Hello, and thank you for standing by. Welcome to the Teradyne Q3 2022 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speaker presentation, there will be a question-and-answer session. To ask a question during the session, star one one on your telephone. It is now my pleasure to introduce Vice President of Investor Relations, Andy Blanchard.
Thank you, Andrew. Good morning, everyone, and welcome to our discussion of Teradyne's most recent financial results. I'm joined this morning by our CEO, Mark Jagiela, President Greg Smith, and our CFO, Sanjay Mehta. Following our opening remarks, we'll provide details of our performance for 2022's third quarter, along with our outlook for the fourth quarter of 2022. The press release containing our third 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. Replays of this call will be available via the same page after the call ends. 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, and 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've 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. Looking ahead between now and our next earnings call, Teradyne expects to participate in technology or industrial-focused investor conferences hosted by Baird, Credit Suisse, and UBS. Now let's get on with the rest of the agenda. First, Mark and Greg will comment on our recent results and the market conditions as we enter the new quarter.
Sanjay will then offer more details on our quarterly results, along with our guidance for the fourth quarter. We'll then answer your questions, and this call is scheduled for one hour. Mark?
Thanks, Andy. Hello, everyone, and thanks for joining us this morning. In today's call, Greg will review our Q3 results and Q4 outlook. I'll then describe the longer-term view, and Sanjay will provide the financial details of the quarter. Greg?
Thanks, Mark, and good morning, everyone. Teradyne's third quarter sales and profits were above the midpoint of our guidance in revenue and profit, with sales of $827 million and $1.15 in non-GAAP earnings per share. At the company level, the quarter unfolded generally as planned, but at the operating level, we cleared some supply bottlenecks, which allowed us to ship more semiconductor test products than planned in the quarter. On the other hand, Industrial Automation was weaker than we forecast in July. Putting the third quarter into context, when we spoke in July, we noted that slowing SoC and wireless tester demand in the second half were related to declines in the end-market shipments of smartphones, compute products, and associated infrastructure.
With that backdrop, we reduced our estimate for the SoC market size and reduced our second half shipment plan approximately $300 million. About 1/3 of that amount was already confirmed in June, and about 2/3, $200 million, was our judgment on how the second half would unfold. In the subsequent three months, about $110 million of the $200 million additional decline was realized, and about $90 million was not, as overall test demand held up better than we expected in July. This incremental strength in test has been partially offset by weaker IA demand as reflected in the Q4 guidance. The highlight in Semi Test is the continued strength in automotive test demand. This year, we're seeing a step increase in our shipments for ADAS processors, silicon carbide drive and charging ICs, and battery management devices.
Our UltraFLEXplus platform is well-aligned to the unique requirements of multi-billion transistor ADAS processors, and our Eagle platform is well-suited for the thousands of volts and hundreds of amps often required for silicon carbide devices. It's notable that these three device types collectively will add over $100 million of incremental SoC TAM in 2022, and they were insignificant five years ago. Shifting now to Industrial Automation. Because only about 30% of our sales are transacted in dollars, and there have been significant changes in exchange rates in 2022, we'll describe results in both constant currency and dollar terms. In constant currency, automation group Q3 revenue grew 7% from last year's Q3 and is up 19% for the first nine months of the year.
In dollar terms, however, revenue contracted 2% in the quarter compared with Q3 2021, and growth is 11% for the first nine months. Looking at the full year, we expect IA growth from 2021 will be about 6% in dollar terms at the midpoint of our Q4 guidance. In constant currency terms, IA growth will come in at about 14%. Even using constant currency terms, this is below our 2022 plan to grow in the mid-30s from the 2021 level and reflects two additional factors. First, slowing industrial activity, especially in Europe, where PMIs dropped below 50 in July and have remained in that contraction zone since. Europe is our largest end market for automation, and this is a 10-point headwind to growth.
Second, labor scarcity continues in our distribution channel, which we expect to reduce growth by about 5 points. While 2022 growth is below our plan, we have a number of bright spots in IA, and I will highlight just a few. At Universal Robots, customer demand for our higher payload, longer reach UR20 has been stronger than expected since its introduction mid-year, and we expect it to be a meaningful contributor to results when it begins shipping in 2023. The UR20 expands our capability in the palletizing, welding, and machine tending end markets. Also, we've continued to grow our OEM channel at UR, including supporting partner expansion from national to international sales coverage. The OEM channel is a powerful growth driver for us.
For example, our welding channel grew over 80% in the first nine months compared with last year, and for this year, we expect to ship well over 1,200 robots in that vertical. At MiR, our sales to large customers continue to expand, with more than 30 customers having fleet sizes greater than 20 robots. The number of customers with these large fleets has increased 30% in 2022. This is significant as our fleet management software is an increasingly important differentiator as fleet sizes grow.
On a related note, at the end of Q3, we merged AutoGuide into MiR as we prepare to offer customers a broad payload range from hundreds to thousands of kilograms, sold by the same team, using the same fleet management software, backed by the same global distribution, same global network of distributors, all while increasing our engineering, marketing, and back office leverage. Finally, MiR service and spares revenue has more than doubled to about 5% of revenue through the first nine months compared with 2021 as we begin to expand our customer support offerings. Across Industrial Automation, the investments we're continuing to make are giving us the foundation to support high growth with differentiated products to global customers through multiple distribution channels.
While we are not satisfied with 2022's growth rate and have taken actions to improve, we remain confident of our long-term strategy in these growing markets. With that, I'll now pass it back to Mark.
Thanks, Greg. I'll comment on the longer term view of our markets and the impact of changes in the regulatory environment. Although better than estimated in our July call, the demand has slowed across our test markets as we enter Q4. As I noted at the time, predicting the depth and duration of these corrections is challenging, but in each of the last three corrections, growth has returned after six to 12 months. We haven't seen anything in the last three months to change that view. While it's too early to offer estimates on 2023 full year market size, based on the historical patterns and current market conditions, we expect next year's first quarter and first half demand to remain relatively weak. That, combined with the weak macro environment in industrial markets, suggests that our Q1 revenue will likely be about 10% softer than Q4.
Looking at the full year 2023, it's too early to make quantitative predictions, but I'll point out how we're thinking about the balloons and anchors for the year ahead. First, the balloons. We expect the semiconductor industry to begin the transition to 3 nm technology and our largest customer to grow from less than 10% of company sales in 2022 to more than 10% next year. However, we expect them to be at a lower than historical past peak level as 3 nm will not hit full stride until 2024. In Industrial Automation, we also expect to grow year-over-year with the exact amount governed by macro conditions. Balancing these balloons are a few anchors. Semiconductor unit volumes and revenue are expected to decline in 2023 due to inventory digestion and macro headwinds.
In Industrial Automation, slowing industrial growth worldwide and a pronounced slowdown in European manufacturing are additional anchors. We're keeping a balanced view of next year, and we should have a better view of the full year in January. Looking longer term, the case for growth in both test and IA remains intact and compelling. In Semi Test, we're operating on a market growth trend line of 10% since 2017. Growth in our test businesses is driven by unit growth and complexity. We expect those drivers to remain in place as new semiconductor technologies, such as 3 nm, gate-all-around, and chiplets ramp over the next few years. These technology building blocks enable applications to continue to grow in complexity, including emerging applications like AI, ADAS, and new compute architectures. In Industrial Automation, we remain bullish on the future.
While we're not immune to macro conditions, we recognize the underlying drivers of demand are favorable for cobot and AMR automation. Labor demographics and economics and the growing range of tasks our cobots can perform ensures an expanding market opportunity, and our strategy enables us to capitalize on that opportunity for many years to come. Regarding the recent changes in trade regulations, indigenous Chinese memory makers have been a significant and growing part of our memory business, and while the new regulations mainly focus on fab equipment rather than test, tighter restrictions combined with potential reduction in their wafer volume will likely impact our business. This situation is dynamic, but as of now, we see these regulations as about a $75 million-$100 million headwind going into 2023.
On the supply side of the trade equation, we have a large manufacturing footprint in China, and our global support team has a professional services operation in place to support local and global customers. As part of our resiliency planning, we have been enabling additional global sites for these activities. Given the current environment, we will be accelerating the capacity growth of these other sites. Sanjay will take you through more details of those plans. Stepping back to look at the full year, despite the drop in demand in the second half, we're on track to deliver our second highest revenue in history. We balanced our growth investments with financial discipline, and we positioned Teradyne for high growth as our core test and IA markets improve. Sanjay will now take you through the financial details. Sanjay.
Thank you, Mark. Good morning, everyone. Today, I'll provide details on our Q3 results, update you on the impact of the new trade regulations, and describe our Q4 outlook. Now to Q3. Third quarter sales were $827 million with non-GAAP EPS of $1.15. Non-GAAP gross margins were 58.7%, and our non-GAAP operating expenses were $247 million, about flat with second quarter OpEx due to implemented spending controls. Non-GAAP operating profit was 28.8%. We had one 10% customer in the quarter. The tax rate, excluding discrete items for the quarter, was 18.7% on a non-GAAP basis. Please note you should now use 17.25% for the full year non-GAAP tax rate.
Looking at the results from a business unit perspective, Semi Test revenue was $576 million, with SoC revenue contributing $451 million, and memory of $125 million. We cleared some supply constraints which enabled the highest level of memory shipments in the past two years. In memory, the transitions to new memory standards in both flash and DRAM, along with strong demand from indigenous China customers, drove the results. In SoC, the increasing semi content for EVs, autonomous driving, and infotainment in autos, along with continued strength in analog industrial, contributed to the demand. System Test Group had revenue of $116 million, with storage test delivering $70 million, and defense and aerospace and production board test $46 million.
In wireless test at LitePoint, revenue was $46 million in the quarter, down 33% year-over-year, primarily because of lower Wi-Fi revenue tied to declining PC and smartphone end markets. Now to Industrial Automation. IA revenue was $89 million, with UR contributing $73 million and MiR $16 million. MiR results include AutoGuide. As Greg noted, a large portion of IA sales are outside the U.S. For the first nine months, 41% of IA sales were in Europe, 29% in the U.S., and 11% in China, and the remainder in the rest of the world. From a financial perspective in IA, the group was slightly above break-even on a non-GAAP operating basis in the third quarter, and for the full year, we expect to be about break-even, below the 5%-15% profit range plan for the group.
As growth has slowed this year, we've also slowed the growth in OpEx. Shifting to supply. Looking at Q4, we continue to face supply issues, but the supply trend continues to improve relative to three months ago. These supply bottlenecks noted at the beginning of each quarter have moved from $50 million in Q3 to about $15 million in the fourth quarter. These amounts are outside the high end of our guidance range. Switching to a strategic issue. We manufacture certain products and deliver certain engineering services in China for our global customers. While recent U.S. government actions have no immediate impact to these operations, due to concerns about potential future impacts, we are accelerating our strategy to expand our manufacturing and engineering service capabilities in other countries.
In manufacturing, this work has been underway for several years, and we're accelerating the work to complete our expansion strategy in the next several quarters. In engineering services, the transition will likely take most of 2023 to achieve. Shifting to the balance sheet and cash flow.
Ladies and gentlemen, please stand by. Ladies and gentlemen, please stand by. Your conference will resume momentarily. Please stand by.
Did John say where we lost it?
Ladies and gentlemen, please stand by. Your conference will resume momentarily.
Hey, Andrew?
Yes, please go ahead.
All right, we're ready to go?
Yes, we are live with the audience. Please go ahead.
Back to you, Sanjay.
I'll go back to shifting to the balance sheet and cash flow. Our cash and marketable securities at the end of the quarter totaled approximately $900 million, about flat with Q2. We had $233 million in free cash flow in the quarter, and our DSO came down from 74 to 58 days. Uses of cash in the quarter included share buybacks of $217 million, dividend payments of $17 million, and debt retirement of $10 million. Due to opportunistic buying through the year, we finished our $750 million share repurchase plan for 2022, a full three months ahead of the original schedule. Our board has authorized continued opportunistic share repurchases in Q4.
Since our program began in 2015, we've repurchased 74 million shares at an average price of $45.85. Regarding debt, to date, $397 million of convertible bonds have early converted. Now to our revolver. During Q3, we worked with our banking partners and have increased the capacity of the revolver to $750 million from $400 million. Our credit facility is in place until December 2026. Now to our outlook for Q4. As Greg noted, our outlook is ahead of what we expected three months ago, driven mainly by Semi Test. Sales in Q4 are expected to be between $670 million and $750 million, with non-GAAP EPS in a range of $0.62-$0.86 on 164 million diluted shares. Fourth quarter guidance excludes the amortization of acquired intangibles.
Fourth quarter gross margins are estimated at 56%-57%, down from Q3 due to lower volume, product mix, and spending on accelerating our manufacturing and service delivery resiliency. OpEx is expected to run at 34%-38% of fourth quarter sales. On a non-GAAP operating profit rate at the midpoint of our fourth quarter guidance is 21%. Looking at the full year of 2022. At the midpoint of our guidance, 2022 revenue will be $3.1 billion with an EPS of $4.08. Gross margin for the full year should be approximately 59%. Regarding OpEx for the full year, we spent a bit lower than planned in Q3 in both test and IA businesses.
We expect the full year OpEx will be about $1 billion, up about 3% from 2021 at the midpoint of our Q4 guidance. Recall our operating model flexes OpEx with revenue, so growth has naturally trended down through the year. A couple of points looking ahead at 2023. First, gross margins. I will note that because of the additional spending on supply resiliency and forecasted product mix, we expect Q1 gross margin will be similar to Q4's level. Second, tax rate in 2023 is forecasted to be at 17%. We'll update our midterm operating model in our January call. Summing it all up, we're expecting higher revenue and profit in the second half of the year than prior update in July.
Declines in our test portfolio were less than forecasted, offset by weaker IA business, driven by FX, continued economic challenges in Europe, and labor scarcity in the channel. For 2022, OpEx spending will increase year-over-year approximately 3% versus the 11%-13% anticipated at the beginning of the year, as revenue forecasts have declined from our view in January. In a challenging environment, regardless if the driver is economic downturn, regulatory changes, supply shortages, labor scarcity, or whatever may come our way, we're confident we have a great team, strategy, operating model, and experience to lead us through whatever lies ahead. We will continue to focus on our customers and the long-term opportunities in test and industrial automation markets. With that, I'll turn things over to Andy.
Thanks, Sanjay. Andrew would now like to take some questions. As a reminder, please limit yourself to one question and a follow-up.
Certainly. Also, as a reminder, to ask a question, you will need to press star one one on your telephone. Please stand by while we compile the Q&A roster. Our first question comes from the line with Krish Sankar. Pardon me, Krish, please check your mute button.
Hi. Can you hear me?
Yes, we can.
Yeah. Hey, thanks for that. Transitioning to my question. I have two of them. First one, Mark, I'm just kind of curious. You know, you're seeing industrial automation slow, you're seeing mobile compute slow, and then I understand the demand drivers for ADAS and auto. Do you worry that auto and industrial test is the next shoe to drop? Or do you think that it could be resilient into 2023?
Yeah, I think that, you know, industrial is probably something we're looking to soften a little bit in 2023. Automotive at the moment is just. There's no sign of that. It's very strong. We're at a year where the market size for automotive semiconductors is all-time peak. The ADAS and EV content is on such a dramatic rise that it could very well be that next year has a flattening. At the moment, we don't see it softening, certainly not through Q1.
Got it. That's very helpful. As a quick follow-up, I think you mentioned that the China impact was probably $75 million-$100 million headwind in 2023. I'm kind of curious, you mentioned a lot of the exposure was through memory. What about the China OSATs? I thought you had pretty high exposure over there too. Is that part of this $75 million-$100 million headwind?
Most of that headwind is memory. There's not really a lot of incremental restrictions that we have related to China OSAT. That's not really a headwind for us.
Got it. Thank you very much.
Thank you. Our next question comes from the line of C.J. Muse with.
Yeah. Good afternoon or good morning. Sorry. Thank you for taking the question. I guess I was hoping to get a little bit more detail on gross margins. I think you talked about manufacturing, engineering, resiliency, and kind of investments outside of China negatively impacting margins into Q1. I guess can you give a little bit more color on the plans there and how we should think about the impact to overall calendar 2023?
Sure. Hi, C.J. It's Sanjay. Addressing gross margins, and I'll get to the resiliency plans. Think of Q4 as roughly a third volume decline, a third product mix, and a third resiliency spending of, you know, the bridge from Q3 to Q4. A lot of that continues into Q1. Getting into the resiliency plans. You know, for the past couple of years, we've been working on what I call a multi-country manufacturing approach. We've obviously grown over the last several years, and we had to increase capacity. We fundamentally took the approach to have redundant capacity in both different geographies, and not just for manufacturing, but from a supply chain perspective. From that, we've chosen to accelerate over the next couple of quarters.
What you're seeing is an accelerated spend on the manufacturing. From an engineering services perspective, again, to create capacity in different locations. That'll take a little bit longer into the second half of 2023.
Very helpful. As my follow-up, you talked a bit about 2023 outlook and highlighted your top customer accelerating higher, but not getting back to prior peak. I guess, you know, any color there in terms of kind of the diversification of the business, and/or when you'll have, you know, clear visibility to what kind of the magnitude of the growth might look like?
Yeah. As usual with our largest customer, the visibility comes pretty late, you know, April of next year. We do get signs as early as this time of the year, which is why we have, you know, the sort of commentary on it going back above 10% for them. You know, it's not gonna go back to greater than 20%, we don't believe, where we've seen prior peaks, primarily because the 3 nm move in 2023 will be for a fraction of the phone models that will be introduced. That will expand as we go beyond 2023, and we would expect it to continue to grow, therefore, that business piece. Outside of that, you know, we've talked a bit in these calls about design wins in emerging vertically integrated non-traditional semiconductor companies.
That should be, you know, another growth engine for us as we look into 2023. I didn't call that out as a specific balloon, but that sits there as well, on top.
Thanks so much.
Thank you. Our next question comes from the line of Timothy Arcuri with UBS.
Yeah. Hi. Thanks. I'm wondering if, Mark, you know, maybe you can help bridge the gap between Q4's better because things aren't as bad as you thought when you guided, particularly on this, you know, Semi Test side, you know, mostly sounds like because of supply, but then Q1 being worse. It, you know, to get Q1 down 10%, you have to assume a, you know, pretty significantly worse than normal Q1, which I assume would mostly come from Semi Test. Can you sort of help us bridge that gap? Thanks.
Well, you know, first of all, in Industrial Automation, our Q1 is always down from Q4. That is a factor here. It's, you know, reasonably significant. That is one thing. In the case of Semi Test, we do see some incremental weakening in industrial, which we spoke about. The China headwinds we talked about around memory will start to impact us in Q1 as well. The combination of those two are kind of what leads us into that estimate for Q1.
Got it. Okay. Can you also talk about, you know, I know that it's a bit early to kind of talk about what the outlook is for the full year next year, but can you talk about any sense of directionality for the SoC TAM? I assume that the memory TAM will be down clearly. You know, you have front-end WFE down 40% plus next year for memory. So I assume that the memory test TAM would be down, but front-end foundry and logic WFE TAM is not down that much next year. It's down probably 5%, you know, maybe 10%. So do you have any sense in terms of where the SoC TAM could go next year, at least from a directional perspective? Thanks.
Well, it's interesting. First on memory, you know, although you're right on the WFE estimates, a lot of what's gonna happen in memory, especially DRAM next year, is the shift to DDR5 finally for server applications. That requires a brand new test platform. We actually think that memory next year for a test will actually be pretty good. It may be, let's say, flattish with 2022 levels. It could, you know, because on the NAND side, it could come down a little bit, but we're pretty optimistic given what we're hearing from our customers because of the DDR5 transition.
In SoC, from a TAM point of view, I would expect the overall TAM to be a little bit weaker because the late in the year declines in the PC markets, the compute markets, I think will precipitate into a lull in capacity adds early next year. That will get offset, of course, by, let's say, you know, some mobility growth related to 3 nm. But the fall off on the PC side, the impact of the fall off on the PC side might be more pronounced than the growth on the mobility side. You know, let's say automotive remains relatively strong, industrial weakens a bit. I could see it being marginally down next year. The first half, much more so.
You know, I think that's where we're gonna see the sort of digestion around capacity for PC test and such come to fruition.
Great. Awesome. Thank you so much.
Thank you. Our next question comes from the line of Samik Chatterjee with JP Morgan.
Hi. Good morning. Thanks for taking my questions. For the first one, if I could start on gross margin. You gave us some color for Q1. I was just thinking if you can give us a bit more color on the trajectory for the year, particularly the implications of the drivers of growth that you're outlining, which seems to have got a lot of the growth will come from your primary customer next year and some from IA, but there's obviously demand slowdown across the board as well as the memory headwinds that you outlined. And there's also, I guess, supply improvement. As we sort of look at the puts and takes, how should we think about the trajectory of gross margin given the growth drivers? I have a follow-up. Thank you.
Sure. First, let me ground you in, as I've said in my prepared remarks, this year, we plan to come in approximately 59% or on model. You know, there's no change to our operating model that we're managing the business at 59%-60%. I think I outlined in Q1 that we expect some headwinds tied to product mix as well as spending on resiliency accelerations. We're just going through the planning phase, you know, right now for 2023, and we'll be able to provide a little bit more color in the January call. You know, we fundamentally manage the business at 59%-60% gross margins. In any given quarter, those are gonna oscillate tied to product mix, and kind of what's happening in the environment.
Okay. For my follow-up, I think I heard you say for the Industrial Automation slowdown that you're seeing some level of that slowdown come through because of labor shortages. I'm not sure if I heard you correctly there, but I'm just curious if what you're hearing from your customers is more of a pullback in terms of capital investment plans that's driving the slowdown that you're seeing right now, or is that sort of more of a next step? Right now it's more about labor shortages that you sort of outlined. Thank you.
I pointed out there were sort of two factors involved. There is a slowdown in spend primarily in Europe due to economic conditions. There you do see some customers that are implementing capital freezes or limiting their investments. In other regions, the labor scarcity in the channel is the primary factor. What we see with our partners is that they actually are maintaining very high project backlogs, and they're trying to work through those projects and add to their capacity at the same time so that they're able to support a higher level of business. We haven't seen a pullback in regions other than Europe. It's much more a constraint, a capacity constraint in our traditional distributor channel.
All right. Thank you for the clarifications. Thank you.
Thank you. Our next question comes from the line of Mehdi Hosseini with SIG.
Yes. Thanks for taking my question. The first one for Sanjay. In a scenario, and I'm not asking for a guide, in a scenario where your 2023 revenues would be flattish, would you expect OpEx flat, up or down? And I have a follow-up.
Yeah, good question, Mehdi. You know, as I said earlier in the Q&A, we're currently preparing our plans for 2023, and we'll provide an update. You know, I will say in January, we thought OpEx was gonna grow for this year 11%-13%. Our operating model, obviously, with the macro environment and our projections coming down for the back half of 2022 and then for the full year, you know, our operating model has a very little compensation declining OpEx, and we also implemented spending controls. What I can guide is that we're gonna have a plan that's integrated and ties to a revenue and a market view. Then if the environmental conditions change, we'll course correct, as you've seen us do this year.
Okay. Let me come back to the question in a different way. You guys have been very proactive in helping us better help investors. You were the first one nine to 12 months ago to talk about 3 nm pushout. This morning, again, you're resetting expectation with 3 nm. Nine months ago, you were referring to a significant front-end capacity being added for MCUs analog. Now I'm looking forward, and some of those front-end equipment being added are may not be fully utilized. Why not take opportunity and be more conservative? Now, I understand you're helping us by giving an overview of Q1, but why not just be more aggressive in being conservative?
Well, I don't think we're being conservative or optimistic in terms of beyond Q1. We just don't have visibility. We could be speculative maybe, but it's just the you know, being a little bit prudent of what we know versus what we speculate. You know, I think the days of inventory in the semiconductor supply chain suggest there's just digestion needed. That should be something that's already starting and perhaps works its way through by next summer in many of these markets. The bottom line is we don't know, and so we don't wanna go out too far on a speculation.
I'm not asking for speculation, but if I go back to nine months ago, we're thinking of yes, considerable amount of front-end investment being added.
Yeah.
Now we're finally looking at the inventory situation that is problematic, 25-year high, and your customers have to cut back on production. Even if they have added investment for front end, those equipment are not gonna be fully utilized. Wouldn't non-compute SoC be weak into mid-year as your customers become more aggressive in working down inventories?
Sure. It's likely that's the case. On the other hand, there's offsets to that I mentioned. Our largest customer probably is gonna be stronger. You know, there's offsets to what is, let's say, likely to be digestion in some of those markets. In, you know, our neck of the woods around mobility, 2022 was unprecedentedly weak because of the factors around product migration and complexity growth. That worked against our neck of the woods. That's gonna work in our favor a little bit next year to offset some of what you're referring to.
Got it. Thank you.
Thank you. Our next question comes from the line of Vivek Arya with Bank of America.
Thank you for taking my question. First one on smartphone test. I'm curious how much of the existing capacity that's for 5 nm can be repurposed for 3 nm test? Is there a large reuse element that we should be aware of? And kind of related to that, I thought that one of the incremental growth areas for next year was the possibility that, you know, if your large customer would insource their 5G modems, that would provide you incremental opportunity, but that seems to be pushed out as well. Just wanted to get your insights on the reuse aspect and, you know, how large customer retaining the 5G modem, what impact does that have on how you're thinking about next year?
Yeah. First of all, testers reuse, yes, 100%. There's no obsolescence in the short term on the SoC AP test side. This has been true for 10+ years. The testers do not often obsolete themselves. We will come to that point at some point in the next, you know, four or five years where that will begin to happen, but not in the short term. On, you know, the drivers of test demand at our largest customer subject for a little bit. I'm gonna answer it a little bit broader than just the modem because I don't wanna speculate or get into our customers' introduction plans on any specific device.
In general, when we look at 2022, you know, now that we understand and it's public that the new processor for phones that was introduced in 2022 only went to the high-end phone models, and that new processor being the last processor built on 5 nm was only a 6% growth in transistor count, so very modest complexity growth really muted the demand from that, our largest customer this year. Drove them below 10% and, you know, that's all kind of clear. Now, if we look forward, the combination of what you described, the introduction of a new line of silicon, when that happens will be a bump. I'm not gonna say if it's when, but that's obviously in the pipe. Another bump is 3 nm for the apps processor for 2023 phones.
If in fact that occurs, you know, it's likely that it'll be a pretty significant complexity bump, not 6%, but probably back in the 20%-30% complexity growth that we've seen historically with a new node. That gets muted a little bit though, because it's likely it will still only go into the top-end phone models, while this year's 6% incremental complex part waterfalls down to the other models. The other models in 2023 aren't gonna see much of a bump, but the high-end phone will. That's what kinda gives us this view that they could go back to being greater than 10%, but less than historical peaks. Let's look forward to 2024. You know, we'd see another significant bump in complexity at the high end.
That phone app processor that was introduced in 2023 would likely waterfall down to the other models, and then the full product line in 2024 is likely to see a significant jump in complexity. This is sort of how the movie plays out, and then on top of that, there's new silicon, there's also high performance computing, you know, all of that moving to 3 nm, but it's just gonna, you know, partially, I think in 2023 and probably more fully in 2024. I'm sorry if that's too long an answer, but
No, no. Very helpful. Thanks, Mark, for that. My follow-up question is, it's interesting that, you are keeping your, calendar 2024 earnings model of $8, which is roughly a doubling from the 2022 level, even though, 2023 seems to be facing some of these, macro headwinds. Mark, what's your gut on what is kind of the bottom of the cycles for you, right? Is it Q1, Q2 next year? Like, when do you start making the recovery to your 2024, plan? Or is the 2024 plan looking ambitious at this point?
We'll update you on the 2024 plan in January, but what I just described gives you some understanding of why we think 2024 is probably a pretty strong year. In 2023, you know, as usual, our quarters are lumpy, so our largest customer tends to tool in Q3 and Q4, so we would expect, you know, a bit of a bump there. The one thing we can't predict very well is the macroeconomic conditions that are gonna occur. You know, the digestion of inventory in the channel and those things, we have some ability to think about and model. But, you know, if the world's gonna be in a recessionary environment, or and the effect of that on both semiconductors and IA is a little bit harder for us at this juncture to suss out.
What I did say, though, is that 12 months is kind of a normal cycle here to go through, so that would suggest come third quarter next year, we should start to see a broader recovery, if it's like anything in the past.
Thank you.
Thank you. Our next question comes from the line of Joe Moore with Morgan Stanley.
Great. Thank you. We're gonna talk about memory tests from the perspective of, you know, the DRAM and NAND vendors are building a lot of inventory right now. Is that untested? Is that like, as their volume picks up, do you see a recovery in test volume that maybe helps you next year? Then on the flip side of that, you know, yeah, I think you talked a little bit about what the deceleration in WFE might mean, but, you know, do you have any kind of sense of what memory supply does next year above and beyond the DDR5 effect? Thank you.
Yeah. I think DDR5 is the only bright spot next year. The rest is kind of very tenuous and, you know, I would expect the first half of next year, we're gonna see, other than DDR5, pretty weak demand as people are trying to balance fab capacity inventory with end market demand. You know, earlier I said there's an argument that memory will hold up next year as a market for test because of the DDR5 transition. I think that's possible. I think it's more likely, though, we see that happen, but NAND fall off a bit stronger, and you know, a little bit more of an anchor on the market next year for memory than a balloon because of DDR5.
We know we have the specific issue around China. Our overall market share in memory test is somewhere around 40%, but in China it's been north of 50%. This issue with, you know, one of our customers in China, which we may be prohibited from selling to, is kind of going to be, for Teradyne, probably a net negative for memory test revenue in 2023.
Okay. On those large inventory builds on the producer balance sheets, is that die bank that has yet to be tested or is there already some test that you think has occurred on that inventory?
There is some test that's occurred, but a lot of it is in that sort of die bank area. You know, keeping the fabs running is kind of a something that makes economic sense, but finishing the product doesn't make sense till the demand materializes.
Great. Thank you very much.
Thank you. Our next question comes from the line of Toshiya Hari, Goldman Sachs.
Hi, can you hear me okay?
Yes, we can.
Okay, great. I had one clarification and one question. In terms of my clarification, just on the China export controls, you talked about the headwind in 2023 being $75 million-$100 million. Just curious how significant China memory test was in Q3. I assume that number should be zero in Q4. Am I thinking about that correctly? Can you speak to the profitability of some of these China memory customers?
Yeah. Well, it's not zero in Q4 or next year. That's maybe one comment. The profitability is not distinctively different than our global, you know, margin, so there's no difference there. There's a portion of the market in China we'll continue to serve. You know, China overall, I think Sanjay mentioned, is, you know, roughly 15%, 16% of the company's revenue. Half of that is multinationals that really are not subject to any restrictions around this set of regulations. About 8% of our annual sales, let's say, are to indigenous Chinese companies. And then, you know, roughly half of that might be memory, and then a fraction of that is this headwind of this one customer.
Okay. Mark, just a quick follow-up. I meant local China memory. That should be near zero or zero going forward, correct?
No.
The multinationals you'll serve.
We'll serve the multinationals, but there's a portion of the indigenous China memory market that we still can serve under the current regulation regime.
Okay. Okay, got it. My question is on the IA side, maybe one for Greg. I realize it's hard to predict what 2023 could look like, but hoping to get your preliminary views there. You talked about labor scarcity outside of Europe being a 5 percentage point headwind. You talked a little bit about, you know, some of the idiosyncratic stuff that you're working on in terms of, you know, the new product, the higher payload, longer reach product. You talked about the OEM channel. Could those be sort of impactful enough for, you know, your growth to accelerate in 2023 or is it a little premature to talk about those things? Thank you.
It is a little premature. We haven't fully worked out our complete plans, but I think you have hit some of the major factors that are gonna go into that equation. One thing is that the reason that we were talking about our results in constant currency is that we do expect a more stable FX environment in 2023, and that alone would add significantly to the growth that we saw in 2022, because that was a like a 10-point headwind to where we are. Beyond that, we do expect significant impact from new products, both from UR and from MiR in 2023.
We also are going to continue to grow our complementary channels like OEM at similar rates to what we saw in 2022, but off of a larger base. Overall, they're gonna have a greater impact on growth in 2023 than they had in 2022. In other words, we are expecting these new things to help out with growth. We believe that our distribution partners are gonna work on labor scarcity, but we are not counting on them to completely solve it in 2023.
Thank you.
Thank you. Our next question comes from the line of Atif Malik with Citi.
Hi, thank you for taking my questions. Mark, your comments on next year's end markets, they all make sense. Memory, you know, stable on DDR5, industrial coming down, mobile improving on the large customer. The only one that I struggle is on the auto side. The auto TAM, you know, around $700 million this year from historically $350 million. Now, $100 million of that, like you said, is coming from new opportunities in ADAS, silicon carbide, power management. But what about the opportunities outside those areas? Like, won't that see a correction next year if there's a recession?
Yeah. I think that's likely. You know, you're right. I think some of the things that have contributed to getting the auto market up to the $700 million TAM level are a lot of these new technologies. Underneath that, there's a kind of a core piece of legacy technologies that I would expect would soften as we get out of this sort of frenzy of trying to replenish the supply chain on those parts. What I was saying is we don't see that coming in Q1. It's kinda not something we baked into Q1. For 2023, that underlying, let's say, legacy piece of automotive technology is likely to come down.
Got it. Sanjay, can you talk about what are the other geographies you're looking at in improving your supply chain resiliency moving out of China?
Sure. From a manufacturing standpoint, you should think about it from both manufacturing and supply chain. Think about other countries in Southeast Asia, Malaysia, Thailand. From an engineering services resiliency, think about Southeast Asia, but also think about other parts of the globe, like in Europe.
Thank you.
Operator, we have time for just one more question, please.
Certainly. Our next question comes from the line of Steve Barger with KeyBanc. Pardon me, Steve. Please check your mute button. Okay, I will now turn the call back over to Vice President of Investor Relations, Andy Blanchard, for any closing remarks.
All right. Well, thank you. Thank you, everyone, for joining us today. Apologies for the technical pause along the way there. We look forward to talking to you in the days and weeks ahead. Certainly, if you have follow-up questions, reach out to me directly. Thanks so much.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating, and you may now disconnect.