Good morning. My name is Tia, and I will be the conference operator today. At this time, I would like to welcome everyone to the Teradyne Q2 2016 earnings conference call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question and answer session. If you would like to ask a question at that time, simply press star and the number one on your telephone keypad. If you would like to withdraw the question, press the pound key. Thank you. At this time, I would like to turn the conference over to Mr. Andrew Blanchard. Sir, please begin.
Thank you, Tia. Good morning, everyone. Welcome to our discussion of Teradyne's most recent financial results. I'm joined this morning by our CEO, Mark Jagiela, and our Chief Financial Officer, Greg Beecher. Following our opening remarks, we'll provide details of our performance for the second quarter of 2016 and our outlook for the third quarter of this year. Press release containing our second quarter results was issued last evening. We're providing slides on the investor page of the website that may be helpful to you in following the discussion. Those slides can be downloaded now, or you can follow along live. Replays of this call will be available via the same page after the call is over. The matters that we discuss today will include forward-looking statements that involve risk factors that could cause Teradyne's results to differ materially from management's current expectations.
We encourage you to review the safe harbor statement contained in the earnings release, as well as our most recent SEC filings. Additionally, those forward-looking statements are made as of today. We take no obligation to update them as a result of developments occurring after this call. During today's call, we'll make reference to non-GAAP financial measures. We posted additional information concerning these non-GAAP financial measures, including reconciliation to the most directly comparable GAAP financial measure, where available on the investor page of our website. Between now and our next earnings call, Teradyne will be participating in investor conferences hosted by Needham & Company, Citi, and Deutsche Bank. Let's get on with the rest of the agenda. First, Mark will comment on our recent results and the market conditions as we enter the third quarter.
Greg will then offer more details on our financial results, along with our guidance for the third quarter. We'll then answer your questions. This call is scheduled for one hour. Mark?
Thanks, Andy. Good morning, everyone. Today, I'll cover three main topics. The highlights of our second quarter, as well as our second half outlook, our reassessment of the near-term LitePoint wireless test market, and a closer look at the developments at Universal Robots. Greg will provide additional details, including an update on capital returns and the path to our $2 per share EPS target. Building on the momentum from Q1, continued strength in Semiconductor Test led to another strong quarter in sales and earnings. Teradyne's second quarter revenues of $532 million were at the highest level in 4 years, and when combined with Q1, our revenue total of $963 million in the first half was at the highest level in 15 years. Complexity growth in mobile devices continues to drive test time increases that have more than offset slowing unit growth.
We are also seeing an increasing number of test seconds devoted not just to weeding out defects, but directed at trimming and optimizing the device to improve performance. In these cases, the tester permanently programs into the device tweaks that enhance accuracy, lower power consumption, or tune the part to better meet specifications. This is becoming an economical way for manufacturers to get higher yields from increasingly finicky lithography nodes. In 2016, despite the weak analog and microcontroller test demand, we see the SoC Test market in the $2.2 billion-$2.4 billion range, or up about 10%, $200 million from 2015 at the midpoint. We expect to capture about half of that market growth. Much of that has already been realized in the first half of the year.
If you recall, the SoC tooling ramp for 2016 shifted about one month earlier than normal, resulting in a revenue profile that is more first-half concentrated. As a result, first-half Semiconductor Test shipments are running about $100 million ahead of the first half of last year. In memory test, strength in flash test demand has been more than offset by weak DRAM capacity needs. For the year, we expect the memory test market to be in the $400 million range, down about 20% from 2015, while our share is projected to be up to about 30%. The bright spot in memory test continues to be in the area of higher speed flash interfaces, such as Universal Flash Storage or UFS.
Higher bandwidth requirements for both SSD and mobile applications have resulted in the growth of new interface standards like UFS, which in turn have driven the need for new testers to handle both the bit growth and replace obsolete testers in the field. UFS and similar technologies are at the very early stage of adoption. Production will continue to ramp over the next few years as data speed increases complement density increases. Our Magnum tester has successfully captured early production at four of the five manufacturers of this new class of leading memory technology, leading to our largest quarterly orders for the Magnum product line. This same Magnum architecture has now proven leadership in all variants of flash testing, as well as DRAM wafer probe. Our system test business continues to perform well.
Although group sales will likely be down about $20 million on lumpy storage test business, group profits should be up for the year. Universal Robots had a very strong second quarter. We are on track to meet or exceed our 50% growth target for the year, delivering $90 million-$100 million of revenue for the full year. In fact, through the first six months of 2016, revenue is running over 80% ahead of the same period last year. Second quarter revenue of just over $25 million was more than double the same period a year ago. We continue to invest in distribution, channel partnerships, and R&D to maintain a high rate of growth. Our unique combination of deployment ease, safety, and low cost continues to resonate across a broad variety of industries and applications.
Growth is dependent on increase in velocity through our distribution channels and the growth of system integrators capable of delivering solutions to our targeted small to medium-sized enterprise customers. We plan to stay ahead of the curve in development of these channels to maintain our market lead. I will come back to one aspect of facilitating this growth at the conclusion of my remarks. At LitePoint, we have been confronted by a further deterioration of the near-term outlook for the wireless production test market. A combination of slowing handset growth and a slowdown of new technology adoption are two universal factors causing this decline. This combination of factors lowers our anticipated market size from $400 million-$200 million for 2016 and 2017. Consequently, LitePoint revenue will be likely down proportionally, resulting in the goodwill and intangible asset impairment charge this quarter.
Having enjoyed the surge in revenue and profits in 2012, shortly after our acquisition, we have since seen slowing demand. Although we have gained share all along the way, the market shrink from about $1.2 billion in 2012 to a projected size of about $200 million in 2016 is obviously too large to offset. Up until 2016, profitability, both on the ride up and down, has been above model. We have had good execution on product development, share gains, and profitability, but we misread the rate of both unit growth and technology change. We will be restructuring the group for continued profitability going forward in the smaller near-term market and continue to focus on leadership for the next technology inflection. The next technology inflections in wireless will be millimeter wave, Wi-Fi, and 5G cellular. Both will require significant retooling of existing equipment and should produce another surge in the test market.
This cycle will likely start in the 2018 period and run with intensity for three to four years. In the meantime, smaller technology changes due to 802.11ax, IoT, and the increased use of MIMO technologies will provide some stimulus, but the market will likely be in the $200 million-$300 million range up until 2018, when 802.11ad millimeter wave Wi-Fi should begin to ramp. Returning to Universal Robots. One of the key features of our cobot product that has propelled the market expansion is the ease with which an automation solution can be implemented, much easier than a traditional industrial robot. The first component of ease translates to less time to program, or more accurately, train the cobot, as complex programming languages have been replaced with tablet-based, icon-driven instruction, as well as hands-on direction of the cobot's arm.
People with no programming experience have often been able to train the robot in this fashion in less than an hour. The second component of ease is the inherent safety of the cobot that removes the need for a costly and restrictive enclosure. These ease-of-deployment features, combined with low cost, results in a fast payback time and opens up automation not only to large enterprises, but also small to medium-sized enterprises that typically don't have in-house automation expertise. In June of this year, we introduced a new platform approach called Universal Robots+, which extends our ease-of-use advantages to third-party partners. These partners supply add-on products to our Universal cobot that tailors it to the specific customer application. This includes peripheral products like grippers, actuators, optical guidance, measurement systems, and other enhancements.
Through Universal Robots+, this third-party ecosystem can take advantage of software API links that bring their products directly into the same easy-to-use environment that our cobot enjoys. This further reduces the user's time to solution and time to a positive ROI while enhancing our competitive advantage. In summary, our SemiTest, System Test, and Universal Robots businesses are performing very well. Complexity increases in semi devices are driving growth in our SemiTest market. System Test is a steady profit contributor. Robust growth in excess of 50% continues at UR. While the dramatic fall of the wireless test market is disappointing, we are restructuring our LitePoint business to focus on the next technology waves in connectivity and cellular. I'll turn it over to Greg.
Thanks, Mark, and good morning, everyone. I'll start with some brief comments on the year, I'll update you on the path to our midterm $2 EPS plan. I'll cover the second quarter results and third quarter outlook. First, though, let me explain the $338 million non-cash goodwill and intangible asset impairment charge recorded in our Wireless Test segment, which consists of LitePoint, acquired in 2011, along with our much smaller ZTEC acquisition in 2013. This year, the wireless test market is expected to be down about 50% from last year. There are multiple wireless test market headwinds, including a lull in the adoption of new wireless standards, slower smartphone growth rates, and most significantly for us, our large customer, who has oscillated between 51% and 73% of our annual wireless test sales, is expected to buy substantially less test equipment as a result of numerous operational efficiencies.
At this point, it's probably safe to assume a contracted wireless test market until 2018, when two new Wi-Fi standards should make their way into volume production. This sharp wireless market size reduction, along with our wireless headcount action in the second quarter, triggered the impairment review under GAAP, which resulted in the revaluation or write-down of our wireless segment goodwill and intangible assets. I should add that we've had other businesses that have faced tough market conditions, such as storage tests in 2013 and semi tests back in 2009, which were both fully renovated back to health and growth. We plan to do the same for our wireless business, which has had a track record of solid execution.
Notwithstanding the wireless test segment, which will deliver sales about $90 million-$100 million lower than 2015, we're on track to grow both sales and non-GAAP EPS for the total company this year. At the midway point, first half total company sales of $963 million are up 13% from the first six months of 2015. On the bottom line, non-GAAP EPS totals $0.86, up 23% from the first half of 2015. Universal Robots and SemiTest are driving the top-line growth. Starting off with Universal Robots. First half sales were $42 million versus $4 million last year in the first half, as we acquired UR in June of last year. On the more important standalone comparison, first half sales are up 82% from the prior first year, first half sales of $23 million.
We expect UR to be over $90 million in sales this year, with an operating profit rate of about 10%, as we've upped the investment level for distribution to extend our leadership position. Next year, we'd expect UR to be at approximately 15% operating profit on sequential sales growth of 50% or greater. Our total company model of needing about $390 million in quarterly sales to hit a 15% operating profit remains intact this year, despite the added UR OPEX investments. We're also leveraging our supply line resources to lower our cobot material costs. These material cost savings should be a hedge against inevitable competition, so that the cobot gross margins remain in the low 50% range. To date, we haven't seen any meaningful competition as the market for cobot automation is so broad and growing so fast.
UR cobots are performing a very long list of dull, dirty, and dangerous jobs around the clock. These include machine tending, assembly, pick and place, polishing, gluing, medical processing, food handling, inspection, packaging, welding, materials testing, painting. Well, you get the idea. Our first to market product lead is now being amplified by a growing ecosystem lead as well. We are growing our distribution pipeline in both quality and vertical coverage across regions, so that as competition increases in the future, our distribution partners will be more experienced, more capable, and armed with more proven and lower-risk solutions for customers than our competitors' partners. Finally, in terms of Industry 4.0, UR will continue to put control back in the hands of the operator on the floor, so that flexible manufacturing can be achieved at lower cost, whether for small and medium enterprises or large companies. Shifting now to SemiTest.
We're catching the expected mobility buying wave, with increasing complexity driving test intensity up. Despite projected semiconductor unit growth of around 3%, the test market is growing in 2016. Each year, apps processors have more transistors, RF chips have more bands, and devices of all types continue to grow in complexity. This complexity drives higher test times, which in the past was masked by improvements in tester productivity. As we've noted before, we see the impact of these productivity improvements for complex mobility devices on the tester market diminishing. In addition, as Mark noted, testers are increasingly used as dynamically tuned devices for optimum performance. These trends are a recipe for solid test demand despite slower semiconductor unit growth. We see these positive trends moving the long-term buy rate trend line in the SoC tester market.
In memory test, the trend to higher speeds plays into our Magnum sweet spot with our high-frequency, low-cost architecture. We have over 50% of the flash final test market and are well-positioned to benefit from the anticipated NAND growth from the fab investments being made this year. Regarding the market environment, as we have noted before, share is very sticky, with about half of our past gains coming from being aligned to growing segments versus head-to-head competitive battles. Therefore, we will continue to be selective in the segments we target in SemiTest. Shifting to System Test group. Overall, we operate at model profit or better in the first half. While storage test will be down from last year on a full year basis, we expect it to deliver model profitability for the year.
Defense and Aero is expected to resume growth this year as the multi-year sequestrations have ended, and production board test is gaining traction with our dual head high throughput solutions, particularly in the growing automotive electronic segment. I'll now comment on the $2 non-GAAP EPS midterm plan. If you recall, our 2015 non-GAAP EPS was $1.27. Starting from that point, we'll need to grow earnings at about 10% annually to achieve $2 in 2020 or sooner. This EPS growth should come from three main contributors: SemiTest, industrial automation, and capital return. In SemiTest, while we have historically done well gaining share very selectively, market contraction has offset some of these gains, yielding very little growth. Going forward, we expect the trends noted earlier to result in a 0-2% trend line market growth over the midterm, assuming a 3-5% unit growth.
We'll still see some even year, odd year swings in market size, but these swings are moderating with complexity advancing more steadily. This, combined with annual continued share gains of about a point, should contribute about $0.35 of the EPS growth in our $2 EPS plan. Much of our SemiTest growth will be offshore and taxed at our low Singapore tax rate. At Universal Robots, we expect the cobot market and UR continue to grow at 50%+ annually over the midterm, similar to what we have seen over the last several years. Rising labor costs, high turnover, and a shortage of workers in places like China, combined with a fast ROI, should drive ongoing UR growth and contribute about $0.30 towards EPS growth. This growth will be taxed at about 22% as the IP is in Denmark.
There is also significant upside to this estimate given the third-party report that pegged the cobot market size at $3 billion+ by 2020. If I conservatively assume a market half this size, $1.5 billion, and assume our share drops from 60%-40%, that's about $600 million in sales for Teradyne, far above what we've included in the $2 EPS plan. The remaining $0.08 should come from our continued return of capital through share buybacks funded by a strong annual free cash flow and some incremental growth in our other test businesses. Shifting back to the company level, we paid $12 million in dividends and used $29 million to buy back 1.5 million shares at an average price of $19.78 in the second quarter. This leaves us with $143 million remaining under our $500 million stock repurchase authorization.
Our cash and marketable securities total $1.1 billion, up $131 million from the end of the first quarter. We have $438 million in the U.S., and the balance is offshore. I should quickly point out that an increasing portion of our annual cash generation will be offshore. This year, it's expected to be about 80%. We'll continue to report this increasing foreign mix as it factors into capital allocation. A reminder on our 2016 capital allocation plans. We plan to buy back a minimum of $100 million and up to $200 million of our shares while returning about $50 million in dividends to shareholders. As to our M&A strategy, several years ago, we concluded that we could secure long-term growth with far less volatility in certain high-growth white spaces, such as collaborative robots, while also returning significant capital. That is a shift we executed on in 2015.
As the leader in automated test equipment, or ATE, we went down the A, or automation path, as we didn't see any attractive T, or test companies available, and many of our electronics customers were asking for help in automation. This customer pull and the broad applications for cobots in many different industries is what attracted us to the market leader, Universal Robots. Moving now to the details of the second quarter. Our sales were $532 million. Gross margins were 53%. The non-GAAP operating profit rate was 23%, and non-GAAP EPS was $0.55. We had one 10% customer in the quarter. You'll see our non-GAAP operating expenses were $158 million, up $5 million from the first quarter due to higher variable compensation accruals on increased profit levels and the continued expansion of UR's distribution programs.
Total OpEx in the second quarter was up from the year-ago second quarter, $5 million in total, which consists of the inclusion of Universal Robots' OpEx, now running at $11 million a quarter, net of cost reductions in our test businesses. We expect our full-year OpEx, excluding Universal Robots, to be down, and UR's full OpEx will grow year-on-year to the $40 million-$40 million range. We plan to keep spending flat to slightly trending down in our test businesses in the aggregate and to increase Universal Robots spending, particularly in distribution, to fan out our coverage and stay on the 50% or greater growth trajectory. Moving to the segment level detail. SemiTest bookings were $391 million, with demand principally driven by mobility. SoC test orders were $338 million, and memory test orders were $53 million. Memory test orders were flash driven.
SemiTest service orders were $73 million of the total. SemiTest sales were $435 million in the second quarter, with SoC making up $394 million and memory test the balance. SemiTest service revenue totaled $57 million in the quarter. Moving to System Test, orders were $30 million in the quarter, and sales were $49 million. Shifting to Wireless Test, we booked $23 million and shipped $22 million in the second quarter. As previously outlined, a significant decrease in buying from our large customer and slowing smartphone growth rates created a difficult demand environment. At UR, orders in the second quarter were $26 million, and sales were $25 million. Sales for the third quarter are expected to be between $375 million and $405 million, and the non-GAAP EPS range is $0.23 to $0.30 on 204 million diluted shares. Q3 guidance excludes the amortization of acquired intangibles.
The third quarter gross margin should run about 55%, up from the second quarter due to improved mix, and total OpEx should run from 38% to 40%. The operating profit rate at the midpoint of our third quarter guidance is about 16%. Shifting to taxes, our full-year tax rate is expected to be about 13%, down from prior guidance of 17%, due to a higher mix of offshore profits where we have favorable tax rates. At the midway point of the year, sales and non-GAAP earnings are up from the first halves of both last year and the last even year, 2014, driven by our alignment to the high-growth segments in SoC test and the addition of Universal Robots. Combined with solid performance from our memory test and system test groups, we expect growth this year despite the headwinds facing our wireless test business.
We remain committed to our balanced capital allocation strategy of both returning capital to owners as well as prosecuting our M&A pipeline very selectively, with an emphasis on automation, and remain on a path to $2 of annual EPS by 2020 or earlier. Thank you. Turn the call back over to Andy.
Thanks, Greg. Tia would now like to take some questions. As a reminder, please limit yourself to one question and a follow-up.
At this time, I would like to remind everyone, in order to ask a question, to press star one on your telephone keypad now. We'll pause for just a moment to compile the Q&A roster. The first question will come from Toshiya Hari with Goldman Sachs.
Hi, good morning. My first question is on UR. You talked about operating margins potentially improving from 10% this year to, I think you said 15% in 2017. Is that primarily a function of OpEx as a percentage of sales coming down from 2016, or is it more a function of potential gross margin expansion, or both?
This is Greg. It's a function of OpEx growing slower next year than what it grew this year. We grew OpEx quite significantly this year, purposely, to get further ahead in distribution and have a stronger ecosystem. We don't need to continue at that same aggressive step-up next year. We would expect gross margins to be about the same next year. We've got some actions to lower material costs, but we're going to assume for now that that will offset any competition we see. We see the sales growth very high. The higher sales growth and modest OpEx increases next year should end us close to the 15% target. You might recall, when we bought Universal Robots, they were running at about 15%. Again, this year, we wanted to get further ahead in distribution.
Okay, great. My follow-up is on SemiTest. I realize predicting 2017 at this point is difficult, if not impossible. What are your thoughts on the on/off or odd/even dynamic going into 2017? Are there any fundamental or structural changes that could potentially allow 2017 to be an on year?
Yeah, I'll take that one. You're right, it's very difficult to call 2017 this far away. It's sometimes difficult even in the year. There are a few things worth noting, I think. On the memory side, this year is quite a bit lower than it's been running in the past several years. There's a lot of tooling going on around R&D and pilot production for these high-speed interface non-volatile memories. It could be that if widespread adoption in mobile devices happens next year, that the memory market would not only snap back to a $500 million level but could be a $600 million level. That depends on close-end decisions in mobility device makers in terms of rate of adoption, which we wouldn't see yet. That's something that's certainly more probable than it's been in any of the past five years.
On the SoC side, what Greg mentioned about complexity growth is true. We're seeing that the test time increases, due to both complexity and something I alluded to, which is more of this sort of performance tuning algorithms that are being run on the tester, will moderate some of the even/odd swings we've seen in prior years. I think there may still be a bit of that, but it could very well be less steep than we've seen in the past.
Very helpful. Thank you so much.
The next question will come from Timothy Arcuri with Cowen and Company.
Hi, good morning. This is Karl Ackerman on for Timothy. If I may, I'd like to address the wireless test business. When I think about the short-term disruptions in the wireless test space versus the long-term opportunity, how do you guys frame that up in your mind? Obviously, you talked about the issues that are driving the market this year and even next year, but with that in mind, how do you think about the long-term reward to live with some of the short-term volatility in that market, particularly as it seems this business is now losing money and remains highly competitive?
Yes. What we said earlier around the near-term outlook, which is the rest of this year, even extending into next year, is that the rate of technology change in cellular and in connectivity looks to be quite low. However, millimeter wave technology, 28 gig up through 70 some odd gigahertz, opening up that band for both backhaul and mobile device applications is the next tooling wave. It's a bit of a judgment as to when that will start. We're currently thinking that's 2018 for Wi-Fi and more like 2020 for cellular. That will result in another round of tooling surge in the market. If you recall, I mentioned that back in 2012, the tooling surge peaked at a $1.2 billion market for cellular and Wi-Fi combined, down to $200 million now. Huge volatility, huge swing.
The next phase of this, we don't believe will get back to that 1.2 level, because that 1.2 level was a combination of both technology and unit growth. It could very well get the market back up in the $600 million, $700 million range for that
Wave across that several-year horizon. That's how we're thinking about it. We still have very strong products, gross margin differentiation, but just a very anemic environment. We'll be focusing on that next technology wave.
Got it. If I may just ask one follow-up. Looking at your guidance for the third quarter, it looks like margins imply they tick higher by about 200 basis points. Is that all related to wireless tests? Because I think your margins have fallen seasonally seven of the last 10 times, Q2 to Q3. Are there other things impacting the mix there, as we look into Q3?
No, it's not really wireless tests. Wireless tests is down $100 million for the year, so they're not a meaningful contributor to the top line or affecting margins quarter to quarter to quarter. What really is happening is we've talked about our large customer, concentrated buyer, and a lot of those tests were shipped earlier in the year. Obviously for a large customer, the pricing is different than for other customers. Those systems were pulled in and shipped largely through the second quarter, so now we're going back to a more normal mix.
Understood. Thank you.
The next question will come from CJ Muse with Evercore.
Good morning. Thank you for taking my question. I guess first question, I think, Greg, I'm not sure if I heard you correctly, but did you say that you expected revenues to grow year-on-year here in calendar 2016? I guess as part of that, how should we be thinking about the contribution from LitePoint?
We expect revenues to grow year-over-year, but LitePoint will be down anywhere, $100 million neighborhood. It's down considerably. Their market cut in half from $400 million to $200 million. It's the opposite of what we saw a couple of years ago, and this can happen when you have a very large customer with 50%-plus concentration. You can get swung around quite severely. In the short term, we're going to remodel, renovate LitePoint, and prioritize where we place our bets. Like we have with other businesses, we'll get this one back on health and focus towards growth opportunities. That's what we need to do as an operator of wireless tests.
Great. I guess the second question, in terms of your gross margin guide, roughly up 200 basis points quarter-over-quarter, you alluded to mix shift there. Is there also savings from the write-down of LitePoint, or is this really more of moving away from mobility into microcontroller, core analog, et cetera?
It's simply mix in SemiTest is the piece, because our large concentrated buyer has taken most of their testers earlier in the year, Q1, Q2. Now we're back to normal mix. If you guys looked at our mix sometimes back in time, we used to talk about 54% or 56%. We'd bounce back and forth whether we had an even or an odd year, because even year, we had the large concentrated buying. That got thrown off last year with the leases. Generally speaking, when we have a very large concentrated buyer, that pulls the margins down. When they're not buying a particular quarter, we get back to a more normalized margin.
I guess the question is: Are there savings that can accrue to the gross margin line, gross profit line, as you are looking to, I guess, reduce the footprint of LitePoint?
There could. I wouldn't expect those to be significant, CJ. LitePoint has very strong gross margins, even with all these headwinds. They have excellent product differentiation. The market headwinds are so strong. There might be a little bit. The margins are very healthy. I think the opportunities in LitePoint are more in OpEx.
Okay. Makes sense. Thank you.
The next question will come from Krish Sankar with BAML.
Yeah, hi. I had two quick questions. One is, can you characterize the NAND solid-state drive test opportunity you see over the next couple of years? Would that flow through both your, I think your storage test and your memory test business? I had a follow-up.
The extent that applications for NAND or non-volatile memories grow, it universally helps our memory test business, and that's an area of strength for us in flash, especially these new high-speed flash interfaces. As it relates to the SSD itself, the unit, that one's a nascent market still. It's been roughly 20%-25% of our storage test business the past few years. We are hopeful that that would, at some point here, start to move north. At least for the most recent, let's say, two- to three-year period, it's been steady at about that level. The real question will be, as more enterprise applications of SSDs grow, which tend to have higher capacities, higher test times, longer test times, will we see the market start to turn there?
Right now, it's, I'd say, still a pretty small piece of market and piece of our business.
Got it. I think I missed the earlier comment. Did you guys say in calendar 2017, your SoC test revenues would grow, or the SoC test market would grow? I forgot. I'm sorry, I didn't catch that comment.
We didn't talk about 2017. We didn't give numbers. I think the question was, do we expect 2016 to be up from 2015? We said, yes, 2016 should be up from 2015, even with the wireless headwinds, wireless being down $100 million.
What about, do you have any view?
On 2017?
Yeah, on 2017, given that, the usual odd year, looks like there are some positives in terms of test time and advanced packaging.
Right. I'll let Mark take that. He took a crack at that early one.
Yeah. What I said earlier was, in memory, there's definitely some upside next year. I mentioned that, going back to $500 million market looks pretty good, and could be as high as $600 million, depending on the rate of adoption and mobility of some of these new flash technologies. In SoC, we certainly see moderation in the even odd year cycle because of complexity growth giving us more test time, less parallelism, as well as this performance tuning aspect of the tester. I gave a range of the market, $2.1 billion-$2.5 billion, kind of. Beyond that, it's just too early to narrow it.
Got it. Very helpful. Thank you guys.
The next question will come from Edwin Mok with Needham.
Hi. Thanks for taking my question. First question, on the prepared remarks, you mentioned something about the analog and microcontroller market is down this year. Can you give us some color around that and where you guys stand with your product?
Yes. Both of those markets are relatively weak compared to prior periods like this. Our market share and our product position there is very strong. It turns out we are disproportionately impacted if those are soft markets, and we're disproportionately benefit when they strengthen. The issues, there was some significant tooling last year in power linear. We had a very strong year there. I think this year, in that aspect, there's been some digestion. In MCU, it's really just low growth. The end pull-through demand for MCUs seems to just be, at the moment, flat lined. There is some expectation that applications in IoT would propel growth of MCUs and MCUs with embedded RF. That's something we bet on. It's something we've got a good product position for, but it's not a market that's truly taken off yet.
Last year, big year in power, this year down. MCUs kind of flat year-over-year, but running below the trend line.
Okay, that's helpful. Then on UR, I think you guys said that you expect your revenue to be somewhere around $90 million-$100 million this year. If I did the math on the first half of the year, and I picked a midpoint of that's like a 6% growth in the second half and the first half. That seems pretty slow, given that last year, I think you guys were 60% from first half to second half. Can you give us some color why there's a deceleration of growth, or am I reading too much into that number?
Yeah, I'll take the part of that. It's the Universal Robots UR3 was introduced mid-year last year. We're comparing against the six-month period in 2015, and there was a very small amount of UR3 Cobot sales. A new market was opened up, this tabletop UR3 market. Now, that market, we had sales in the second half of 2015. I think the growth in the second half of the year relative to the year-ago period. But for the year, we would expect to be 50%-plus growth.
We did in the first half of the year about $42 million of sales in UR. To get to 90, we're talking about $48 million in the second, to get to 100, we'd be talking $58 million. At this point, that looks very doable, there's certainly, given the growth we saw from Q1 to Q2, there's certainly upside to that as well. At the moment, that's how we're modeling the second half.
Okay, great. Thank you.
The next question will come from Farhan Ahmad with Credit Suisse.
Hi, thanks for taking my question. My first question is just regarding your long-term model. Thanks for providing details around the $2 target. If I look back at your revenue and EPS in 2010, you already had $2 of EPS back in 2010. Since then, your revenue has grown basically low single digit from 2010, but the OpEx has gone up by 40%-50%. If I look at every even year to even year, your operating margins are declining. If I think in that context, why are you not more focused on cutting your OpEx and using that to drive earnings growth? It doesn't seem to me like strategy that you have going forward is very dramatically different from what you had for last five years, which is acquiring companies outside of SemiTest and driving growth through that.
if you could just talk about why is the operating model not expected to be higher profitability, just from why is the OpEx so high relative to your revenues right now, and why shouldn't the operating margins be higher?
Okay, good question. Let me take that one. Our operating margins, our non-GAAP operating profit, the last three to five years has averaged 19%-21%. This year, we'll be close to that. you mentioned 2010. 2010, our operating profit was 28%. keep in mind, 2010 was the year after many companies did very severe cutting in 2009, and it was questionable what the world would look like down the road. we cut very deep in 2009, and therefore, in truth, 2010 had a lot of makeup revenue and also a very low cost structure that we had. back in 2010, we also lowered engineering quite a bit. We lowered a whole bunch of other costs, and frankly, we were operating at very healthy profits in 2011, 21%; 2012, 23%. we have to also factor in what is our major competitor doing.
Our major competitor, at the same time, was operating with very low profits. we had to put some defenses or put more money back in R&D so that we could continue our leadership. there's a real kind of challenge that as much as we'd want to get profits mid-20s or something like that, we can't do that at the risk of opening up various accounts. part of our efficiencies, while they're very good compared to peers, we're always looking to improve them further, but sometimes we're constrained by what the other guy is doing. How much he's investing in R&D, how many apps people he has at various accounts. I hope that helps answer your question. The other thing I think different going forward is we talked about the market had been declining for a long time period. We see that changing.
we're going to continue on our same playbook that is gain share, that share gain didn't necessarily help us because the market declined. Thank God we got it, had we not gotten it would've been a terrible story. for us to grow earnings, if we finally have a market that's not declining and it's neutral to 2% or somewhere around there, that's a big change for us year after year after year.
One other thing I would add is, as Greg mentioned, we have turned the corner on OpEx in our core businesses. If you look at this year, our core business' OpEx will be down. UR, the new opportunity with high growth, is up. Across the next several years, we do expect OpEx improvement in our core businesses. That is part of the equation here.
Got it. Thank you. Then one question related to the wireless. Why are you so sure that there's no 802.11ax 802.11ad in 2017? Is there any chance that you may just get an earlier ramp next year?
We're not entirely certain, we want to be cautious with this community because if it comes earlier, great. Our guess is there will not be large volume production buying. If there is, that'll be good news. We just want to be cautious.
Got it. Thank you. That's all I have.
The next question will come from Weston Twigg with Pacific Crest Securities.
Hi. Yeah, thanks for taking my question. I was wondering if you could just remind us what your memory exposure is between NAND and DRAM, then more specifically, just wondering with the big ramp in 3D NAND this year and next year and the potential for very high bit growth next year, why maybe you're not a little bit more bullish on the memory test market?
We're heavily concentrated in flash versus DRAM. I think, on a revenue breakdown, we're probably, it's 80/20 this year toward non-volatile. That's the first part. Second part is next year could be a big year. I mentioned it could go from $400 million this year to $600 million next year. That's, on a percentage basis, a pretty big jump over 15%, and on a normalized $500 million basis, a 20% bump. That's a rough bogey, but what would drive that? It would be a majority of cell phones adopting these high-speed interfaces. The bit rate growth is certainly one thing, but we've been in a high bit rate growth environment in NAND for many years, and that's yielded this stasis of about a $500 million memory market.
The interface change is the real interesting technology shift here, and the adoption of that is the wild card for next year. If it doesn't happen next year, it'll be the year after. It's coming. It's hard to call next year.
Okay. That's really helpful. The other question I had was in wireless test. I know you're talking about the market coming down quite a bit, but how can you be so confident that you haven't simply lost some market share? I think you said that you're fairly confident that you've continued to gain market share in wireless test.
We have good relationships at the major buyers, and we see who's getting what. What really has happened this year, mathematically this year, we will lose market share because our large customer, where we have high market share, is buying a lot less. Mathematically, this will probably be the only year we lose market share. That's because we're in with one customer very tightly, and all the factors that we talked about earlier, including operational efficiencies, are playing out in that account in this year.
Okay. Got it. Thank you.
The next question will come from Patrick Ho with Stifel Nicolaus.
Thank you very much. Mark, in terms of the increase in capital intensity trends that you're seeing on the mobility side of SemiTest, I kind of understand it from the complexity of the devices and the higher test times. How do you see that thesis or trend migrating to other devices, whether it's analog microcontrollers? Is that still some time away when they reach that kind of inflection point where complexity for them requires an increase in testers?
Yes, it is, frankly. The things that drive the complexity growth tends to be centered on large digital content devices and their peripheral devices. Lower lithography nodes enable that and also create this larger, let's say, process variance equation that has to be solved with this other trend I mentioned, which is more extensive trimming and optimizing the part it tests for performance. All of that gets concentrated around the lithography node migration and growth of transistor count. In analog and in MCUs, really there hasn't been as quick a migration to lower lithography nodes and more transistor counts. It's a slower trend line there.
Maybe as a follow-up to that question, at some point, you'll reach that kind of balance once again on the mobility side, where your enhanced tester performances will meet some of the complexity tester capabilities. What kind of runway do you see right now in terms of this change that you're seeing where the tester buys may increase? Is this kind of a two- to three-year phenomena, or is this something that can last a little bit longer?
I don't think this is an issue of the tester will catch up. The question is how long will the transistor growth continue in these devices? Because the transistor growth and the lower lithography node migration are the two key factors. Certainly over the next, how far we have visibility on that, at best 18 months out. We have roadmap information that suggests it could go another three to four years beyond that. Frankly, after that, nobody, our customers don't have a lot of clarity there. I would say pretty good certainty over the next couple of years, reasonable, maybe three to four, and after that, we're really speculating.
Great. That's really helpful. Thank you.
The next question is from Steven Chin with UBS.
Good morning. This is Neil on for Steven. My first question is about the $2 in EPS target. You said about $0.35 of this will come from SemiTest. Can you give us an idea of how much of this will be SoC and how much will be upside from memory?
The majority is SoC test. The SoC test is a larger market. We have higher share there. That's where we see the significant majority of it coming from.
All right. Thank you. If I could have a follow-up related to future M&A. After the goodwill write-down in the wireless segment, does this change your approach to future M&A at all?
As I mentioned in some of my comments a couple of years ago, we did shift our M&A focus to the A in automated test. We saw automation as a better place for us to get new growth because there are markets such as collaborative robots, where there's very broad applications and there's many different industries. It's basically the opposite of wireless test, where you had a 50%-plus customer. We didn't see a competitive landscape at this point. It's sort of a new white space. We are looking much more towards automation because we see automation can work next to our testers, whether it's at wireless test, storage test, or printed circuit board. Automation next to a tester is a small part of the automation market, so we like that factor, too.
All right. Thank you.
The next question will come from Atif Malik with Citigroup.
Hi, thanks for taking my questions. The first question, understand the concentration with the customer on the wireless test side. Even if the units are up year-over-year next year, and the standards, I understand, are difficult to call. Even if the units are up next year, you would expect the wireless test opportunity to be down next year?
Look, it's hard to be precise, but it depends. Are units up? Who gets those units? There's a lot of factors. What's going on with test time? What can be further done with operational efficiency? It's less clear, but we, at this point, just want to signal that we would expect probably a flat market with this year, next year. We may be surprised. There may be upside because units are up with a new product, and that would be wonderful. At this moment, we'd just rather be cautious in terms of what we have high confidence in the next 24 months.
Okay. As a follow-up, can you help us quantify the opportunity of the advanced packaging side, integrated fan out, slim, these new packages? If you kind of zoom in on that opportunity, what kind of growth are you expecting this year over last year? What should we be thinking about the adoption of these packages year-over-year for next year?
Yeah. There's a variety of advanced package types, but in general, that change or that shift is a relatively small component of what we see driving the business. It does inflate the tester demand for a small class of devices by maybe 10%, because of complexity issues with that package, as well as some efficiency of parallelism with that package. It's still, we're talking about 10%-ish of the market. Even as it proliferates into other types of devices and other manufacturers, I don't think that's going to be a key driver, whether it happens quickly or slowly to our business. The main key driver is this complexity growth, the reduction of the impact of parallelism, and this performance tuning aspect of the devices. Okay, operator, we'll take our next question, please.
Yes, sir. The next question will come from Sidney Ho with Deutsche Bank.
Oh, thanks for taking my question. Related to the SoC test, this is for next quarter, how much conservatism do you think you have built into the guidance? If I look at your past five years, on average, your orders in the previous quarter, in 2Q, translate to about 100% of revenue in the following quarter. Is there something unique this quarter that is different, or is there a seasonality factor that we should consider?
Well, yeah, the key seasonality or core issue this year is that everything has been shifted by about a month. Because our revenue swings can be 2x from the trough to the peak, it has a significant impact at the quarter boundaries. A little bit of what traditionally would have been third quarter orders and shipments moved into second quarter orders and shipments this year than in the past several years.
Okay. That's helpful. Going back to the LitePoint side of things. In the past, you talked about LitePoint had pro forma operating margin similar to corporate average. Obviously at a much lower revenue level today, margins have to be low and probably a loss. What can you do to get your margin back to your target on a pro forma basis? Is it just growing your revenue into it, or is there some OpEx reduction that's left to be done?
Historically, LitePoint has had higher gross margins than the company's average. I think of LitePoint more as a software business with hardware, but heavy software content. They've had very good gross margins, always have. Where gross margins have been under pressure a little bit is cellular, but when you put the whole thing back together, the gross margins are strong at LitePoint. Where we see the opportunity to get the operating margin back to health, now, by the way, the operating margin, after the OpEx, has been above our 15% target ever since we bought LitePoint. Historically, it's done well. What we now need to do is look at our OpEx and make some priority changes in terms of what programs, what technologies, what accounts, what are we going to pursue because it strategically fits our strength and what do we not do?
We've got an exercise underway to figure out how we reposition wireless tests or LitePoint back to health and growth.
Great. Maybe just to clarify. In your Q3 guidance, which implies down roughly $5 million-$6 million, should we expect a more step down with OpEx from the LitePoint business going forward? Or is that the right revenue of the OpEx base to drive growth from here?
We will be trimming LitePoint expense, going forward. When it hits the numbers, we'll probably hit the numbers in the fourth quarter. We would look to trim those expenses so we can have a healthier business.
Okay, great. Thanks so much.
Operator, we have time to just sneak in one more quick one if you could, please.
Yes, sir. The final question is a follow-up from CJ Muse with Evercore.
Thank you. Just a quick question. How should we be thinking about tax rate going into calendar 2017 and 2018?
Yeah, that's a good question because it's getting pretty tricky, the tax rate. We have a mix of businesses that have very different tax rates. You can see what can happen in a year like this year where LitePoint, a U.S. business, is down. SemiTest is doing quite well. Universal Robots is growing. That all pulls our rate down. One, it's sort of a good news, bad news. Our tax rate has gone down considerably to about 13% this year. We are doing work to kind of figure what is it long-term when you normalize the businesses, but our rate probably will be closer to 19%, 20% on a long-term basis. Perhaps less than what we thought earlier, because we're finding more and more of our profits are offshore.
In long-term, you're referring to 2017 or that's further out?
I'm referring to 2017 to 2020, CJ.
Okay.
2017, I could be a little surprised if it's possible it looks like this year, but I think LitePoint will be healthier next year, and that's U.S. income.
Got you. Great. Thanks so much.
Okay. Thank you, everyone. This concludes today's call, and we appreciate your interest in Teradyne and look forward to talking with you down the road.
Thank you.
Ladies and gentlemen, thank you for participating in today's conference. You may now disconnect.