything within our power to regain compliance as quickly as possible. However, the process is largely beyond our control at this point, and it is difficult to predict when we will be able to apply for relisting on NASDAQ. As I mentioned earlier, our best estimate at this point is that the process could take another six to 12 weeks.

In the meantime, my primary concern is that we remain focused on our business as we have done throughout this process. Our second quarter results were very strong, and we expect robust growth again in the third quarter. We are as excited as ever about the trends in our business and the growth opportunities ahead of us that will not allow and will not allow this temporary setback to throw us offtrack.

Looking more closely at our second quarter results, revenues grew 10% sequentially excluding the impact of shipment debit settlements. All four major end markets contributed to the growth with communications and industrial markets leading the way, each growing in the mid-teens sequentially while the consumer and computer markets each grew in the mid single digits.

Our revenue diversification efforts continue to pay off. Cell phone charges, the market where we see the most intense price competition, now account for barely 20% of our total revenue, while other communications devices such as broadband modems, cordless phones, and voice over IP phones are nearing 10% of our total revenue.

Meanwhile, the highly fragmented industrial segment and consumer segments have grown to 15 and 32% of total revenue, respectively. Our diversification into consumer, industrial, and non-cell phone communications products has increased our exposure to second and third-tier customers. This has, in turn, helped us achieve greater stability in pricing and contributed directly to the improvement in gross margin over the past two years. Our margin has risen despite an increasing revenue contribution from our LinkSwitch products, which are still relatively early in their lifecycle and carry a lower-than-average gross margin.

In fact, LinkSwitch comprised 9% of total revenue for the second quarter and continues to gain traction. Linear power supplies, which LinkSwitch is intended to replace, are falling out of favor due to higher raw material costs, especially copper, and also due to energy efficiency standards such as emerging standards on external power supplies.

California's mandatory standards have now been finalized and are no longer being challenged by industry groups. After the intense industry lobbying effort, the California Energy Commission granted only a six-month delay for cell phones, PDAs and other IT products. These products must comply with the standards by January 1, 2007. Most of the products, including cordless phones, must adopt energy efficient external power supplies by July 1, 2007, in California.

We continue to be excited about our opportunity created by these new standards. While industry lobbying managed to delay the standards, the CEC resisted pressure to relax standards and did not grant exemptions to any major product category. As a result, less than a year from now virtually all products using external power supplies will be subject to California standards.

The standards are quite stringent and nearly all existing linear power supplies will fail to meet them as will many existing discrete designs. We have a portfolio of lower power products that is extremely well suited for applications most affected by these standards such as cell phones, cordless phones, power tools, and many more. And our sales force is working very hard to educate customers about our solutions.

We are also working hard to capitalize on our win at the International Trade Commission, where we received a favorable initial determination in our case against System General. In approximately two weeks, the commission will issue its final determination, and we fully expect that they will uphold not only judge's findings but also his recommended remedies. The infringing parts and any LCD monitors and AC printer adapters containing them will be effectively banned from importation into the U.S. There will be a two-month presidential review period following the final determination after which the exclusion order is expected to go into effect.

However, during the two-month period, the SG will have to pay its $0.38 bond for each chip that comes into the U.S. So for practical purposes, we believe the infringing parts are effectively out of the U.S. market starting two weeks from now.

Our patent infringement case against Fairchild is also nearing completion. The judge has decided to split the case into two trials, the first, which is scheduled to begin on October 2nd will determine infringement and damages with a verdict likely to come soon after the conclusion of that trial. The second will determine the validity of patents we are asserting. That trial is scheduled to begin December 4th and should be fairly brief. We remain highly confident in the strength of our case, and we look forward to these trials.

Looking ahead, we are expecting strong growth again in the third quarter with revenues expected to be between $42 million and $44 million. While we are mindful of concerns about the [indiscernible] of the cycle and uncertainty about the end market demand, we entered the quarter with healthy backlog thanks to strong bookings throughout the second quarter, and we would reach the midpoint of our guidance with turns orders in the low 60s. We believe we are off to a good start as strong bookings continued through the month of July.

The third quarter and future quarters will benefit from an exceptional list of designs won in the second quarter. As I mentioned earlier, the expected annual dollar value of designs in the second quarter was the largest ever by a substantial margin, more than 20% above our previous high. TinySwitch-III is off to a fast start winning several PC standby designs for Dell as well as several appliance designs, a high-volume DVD design, and designs for LCD TVs and DSL modems.

TinySwitch-II also continues to do extremely well winning dozens of designs in the second quarter including a high-volume design for LD cell phone as well as a high-volume design for XM radio and a cordless phone for British Telecom.

TOPSwitch GX-1 designs across a broad range of applications, including printers, servers, DVRs, security systems, and LED lighting. DPA-Switch won designs for several telecom applications including a Nortel power over Ethernet hub as well as the industrial control application.

LinkSwitch designs included a washing machine for ElectroLux, several LED lighting applications, power meters, and even a fish tank. The diversity and breadth of applications we are winning has never been greater, and we are very excited about the number of designs that are coming from small volume applications and smaller customers. This is precisely what we set out to do, and we are very pleased with the success we are having.

At this point, I'd like to introduce Rafael Torres, our new Chief Financial Officer. Rafael joined us on July 19th, coming over from PLX Technology where he had been CFO for about six years. I am extremely pleased that he has joined our team. Rafael?

Thanks, Balu, and good afternoon. It's a pleasure to be here for my first quarterly call with Power Integrations. The circumstances of this call are certainly unfortunate, but I'm confident that we'll get these issues behind us soon, and I look forward to taking part in plenty of more enjoyable conference calls in the years ahead.

Our second quarter results are certainly a step in the right direction. As Balu noted, revenue was $41.5 million, up 17% both year-over-year and sequentially. This included a pickup from the settlement of ship and debit claims with two of our distributors. These claims related to sales that took place in prior periods and will represent a $2.7 million net cash benefit to Power Integrations as well as $2.7 million in revenue recognized during the second quarter.

Revenue mix by end market for the first half of 2006 is 32% consumer; 28% communications; 18% computer; 15% industrial; and 7% other. Revenue by product family for the first half of the year was 53% TinySwitch; 36% TOPSwitch; 9% LinkSwitch; and 2% DPA-Switch. Excluding the one-time pickup, turns orders represented nearly 70% of revenue for the second quarter. Distributors accounted for about two-thirds of revenues for the quarter, all recognized on sell-through business as usual.

We are unable to provide much additional income statement detail at this time due to the pending restatements. However, directionally speaking, our gross margin expanded in both the first and second quarters continuing the trend of the past two years. This is exclusive of stock option expenses.

Patent litigation expenses totaled $800,000 for the quarter, bringing the total to $3.7 million year-to-date. As stated in our press release, we expect roughly $3 million more through the balance of this year primarily related to the Fairchild trial. Since the trial begins so close to the beginning of the fourth quarter, it is difficult to predict how much spending will fall in each of the quarters. Our best estimate for now is roughly one-third in Q3, and two-thirds in Q4.

Second quarter operating expenses also included $4 million related to the internal investigation and the restatement efforts bringing the total to $5.9 million to date. We expect the spending to continue through the duration of the restatement project.

Our balance sheet is extremely strong and shows evidence of our efforts to return cash to stockholders. Our cash balance at June 30th was $125 million, down $14 million from the prior quarter. Share repurchases during the quarter totaled $19 million for which we repurchase 1.1 million shares at an average price of $18. This concluded the $25 million repurchase program announced late last year. Over the past two years, Power Integrations has repurchased 3.4 million shares for a total of $65 million. This has resulted in a net decrease of 2.2 million common shares outstanding, a decline of 7%.

Regarding the outlook, as Balu noted, we currently expect revenues of $42 million to $44 million for the third quarter, or sequential growth of roughly 8 to 14% excluding the one-time revenue benefit in the second quarter.

And now I'll turn it back to Joe.

Thanks, Rafael. Before we take questions, I would like to note that we do intend to be back out on the road meeting with investors over the next several months as circumstances permit. Currently, we plan on attending the RBC Technology Conference on August 10th in San Francisco, and the Citigroup Technology Conference September 7th in New York, as well as a number of additional conferences later in the year. We look forward to seeing many of you in the months ahead.

And now, Operator, would you please give the instructions for the Q&A session.

[OPERATOR INSTRUCTIONS] Michelle Ha, Ferris Baker Watts.

I'm calling in lieu of Mark Donohue. You mentioned that previously that the first half for the FAS 123 expenses would be in the realm of $5 million for ironing these issues out. So far, you've burned -- so far you've spent 5.9, is that right?

Okay, just to be accurate, it is not the FAS 123 charge, but rather --

I'm sorry, the stock backdating.

So what we had said was that the special investigation would cost us between $4 million to $5 million, but we did not -- that did not include the restatement efforts, which have a lot of additional people involved and also more audit time and so on.

What do you expect to spend for the full year?

The only thing we can say at this point -- it's Rafael -- is that it will continue at roughly the same clip depending on the timelines that we may hit. So our current estimate is that it will take us between six and 12 weeks to complete the whole effort. So that's as much information as we can share with you at this point in time.

So you've spent 5.9?

Correct.

Sumit Dhanda, Banc of America Securities.

On the costs there, the litigation costs, I thought, came in a little lower on Q2, but they're ramping up again the back half of the year. So if you could help us understand exactly what occurred there. And then how comfortable do you feel with the notion that we'll finally see an end to the litigation efforts by the end of this year, because it seems like it's been pushed out a little bit over the course of the last two quarters.

Yes, I think the main reason the litigation costs are higher than we originally projected is due to the splitting of the trial into two trials. The judge has decided that he would have two trials for various reasons. So that does increase our cost.

And as far as we can tell, those dates are agreed to, and it's not likely to change, and therefore by the end of December, so by the end of the fourth quarter, the Fairchild trial and associated costs, to a large extent will be over. There will be very small residual costs related to that trial, going forward, but all the material costs would be completed at that time.

Okay, and then just along the same lines, you know, it seems like a lot of expense has been allocated to the litigation both against System General and then Fairchild, too. I mean, do you expect some of this to be offset by a future revenue stream, which you had potentially lost to these two competitors because of their infringement on your patent? Or not really? Is this just a future hedge against encroachment?

Before we initiated either of these lawsuits, we went through an extensive analysis to make sure it is worth the money and the time in terms of the return it will provide over the long term. So we are very confident both of these cases will have a return that is significantly higher than the money we have spent. This is not an emotional response to infringement.

No, I understand that. I'm just saying can you help us quantify this in terms of what you might be seeing six to 12 months out as a result of potentially this favorable ruling against System General.

Unfortunately, I can't quantify it because it's part of the damages claim, and then we are not allowed to estimate those publicly. So the attorneys have specifically told us not to discuss damages.

And then the restatement, you know, the restatement expenses that relate to the investigation, I mean, again, it seems awfully high -- another 6 million in the back half of the year. I understand all of the special circumstances that you are being subjected to, but it seems like the investigation is going on for an inordinately long time. I mean, help us understand why it's really taking so long. The other companies which are in the same situation, granted, larger in scale, but they seem to have had better success in closing this issue too. Some kind of a preliminary --

Sure, just to clarify. The investigation is over. It was over in May, and we started the restatement work soon after the investigation was complete, and we were very far along before we had this new development, which has basically delayed us significantly. And in terms of the total cost, it is a very complex process, and we want to do it accurately, we want to do it correctly, and it is a number of years, as you know, and it's a very complex set of events and, as I said, we were very far along in that process before these developments came in.

Just let me clarify, Rafael, because you said it is going to continue, the costs are going to be at the same clip as you saw so far, 4 million or so in Q2. Unless you think this is going to spill over into Q4 --

-- in our system --

[inaudible] additional costs before and not 6 million.

Yes, and my comment wasn't that it will be the exact amount in the second half. My comment is it will continue at the same rate. So if you use the assumption that we'll close between six and 12 weeks, the costs would be lower.

Closer to 4 million in that case?

Yes, I couldn't give you a number, but it would be lower than 6.

Okay, okay, and a couple more. I apologize for the multiple questions here -- the design wins that you talked about -- in terms of you gave some color here, but LinkSwitch 9 % of revenues now, but it almost seems like the design win commentary around TinySwitch-III and TinySwitch-II is better, or am I misreading?

I'm not sure why you think that way.

Because you gave us some quantitative data on TinySwitch-III and the customer profile there than TinySwitch-II.

Right, so --

[inaudible] on LinkSwitch here because it speaks to higher volumes [inaudible].

I don't quite understand. I think you are on a cell phone or something. Did you understand the question, Joe?

Yeah, I mean, I think that -- we didn't mean to indicate that LinkSwitch wasn't doing as well as TinySwitch. I think that both products are doing extremely well, and as you've seen, the percentage of LinkSwitch -- LinkSwitch, as a percentage of our revenue, has been increasing steadily over the past few quarters.

Remember, the lower [ASC] products in terms of volume, it's even higher. So LinkSwitch is doing extremely well.

Okay, and then just some numbers, which I may have missed out on. Could you repeat what the [inaudible] growth of the different end markets was in the quarter?

We said for industrial and for communications, it was in the mid teens, and then computer and the other two segments was mid single digits.

Okay, and then is there anything to read in your comments that cell phones are barely 20% revenues. Was the growth within communications less cell-phone centric this quarter, or more, or -- ?

Cell phones also grew sequentially, but the other segments grew a lot better.

Okay, and then, going forward, as it relates to these design wins, call it indirectly or directly related to the CEC mandate, where do you see the most opportunity by end market.

The two major ones are cell phones and cordless phones followed by tools and a large number of fragmented applications, where we are already getting design wins as we indicated earlier.

And the last question I had -- I know there was an issue with a customer program ramp last quarter. Could you just update us on what happened there? Thank you.

That program has gone into production, and our Motorola revenue has substantially increased in Q2.

Steve Smigie, Raymond James.

I was wondering if you could comment a little on what factors might determine the six to 12 weeks versus perhaps longer? Or what would determine the boundaries, the ranges?

Predominantly, it will be the review by SEC. We don't know exactly how long it will take.

Okay, and they're reviewing the new standards that you've used in your information or could you help me understand a little bit better the process that they'll be going through?

Yes, they will review the assumptions that we use on measurement dates. So we will submit that to the SEC for review, and they'll have an opportunity to comment.

Okay, and I was hoping you could comment a little bit on what your year-over-year growth rates look like for the handset and for the computing category?

Year-over-year growth rates, you said?

Yes.

Let's see -- the computing segment, I would say, was down by sort of a high single digit kind of percentage year-over-year. As you know, we've seen some weakness in the desktop PC market. And which was the other segment?

The handset.

Well, I'd say the communications segment year-over-year is up by about a high single digit percentage.

Okay, you guys had doubled your sales force some time ago, and I was just curious about how the pipeline of products is related to that. I mean, if the accounts that the new salespeople have gone onto, have you seen design wins start to roll in yet from that? Is that part of the reason for the significant increase in the design-win dollars that you have?

Yes, that is definitely part of the reason we have had such a strong design-win quarter.

Kevin O'Boyle, KCO Investments.

I guess one more question on this restatement. You've alluded to new developments, and I was wondering if you could run through again exactly what are those new developments? Because it sounds like you have the restatement done, but now you're going through a review process, and I just want to clarify this.

There were really two -- one is -- we got some clarification on the accounting treatment for stock options, which may have different measurement dates. And that came at a relatively late stage, very recently, actually. And then the second one is the strong advice we got to get the accounting treatment reviewed by SEC before filing our statements.

Okay, and so, before, you didn't think you were going to have to go through an SEC review process before filing?

Correct.

Okay. Now, if I could just turn to the business -- you seem to be a lot more confident now that the standards in California are set and the timelines are set, and so I'm wondering if you could give us a sense on how you expect the market to transition from linear transformers to LinkSwitch or something else, and sort of what should we be looking for to sort of monitor the progress there?

Okay, so it varies a lot from customer to customer. Some customers are going forward with the transition independent of the delay in the CEC because, for one thing, they are already way ahead. Secondly, the cost of the linears have gone up significantly because of the copper price and plastics and steel and so on, and they don't see -- think that the linear is an option.

On the other hand, many other customers have decided to slow down in their design, given that they have additional six months or one year, depending on the application. And so, to that extent, you know, they will get compliance that's closer to the deadline.

Now, in terms of how they will roll out their products, you know, talking to the OEMs, what we are hearing, is that they will do it gradually over a period of maybe two or three years. Obviously, they [know how] to take care of California immediately as the deadline comes close, and then they will propagate that to other states within the U.S. because they really can't afford to have a special product just for California. But, at the same time, they do have linear power supplies in the inventory that they have to clear, so it makes more sense for them to do it on a gradual basis. And then, of course, many other countries are considering making this standard mandatory, including Australia, New Zealand, China, Korea, and so on, and as those happen, it really wouldn't make sense to have a linear power supply at all. It would make sense to completely transition to electronic or switch to more power supply. And so our best estimate is that over the next few years, maybe three years or so, the transition will happen.

So would it be fair to say that maybe the first meaningful design win spike filling switch has now been pushed to Q4 '06 or Q1 '07?

No, we already have had a number of design wins with LinkSwitch. If you remember, our LinkSwitch revenue was very small beginning of last year. Now we are at 9% of total revenue, and in terms of unit volume, that's significantly higher because it's a lower [ASP] product. So LinkSwitch is already on its ramp, and what we should expect -- or what we think would happen is it will continue to ramp over the next three years.

Okay, and then a question on -- in the press release, I believe, it was stated that product prices have been stable, and normally the business experience is ongoing price erosion. So I just want to clarify that you mean actual nominal stablity or has it -- what exactly do you mean by stable prices?

It's relative stability, because the competing solution, whether it's linear or discrete, are impacted more by raw material costs increasing. For example, linear has a lot of copper and iron and plastic. Discretes have more copper, more labor, and more raw materials because there are more components. So the raw material cost increase that everybody has experienced has more impact on competitive solutions than it has on us. So because of that, we have not seen as much pressure to reduce our prices as we have seen in the last several years.

Okay.

I'm not saying the prices are not declining, but they're declining at a much lower rate.

Okay, and a couple of more questions, if I may. I believe the last time you reported numbers, that there was a $10 million receivable outstanding to a supplier who was engaged in some technology development in Asia. And I was wondering if that receivable is still outstanding?

You may be referring to a loan.

Okay, I'm sorry, a loan, yes.

Yes. It is to one of our foundries. They [indiscernible] that loan to help them develop a lower-cost process for us in the long run.

And that's still outstanding?

That's still outstanding. The terms are disclosed in the 10-K, the last 10-K, I think, 2004 10-K.

Yes, that loan matures in 2009.

Okay, I will go check that. And then the other question relates to Japan. You recently converted from a distribution relationship with Matsushita to doing direct sales, and I don't have -- I guess I'd like to know what were the sales by Matsushita in 2005, and have all those sales converted to direct sales in '06?

Well, it won't happen overnight. To first answer your question, in 2005 I would estimate is roughly about $40 million to $50 million based on the royalties we receive. So that was -- we estimate that's their sale. And they have rights to continue to sell the products they got license to before we terminated the license, which was the middle of last year. So any product they had before that, they can continue to sell. So their revenue is not going to come down immediately. But they don't get access to any of the new products we've introduced since the middle of last year. And so with the new products, which are more cost-effective, we expect to take share away from them and grow our own business in Japan. And last year we doubled our business in Japan in dollar volume.

And that was, too, what, about $5 million?

Yes, it's –- 2005 was roughly $5 million, and we expect that to continue to grow significantly this year and the next few years.

Can you tell me how much it grew in the first half of '06?

I don't have those numbers with me.

Ross Seymore, Deutsche Bank.

One way to get to the gross margin side of things is in the past you've given what the ASP is in the quarter. Can you give that to us for the second quarter and potentially back to the first quarter?

It was $0.43 in the second quarter, it was $0.45 in the first, and it was $0.43 back in the fourth quarter.

Let me caution you on that, because it is affected by both price and mix. As we sell more LinkSwitches, the ASP will come down, and it doesn't directly correlate with gross margin.

Well, that leads in nicely to my second question, which is as we look forward, I know you're not projecting on gross margin for all the specific reasons, but is there anything that has caused the gross margin to rise in the last six months that you see ending as we go into the next six months?

Well, as we have said many times in the past, gross margin is a combination of a lot of different effects. We will continue to reduce costs in manufacturing. Like, for example, we are making very good progress on the testing overseas. We have shipped more than 50% of our testing is now done in Southeast Asia, and by the end of the year, we should be around 70% testing in Asia. And, of course, that helps the cost. However, we are also subject to the yen exchange rate, which none of us can predict.

Thirdly, the raw material costs also impact us to a lesser extent, as I mentioned, compared to our competition, but they do impact us. You know, the copper -- if you use a small amount of copper and a small amount of plastic, you know, chips, and also the silicon -- the raw silicon costs, are going up slightly, they are expected to go up further in the future.

So it is very hard for me to predict whether the gross margin will continue to improve or stay stable and so on. It depends on many different factors.

Okay, fair enough. As we look to the turns requirement to hit your out quarter guidance, as I look back to the third quarters in the last couple of years, you've done at least 70% turns if not north of that. Is there any reason why you're predicting, say, 62%, 63% turns with today's guidance?

Yes, and the reason is if you look at our bookings, they are longer-term bookings than before. Our turns in the second quarter was 70%, but we had a very strong booking in the second quarter, and they were longer term than usual.

Okay, so why do you think that is? Because people getting ahead of the regulatory environment or is there some sort of lead-time issue on your side of things?

Well, what we understand -- the best we can understand is that the lead time on other components in the power supply have gone out, and so when they are components of a power supply all of them together, so even though our lead times are short, most of them are in the two- to four-week range, they end up ordering parts ahead of time because there are already a lot of other products for the power supply.

Okay, and then, I guess, the corollary to that is since the lead times are extending, you have better visibility into the fourth quarter than you typically would have at this time?

It's not that far out. Instead of ordering most of it for the current month and the following month, they are going maybe three months, but they're not really going into the next quarter. It's not that far out. We will still be expecting low 60s turns business to hit our mid number.

Okay, then, the final question -- excluding all the one-time aspects, has there been anything -- and excluding the DSO expense -- just on a pro forma OpEx level, has there been any major changes in the last couple of quarters versus what you would have expected along the lines of hiring more salespeople so you would have your SG&A going up, R&D rising, any sort of abnormal behavior in those income statement lines?

No, none whatsoever. There is really nothing out of the ordinary that you would normally expect.

[OPERATOR INSTRUCTIONS] Craig Ellis, Citigroup.

Thank you very much, Rafael, welcome aboard.

Thank you.

You're welcome. Balu, can you put some color on where you'd expect the third quarter strength to derive from -- from a product as well as an end-market standpoint?

We would expect strength in all areas, because third quarter is usually our big growth quarter, and third and fourth quarter are always stronger than the first and second. I say always, but some years have been strange, but in most years they've been stronger. I don't have any reason to believe or I have no information to believe any one area is better than other.

And were you talking about end market or products there, Balu?

I'm talking about end markets.

Okay, what about from a product standpoint?

Well, the product standpoint -- LinkSwitch, obviously, has a lot of momentum with the energy standards and linear replacement. So that will continue to grow. Tiny III is a new product. It is more cost-effective than the previous generation. Customers are very excited about some of the feature we have put in there, so I expect to see it grow very strongly in the second half of the year, however, it's still in the early stages. Tiny II will still contribute the largest revenue, and it's still growing even though it's about six years old now, it still continues to grow.

Okay, helpful. And then just as we look at the progress that you're making on the gross margin line, and if we were to exclude the options inquiry expenses as well as the legal expenses, would we be finding that the operating margin would be rising or is incremental sales expense offsetting or more than offsetting the margin expansion that you're getting?

We would expect the operating margin to increase.

Sumit Dhanda, Banc of America Securities.

A couple of quick follow-ups -- first, any thoughts on another share repurchase program? And then, second, any more on design wins with large handset OEMs? I know LG was a design win for you last quarter. So have you made any inroads with any other large handset OEMs?

Okay, so -- what was the first question --

On the repurchase question -- we completed the $25 million repurchase program, and until we become current with our filings, we will not be initiating a second one, mainly because we're not allowed to.

We cannot start a repurchase program until we are current, and as far as whether we'll do it or not, we look at it periodically, the board looks at it periodically and makes a decision on that. But at this point there is nothing we can do.

And so to answer the second question, there is no -- if there is a major cell phone design win, I am sure we will let you know. We (multiple speakers) related to the CEC linear replacement, but at this point we have nothing significant to report.

Let me ask this another way. You've been engaged with a lot of these OEMs in terms of driving design wins post the CEC mandate, so have there been any other decisions reached by these larger OEMs where you fail to garner a design win or is there still design activity but no indication of a design win with one of these larger OEMs?

Well, as I said, some people are going forward, and some people have slowed down because they have more time. So if you're talking specifically about cell phones, there are a number of design activities going on where our customers have designed our product in, but they are waiting to hear from their customer as to whether they won the design. And those kinds of decisions in many of these cases are being delayed because of the CEC delay.

Ladies and gentlemen, we have reached the end of the allotted time for questions and answers. Mr. Shiffler, are there any closing remarks?

Thanks, just to say that a replay of this call will be available shortly on our investor relations website, www.powerint.com, that's the investor info section. A telephone replay is available for one week by dialing 800-642-1687, and the passcode is 3902365. Thanks, everyone, for listening and good afternoon.

Thank you, this concludes the Power Integration's conference call. You may now disconnect.