Cohu Inc (COHU) 2017 Q2 法說會逐字稿

Greetings, and welcome to the Cohu 2017 second quarter earnings call. (Operator Instructions) As a reminder, this conference is being recorded. I would now like to turn the conference over to your host, Jeff Jones, Vice President of Finance and Chief Financial Officer. Please go ahead, Mr. Jones.

Good afternoon, and welcome to our discussion of Cohu's most recent financial results. I'm joined today by our President and CEO, Luis Muller. Following our opening remarks, we'll provide details of our performance for the second quarter of 2017 as well as our outlook for the third quarter of this year. If you need a copy of our earnings release, you may obtain one from our website, cohu.com, or by contacting Cohu Investor Relations. Before we begin, you should all be aware that during the course of this conference call, we will make forward-looking statements reflecting management's current expectations concerning the company's future business. These statements are based on current information that we have assessed, but which, by its nature, is subject to rapid and even abrupt changes. Forward-looking statements include our comments regarding market momentum, growth in our contactor business, market expansion into inspection market, new products and customers' future results including Q3 guidance and any comments we make about the company's future in response to your questions. We encourage you to review the forward-looking statements section of the earnings release as well as Cohu's filings with the Securities and Exchange Commission, including the most recently filed Form 10-K and Form 10-Q. Our comments speak only as of today, July 27, 2017, and Cohu assumes no obligation to update these statements as a result of developments occurring after this call. Further, our comments and responses to any questions will not make reference to any specific customers, as we are precluded from disclosing such information by our nondisclosure agreements. Finally, during the call today, we will also discuss certain non-GAAP financial measures. Please refer to our earnings release for a reconciliation to the most comparable GAAP measures.

Now I'll turn it over to Luis.

Thanks, Jeff, and good afternoon, everyone. For the second quarter, we reported strong results with sales of $93.9 million, up 16% sequentially and 23% year-over-year. And as reported on July 11, it was above our original guidance of approximately $86 million. This was our highest quarterly sales since 2014, driven by continued momentum in automotive, mobility and IoT markets. Non-GAAP earnings per share was $0.48 compared to $0.23 in the year-ago quarter. Orders were a new record and for the first time exceeded $100 million, with increased momentum for industrial and solid-state lighting applications. Systems accounted for 57% and recurring 43% of total orders. Additionally, our contactor business increased to over 10% of quarterly sales due to strong customer demand for a new RF solution in digital contactors coupled with Kita spring probes and is a growing opportunity for Cohu.

Now looking more specifically at the system orders by segment in the second quarter. Digital and mixed signal represented 28% of system orders, mainly for automotive, industrial and consumer markets, with strong demand for our Tri-Temperature Pick-and-Place Handlers for testing microcontroller devices. We continue to gain traction with our products in part due to their wide temperature range capability and expect to have 2 new customers evaluating our handlers before year-end. Processors were 24% of system orders in the quarter.

There was a strong ramp for the Eclipse Active Thermal Control Handler at test subcontractors in Asia, supporting a production launch of new mobile processors from a leading customer that outsources assembly and test. Additionally, we had a repeat order for thermal subsystems for test from another customer's device that will be in volume production in the second half of this year. Our leading microprocessor customer placed orders for capacity expansion at factories in the U.S. and abroad. We're also making good progress at new and existing customers with a high power active thermal control configuration of the Eclipse handler for testing server chips as well as graphic and other high-end processors used in gaming and artificial intelligence applications.

LED and sensors were 21% of system orders. There was high demand for turret handlers from a major LED customer for test and inspection of high-power devices. Sensor system orders were also strong reflecting the market penetration of our products for testing MEMS used in automotive applications. We expect continued momentum in both of these segments in the third quarter.

Analog RF and power discrete were 16% of system orders, mainly for turret handlers, serving a wide range of customers testing filters, amplifiers, power transistors, diodes, precision and power resistors. There were also gravity handler sales where temperature test is required. Power management was 10% of system orders driven by demand for gravity and strip handlers as well as some turret systems.

We expect the Power IC segment to continue growing in the coming years, as increasing global GDP drives more factory equipment in productivity.

Additionally, increasing automation and robotics will be a catalyst for power ICs and other semiconductors. Other segments accounted for the remaining 1% of system orders. In support of our market expansion efforts, in mid-July we announced 2 exciting new products; first, we introduced the PANTHER prober for test and inspection of singulated wafer-level chip-scale packages and bumped dies. Growing demand for semiconductor integration with system-in-package and increasing quality requirements are driving the need for final test of integrated circuits after wafer singulation. These devices are essential in mobile and IoT consumer products where space constraints and functional requirements are paramount and are becoming a growing need in automotive applications. This has been a much-anticipated product launch with initial production units shipping from Malaysia late in the quarter. This prober leverages a unique combination of our established testing technology, while also further increasing our addressable market by expanding into a key vertical.

We also announced the new vision inspection margin for enhanced micro- crack detection on wafer-level chip-scale packages called Aquilae. Advances in functionality and product reliability in the IoT, mobility and automotive markets are driving the need for early detection of defects at a micrometer scale. In order to meet the growing expectation of customers, OEMs and product manufacturers are establishing more stringent requirements and setting new standards for quality across the supply chain. Cohu's new vision and inspection solution was designed to identify micro cracks on an integrated circuit that oftentimes go undetected using only electrical tests, thus minimizing in-product failures. This module broadens our vision inspection capability to include micrometer scale defect detection on our turret handlers as well as on PANTHER.

Longer-term, we see opportunities to broaden applications for Aquilae to larger semiconductors, such as those used in automotive and digital mobile markets. Also during the quarter, we continued to make progress across our new product development pipeline, with plans to ship our first system-level test platform in the third quarter. This quarter we'll also ship a new handler to a recently acquired customer that is a leading Korean memory and mobile processer semiconductor manufacturer. This new product addresses requirements for back-end factory automation and significantly reduces use of operators. This would be the first handler developed for what is commonly known as a "lights-out" factory. And it brings Cohu once again to the forefront of capital productivity in the semiconductor test industry.

Now turning to our contactor business, as I mentioned earlier, revenue from this business increased to 10% of quarterly sales. Orders were up 18% sequentially, due to strong demand for contactors including Kita spring probes. We had several design wins in the quarter across the 3 market segments. First, the digital and mixed signal market had strong order activity mainly driven by capacity additions associated with the MATRiX handler. In the second quarter, we captured a key design win at a leading European semiconductor manufacturer. And also secured a strip test application with Kita probes at a customer with a large installed base of Cohu handlers in Taiwan. We expect follow-on orders for strip contactors as we displace a competitor's product in volume production.

In the power, sensors and LED markets, we secured another win with the same European semiconductor manufacturer, which has the potential to grow to over $1 million in annual sales. The multibeam structure of our new Cantilever products is a real competitive differentiator, especially for small form factor semiconductors that need accurate temperature control during test. We have received follow-on orders for multibeam contactors in our turret handlers and have evaluations planned in the third quarter at customers in China and the U.S.

In the RF Analog market, we made significant progress at a U.S. industrial customer, following a recent design win. The multibeam structure of our cDragon contactor provides superior signal integrity in combination with best-in-class thermal performance. Feedback from this customer has been excellent. And we're rapidly scaling our design and manufacturing capability to satisfy multiple designs to replace their installed base of a competitor's contactor with Cohu's solution.

We have 3 other customers aligned for evaluation starting in the third quarter. The cDragon offers substantial performance advantages over competitors' products and it's quickly establishing the Cohu brand as the leading solution provider for complex RF test applications.

To close out my remarks, I would like to reiterate a few of the key messages we conveyed at the recent corporate access day. First, as a result of our consistent execution over the past 2 years and planned expansion of our addressable market to $2 billion, we increased our midterm model--which we define as the next 3 to 5-year period to include an annual sales target of $500 million, with 45% gross margin and 20% EBITDA, both non-GAAP. This increased confidence and excitement about the future of Cohu is driven by our successful implementation of strategic actions, including streamlining operations and cost structure with over 90% of system shipments originating from our Malaysia facility. Increasing share in our core handler markets, expansion in test contactors as well as executing strategic acquisitions and new product development. Collectively, these have improved our operating model and margins and we believe position Cohu to deliver continued profitable growth and cash flow.

With that, I'll now turn it over to Jeff, for details on the Q2 financials and guidance for the third quarter.

Thanks, Luis. The results for the quarter highlight the strength of the financial model generating non-GAAP profitability of 16% operating income and nearly 18% EBITDA in line with the updated operating model targets as discussed last quarter on sales of approximately $94 million. Record orders in Q2, drove a 12% increase in backlog, which sets the stage for solid Q3 sales. The Integration of Kita is on track and we remain very optimistic about contactor sale synergies and our growth opportunities in this market. For Q2, the GAAP to non-GAAP adjustments include approximately $1.8 million of stock-based compensation expense, $1 million of purchased intangible amortization expense, $341,000 of restructuring costs, $465,000 of costs related to the step up in valuation of inventory acquired from Kita and $56,000 of other acquisition related costs.

My comments that follow, including the Q3 guidance, are all based on Cohu's non-GAAP results, which exclude the impact of these items. Sales for the quarter were $93.9 million, which was above our original guidance of approximately $86 million and our preliminary expectations of approximately $93 million, driven by continued momentum in the automotive and mobility markets. One customer in the automotive market represented approximately 21% of sales during the quarter. We had no other 10% or greater customers in the second quarter. Q2 gross margin was 40.8% and in line with our financial model. Operating expense was $23.1 million and in line with our preannouncement. The non-GAAP effective tax rate for Q2 was 10.2% and lower than forecasted. Our tax rate benefits from profit generated outside the U.S. in countries with lower statutory income rates and in certain countries where Cohu has income tax holidays. The tax rate also benefits from profits generated in the U.S. which do not currently incur a tax provision because of our valuation allowance. In Q2, we generated more profit in the U.S. than initially projected, resulting in a lower overall tax rate. As a result of these factors, we are expecting our effective tax rate to be approximately 17% for Q3 and Q4, both of which are below our prior expectations of 20%.

The cash balance grew by $8.1 million during Q2 and cash generated from operations during the second quarter was $4.4 million. Accounts receivable increased sequentially. Our DSO increased by 2 days to 86. Inventory also increased sequentially, mainly due to the timing of Q3 shipments as well as advanced purchases of inventory to ensure continued production and delivery of products as we transition the ERP system in Malaysia and Switzerland to align with our other principal business locations.

Inventory days increased by 3 to 102. Accounts payable days was flat quarter-over-quarter at 68. And the overall cash conversion cycle increased by 5 days to 119. Fixed asset additions in Q2 were approximately $1 million and depreciation for the second quarter was $1.2 million. Deferred profit at the end of June was $6.3 million, up $1.6 million quarter-over-quarter. And the related deferred revenue at the end of Q2 was $8.2 million, that's up $2.5 million sequentially. Cohu's Board of Directors approved a quarterly cash dividend of $0.06 per share, payable on October 20, 2017, to shareholders of record on August 25, 2017.

And now moving to our guidance for Q3. With strong backlog at the end of the second quarter and Q3 shipments expected to be approximately the same as Q2, we're guiding revenue to a range between $88 million and $95 million. We're providing a range for this quarter due mainly to potential timing variables related to the ERP transition as well as product acceptance from new customers. And as announced in our preliminary results on July 11, we are projecting revenue for the second half of the year to be approximately the same as the first 6 months of 2017.

Gross margin in Q3 is expected to be in line with our financial model at approximately 39% to 41%, depending on actual revenue results. Operating expenses for the third quarter are expected to be approximately $22.5 million, up approximately $500,000 from our recent run rate due to product development costs for a new pick-and-place handler.

Our projection for Q3 non-GAAP gross margin and OpEx, excludes approximately $400,000 of Kita inventory purchase price step-up costs, which will flow through cost of sales. Additionally, any changes to our initial estimates of the fair value of Kita assets, including intangible assets acquired, may result in different amortization amounts being recorded in the third quarter or future quarters.

That concludes our prepared remarks. And now we'll take questions.

(Operator Instructions) Our first question comes from the line of Patrick Ho with Stifel.

Hi, this is Brian Chin on for Patrick, thanks for letting me ask a few questions. First, I just wanted to confirm, I don't think I caught it in the prepared remarks, but whether you are still anticipating the $10 million to $15 million plus in sales from the new wafer level prober and SLT platforms in the second half?

Hi, this is Luis. As you know, we just introduced the PANTHER program mid-July. I think you were there. You saw the product, and we are in the process of ramping production from our Malaysia operation for this product. As you've seen, it's quite innovative. The product includes testing, marking, inspection, taping for singulated wafer-level chip-scale CSP's and bump dies. But at this point I'm not going to give any more specifics about PANTHER, particularly, anything commercially related because I think it's sensitive information and things that competitors will be listening for.

Okay. Fair enough. Also I wanted to ask you, you gave some good qualitative commentary on the contactor business across the 10% threshold like you mentioned. I think you've talked about maybe a 20% CAGR over the midterm period for that particular segment. Is it more realistic, given the momentum you're seeing, maybe to think about that near term, off the lower base, as kind of a minimum growth threshold?

Well, at this point we're sticking to the plan that we have, that we have communicated, which is a 10 -- 20% CAGR over the midterm and we're actually good off the gate here, with the beginning of the year and the first half results and the quarter-over-quarter order momentum. But we're staying with our 20% number at the moment.

Okay. Fair enough. One last quick question for me. I know you don't break it out this way precisely but, given the breakdowns of the order momentum that you saw in the quarter, it seems like maybe industrial and auto in aggregate, which is strong markets for you, might have been down a little bit sequentially Q-on-Q. Can you maybe provide any color in terms of the market commentary around that?

You know, I don't have data handy right now to comment on auto and industrial Q-on-Q but I can tell you this for sure, it was a -- it continued to be a very strong quarter for automotive orders in Q2. It was also a very strong quarter for the mobile segment and IoT. And we had also a pickup in strength in the industrial and LED. So it was actually a very broad-based quarter and that's what led to record orders, like I said, for the first time above $100 million.

The next question from the line of Edwin Mo with Needham & Company.

It's actually Arthur on for Edwin. Just the first question is just on the system level test platform. Congratulations on the progress you've made and targeting to get it out -- launch it this quarter. Can you provide some color on the stage of customer engagements and sort of the applications you're targeting in order to grow this business?

Sure. We haven't quite announced the product yet, publicly speaking, even though we are talking about having a first revenue -- or sorry, a first shipment in Q3. We're looking at this product as an opportunity to get into the space where semiconductor manufacturers are integrating different devices together like processor, memory, potentially RF, PMICs and others into a system-in-package. So the applicability for the system-level test in the near term is more into, if you will, into the mobile space, high-end -- high-end devices in the mobile space. And as we mentioned here in mid-July, there may be a longer-term opportunity for these platforms into the memory market as well.

Got it, thanks for that color. And then given that contactors and handlers pretty much go hand-in-hand, can you briefly talk about some of the sales synergies you've seen with your broad contactor portfolio and whether they have exceeded your prior expectations or remained consistent?

For the first 2 quarters of this year, I would say, it really exceeded our expectations in the sense that, it's going faster than originally anticipated. We had strong bookings in contactors, increasing quarter-on-quarter. A lot of it has to do with the benefits of having a Kita spring probe and the opportunity to leverage that into both into our handler installed base and frankly also, into sales of handlers to new customers. So it's been very positive actually.

Okay, great to hear that. Just last question, with the launch of this -- the Aquilae inspection module, what's the go-to strategy here -- go-to-market strategy here and how do you plan to grow that business?

Well, we have -- just to be clear, we have been in the package inspection for quite some time now through our Ismeca acquisition. Most of our turret handler sales have included package inspection along with electrical test functions, right? It's only more recently that we have seen an expansion or the capability of these products adding die inspection, micro-crack and sort of true physical measurement capabilities, including the Aquilae that we just announced. So what we see today is more of an expansion of vision capability on the turrets, which was aimed for small packages, small semiconductor devices and, as we grow into this, we're also having an increased focus on expanding vision technology, as well as the automation technology, which may or may not be turret in the future, into larger packages like those that are used in the automotive and processer markets.

The next question comes from the line of Dave Duley with Steelhead.

Most of them have already been asked but I'll take shot of couple of others here. Just as far as your guidance goes in the September quarter, could you just talk about how some of the major pieces are going to act? I guess, it's relatively flat or slightly down, whatever the guidance is. But just from a macro perspective, could you just talk about what pieces are going to be up and down in the September quarter?

As you said, overall it will be relatively flat, shipments are expected to be relatively flat. I mean looking at the Q3 orders, I would say, we're going to see strength sort of relative to what we're seeing in Q2. So I mean, automotive, industrial are still going to be strong. Contactors, although lower numbers, they'll have good momentum. I don't know that we're going to see any major shifts in the sort of the makeup of the revenue or the shipments. So I would say it's going to be pretty much strength in auto and industrial followed by growth in contactors and then similar activity in solid state lighting, computing and so forth.

Okay. And then just as a clarification, you mentioned how you gave a revenue range this quarter and one of explanations was customer acceptance and the other was ERP implementation. Could you just help me understand, what you meant by that?

Yes. Sure, sure. So those are the 2, we call them, timing variables that are identified for Q3. It could impact the revenue and -- so the first one, I'll talk about relates to the ERP changeover, Malaysia and our Swiss locations, and so with any significant systems change there is always a risk of transaction processing delays as the new system comes online. And we've done thorough testing of the system. It's also a system that we know very well. It's used in nearly all of our other Cohu locations. So we believe the timing risk is low but still exists. Secondly, we have $2 million to $3 million of revenue which requires customer acceptance, new customers. So requiring acceptance before we can recognize the revenue. We're scheduled to achieve that acceptance near the end of the quarter. So any delay, any slight delay in this process, which, by the way, is a process that we don't control 100%, right? The customer has a very large say in the timing of acceptance. So any delay in this process will move that revenue into Q4.

Okay. As a final question, what is -- can you hear me still?

Oh yes, we can hear you.

I'm sorry, it's seemed like the phone faded out there. Just as a final question, you mentioned the wafer-level chip-scale packaging, just clarify what your statement was that you did make about that? I think you said you were shipping a system from Malaysia into production and so -- I'm not looking for any new information but could you just update us -- repeat what you said about that. And then remind us, you have evaluation systems in the field at customers with this product already, so now you're shipping a production system, is that the point?

Yes, that's about right. We have multiple systems in the field and what we're doing is we're ramping production of this product, the PANTHER prober, from our Malaysia manufacturing operation and should be ready to start shipping the product towards the end of third quarter from Malaysia.

Okay. So I guess, you would expect to be putting systems into production towards the end of the third quarter?

Yes. To be specific my comment was, we'll be shipping out of our production in Malaysia at the end of third quarter. But yes, going into customers.

Our next question is from the line of Craig Ellis with B. Riley.

This is Peter Peng calling in for Craig. Congratulations on the strong execution. Just on a high level view, I recall, a few quarters ago, you provided some industry growth in the low single-digits for contactor and handler. So has that view changed? Or is this more of a share gain story for Cohu?

Well, it's a combination. The view has not changed and we still believe in those growth numbers for the market. In addition to -- we obviously have grown the business both from a share again perspective in core handlers as well as share again in contactors in, obviously, an acquisition that we closed beginning in this year of Kita manufacturing in Japan. So you put all those pieces together and that's what Cohu is in 2017.

Okay. And then just on the cash balance, with a growing cash balance and with kind of the successful integration of Kita, how do you think about using cash for future, either through dividends, M&A and so forth?

Yes, good question Peter. So, as you know, you mentioned we pay a dividend that is roughly $6.5 million a year. The operation requires $5 million to $7 million per year in CapEx. And then we estimate that we need about $35 million in reserves just to operate the business and provide customers with the confidence that we have the liquidity in order to continue to service them during ramps and so forth. So all in all, we view it as requiring about $40 million to $42 million for the operation, the balance of which would be excess that is earmarked currently for investments to grow the business. And we outlined that strategy in great detail during our corporate access day, and it's a combination of organic development, new products in the contactor space, wrapping up the wafer level CSP product, new products coming into the core markets. But then there's also opportunities -- future opportunities for acquisitions as we've done in the past with Rasco, Ismeca and Kita. We will continue to look at opportunities to acquire companies to accelerate the growth to hit the financial targets that we laid out also during that access day, which the midterm target is now $500 million in sales and a 45% gross margin. So that's what that cash is earmarked for. It's earmarked for strategic investments to achieve those targets.

Thanks for the helpful color. And a final question is how should we think about margins as you'll launch these kinds of products -- new products and as you start recognizing revenue from these products?

Well, you should think about it in terms of the model that we've set forth. In the near term, we're still operating under this Cohu 400 model, which is defined as $400 million in annual sales, a 42% gross margin and an 18% EBITDA. Now we've made fantastic progress with respect to achieving that level of profitability even on sales that are slightly less than $100 million per quarter. So as we release new products, they start to gain traction, we grow our revenue. We will be achieving the 42% gross margin at a $100 million in sales. And then, over time, over the midterm, the 3 to 5-year mark, it's continued growth in core markets, it's further expansion in inspection and potentially additions with acquisitions that could accelerate that time frame. But it's the expansion of revenue based on those products which will have gross margins at or above the target model financial profile that will bring up the revenue as -- bring up the gross margin as the revenue increases.

At this time, I will turn the floor back to Jeff Jones for closing remarks.

Okay, before closing up today's call, I want to let you know about upcoming IR events which include the D.A. Davidson investor forum in New York City on August 15, and the Dougherty conference in Minneapolis on September 19. We'll also be hosting roadshows to Baltimore, Boston, Midwest and Dallas. If you would like to request a meeting in any of these cities, please contact either myself or our IR firm Shelton Group. And we look forward to providing an update when we report our third quarter results. Thank you, and have a good day.

Thank you. This concludes today's conference. You may disconnect your lines at this time. Thank you for your participation.