Teradyne Inc (TER) — Q1 2022 Earnings Call Transcript

Thank you, Chris. 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, and our CFO, Sanjay Mehta. Following our opening remarks, we'll provide details of our performance for 2022's first quarter, along with our outlook for the second quarter of 2022. The press release containing our first 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. Replay 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 our 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 those non-GAAP financial measures, including reconciliation to the most directly comparable GAAP financial measure were available on the Investor page of the website. Looking ahead, between now and our next earnings call, Teradyne expects to participate in technology or industrial-focused investor conferences hosted by Cowen, Luke Capital, Bank of America, Stifel, and Bernstein. Now 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 new quarter. Sanjay will then offer more details on our quarterly results, along with our guidance for the second quarter. We'll then answer your questions, and this call is scheduled for 1 hour. Mark?

Hello, everyone, and thanks for joining us. Today, I'll summarize our Q1 results, review current market conditions, and provide an update on Q2 and the full year. Sanjay will then take you through the financial details, including our outlook for the second quarter. The first quarter played out as expected from a financial perspective. We delivered results toward the high end of guidance as we cleared a few more supply chain bottlenecks than expected. As outlined in January, we expect revenue growth throughout the year as we build toward 2023 and the transition to 3-nanometer. Consequently, we expect roughly 7% to 9% sequential quarter-over-quarter growth throughout the year. Since January, we've seen incremental strengthening of demand for Semi Test and wireless in the areas of automotive, memory, WiFi 6c and 7. On the other hand, for the first time in 2 years, supply chain challenges are materially impacting our Semi Test shipments. This is reflected in our wider-than-normal revenue guidance range in Q2. Sanjay will describe these details shortly. From a full year market perspective, we estimate the SOC market is trending to the high end of the $4.6 billion to $5 billion range we discussed in January. Strengthening automotive test demand is one element driving the market increase as both per car content and unit complexity are growing quickly. The accelerating move to electric vehicles is a key part of this increase with EVs growing at a 20% plus CAGR while the chip content of these vehicles is running more than three times compared to gas-powered cars. This year, we are seeing particular strength emerging in high-end automotive ADAS processes. The complexity of these processes is approaching the level of high-end cellphone applications processors. Additionally, high-end ADAS processors have about a 3 to 4 times longer test time due to the more stringent test requirements for devices used in automotive safety applications. Furthermore, there are usually multiple ADAS processors per vehicle, amplifying the unit demand. Combining the enhanced EV semiconductor content with these ADAS trends, we expect the automotive test market to outgrow the core test market for the foreseeable future. Another area of growth is in hyperscaler silicon development. While the material test revenue impact from these complex chips is in 2023 and beyond, both the complexity and pace of development is accelerating. This has been a focused investment area for us over the last 18 months, and we are pleased with our win rate for these new opportunities. Shifting to the memory test market. Our market size estimate for the year remains at about $1 billion. However, within that, we have seen the DRAM demand soften a bit, offset by an increase in flash demand. In System Test, defense and aerospace and production board test groups collectively grew over 30% from Q1 '21 on strong automotive board test demand and program buying in defense and aerospace. In Storage Test, sales were down 28% as expected versus Q1 of '21. Customer concentration here causes this lumpiness, but the long-term complexity and unit demand drivers remain in place. Our wireless test business at LitePoint remains very healthy with sales growing 26% year-over-year in Q1. Strong demand for emerging WiFi 6c and WiFi 7 standards is driving this growth and the outlook for the year. LitePoint's strategy delivers high throughput, easy-to-deploy test solutions for the full array of wireless standards. Our chipset test library approach to programming allows customers to shorten their time to market and achieve superior test economics for these emerging standards. Moving to Industrial Automation. Group revenues were up 29% from Q1 of last year. Demand in North America was particularly strong with 55% growth year-over-year. While we cleared some supply issues at UR rate in the quarter, IA shipments overall continue to be constrained by material shortages. Globally, the PMIs for the U.S. and Europe remain over 50, while in China, the PMI dipped below 50 in March. We are closely watching to confirm that China data point is a short-term COVID lockdown issue and anticipate recovery in Q3 and Q4. Similarly, the conflict in Ukraine slowed EU demand in March and has the potential to impact the European industrial activity throughout the year. While we expect these constraints to continue in Q2, we do expect IA growth rates to increase through the year as supply constraints ease, AutoGuide begins to ramp shipments, and UR and MiR see increased growth in China and EU recovery. At UR, our OEM welding application continued to flourish with Q1 growth of over 100% compared to 1 year ago. The U.S. geography was also strong with over 50% year-over-year growth. Our UR+ program of plug-and-play applications for our cobots is passing the 400-product milestone this quarter. Each of these applications leverages an independent third-party developer's knowledge of a specific market vertical to solve a specific problem. This application is then certified by UR so our customers can deploy automation solutions quickly by spending less time on programming and integration work. The resulting network effect is a powerful differentiator for us, and we continue to invest to support this growing army of developers. At MiR, we are seeing two positive trends emerge as we grow the business. First, we are seeing a steady movement toward higher payload, higher ASP AMRs. In 2020, about 10% of our sales were above the 500-kilogram payload range. This moved to 22% in 2021, and in Q1, we saw over 30% of our sales in this class. This represents a broadening of AMR applications to a wider range of material handling tests and manufacturing and logistics. Second, we are seeing the trend toward both growing fleet sizes and increasing multifactory deployments at major customers. Within these environments, our fleet management software is an increasingly important product differentiator. These larger customers also have demand for uptime and support requirements, which also plays to our strength as MiR is leveraging Teradyne's global support and large account experience to meet these customers' needs in ways that smaller players can't. With one quarter of 2022 in the books, the year is unfolding roughly in line with our forecast. The incrementally stronger demand in testing is being balanced by supply constraints. Longer term, I am encouraged by the continued strength in WFE investments, which fuels our growth; the progress from industry leaders on 3-nanometer technology; the growing opportunity for our human scale automation projects and the resilience of Teradyne's global team and partners that are enabling us to muscle through some very tough supply and operational constraints. Before closing, I want to note an important milestone at Teradyne. Mike Bradley will retire from our Board of Directors in May. Mike's Teradyne career spans over 40 years, including 10 years as CEO, helping shape Teradyne's products, business model and culture. Mike has had an incredible impact on our company, our customers, and shareholders, and I'm especially appreciative for the help he's given me along the way. Thanks, Mike. Sanjay will now take you through the financial details. Sanjay?

Thank you, Mark. Good morning, everyone. Today, I'll provide details on our Q1 results, offer additional color on the operating environment, including the supply line update and describe our Q2 outlook. Now to Q1. First quarter sales were $755 million, with non-GAAP EPS of $0.98. Non-GAAP gross margins were 60.2%, and our non-GAAP operating expenses were $248 million, slightly below our guidance due to slower-than-planned hiring. The non-GAAP operating profit rate was 27.4%. We had no 10% customers in the quarter. The tax rate, excluding discrete items for the quarter was 16% on both a GAAP and non-GAAP basis. Please note, you should now use this for the full year tax rate as well. Looking at the results from a business unit perspective, Semi Test revenue of $482 million was down 9% from Q1 '21 as expected. SOC revenue was $387 million driven by strength in mobility and compute end markets. Memory revenue was $96 million, driven by flash final test and DRAM wafer sort. System Test group had revenue of $119 million, which was down 11% year-over-year, again, as expected. Storage Test sales, including HDD and system-level test solutions, were $68 million in the quarter, down from $95 million in Q1 '21. Defense and aerospace and production board test grew year-on-year. At LitePoint, revenue of $52 million was up 26% from the prior year due primarily to strong shipments of WiFi and UWB test systems. Looking at our test portfolio overall. The positive demand environment, Mark noted, reinforces the trend line growth built into our 2024 earnings model. In Semi Test, growing device complexity supported by no transitions and unit growth will provide a long-term tailwind to that business. At LitePoint and in System Test, similar increases in complexity continue to drive growth. A couple of examples. One, LitePoint revenue in 2021 was approximately double the 2017 level. Two, our Storage Test business. Revenue in Storage Test has more than quadrupled from 2017 to 2021. At LitePoint, we expect growth to continue to be driven by more complex wireless standards for connectivity, sophisticated location tracking and security technologies, and cellular applications. In Storage Test, growing chip complexity combined with higher density, hard disk drives are driving long-term growth. Now to Industrial Automation. Industrial Automation revenue of $103 million was up 29% year-over-year. This was ahead of what we planned in our Q1 guidance as we were able to work through several supply line constraints in the quarter at UR. I'll also note that we expect IA revenue to follow the historical pattern and grow as we move through the year. UR sales were $85 million in Q1, up 30% year-over-year with the highest growth in North America. MiR sales were $17 million, up 22% from Q1 '21. As Mark noted, higher payload products with higher ASPs contributed to that growth, and we expect this trend to continue. Overall demand at both UR and MiR remained strong as labor availability and rising costs make the financial case for our human scale automation compelling. However, as noted, we are watching for the impacts of COVID events in China and the Russian-Ukraine war closely. From a financial perspective in IA, the group was breakeven on a non-GAAP operating basis in the first quarter. For the full year, we expect to operate in the 5% to 15% range we discussed in past calls, with 2022 trending towards the low end of the range. Recall in 2021, IA delivered a 4% non-GAAP operating profit for the full year. As noted in the past, with market penetration for cobots and AMRs below 3% and high market growth, our strategy is to prioritize growth over obtaining model profits. In simple terms, we are investing to capture these markets as we see significant growth beyond the midterm. Shifting to supply. While the demand environment is strong, supply issues are getting progressively worse. We continue to manage through a very challenging supply environment as we have for the past 2 years. Our supply management operations teams and partners have done an incredible job, including navigating through a multi-week plant shutdown at one of our major contract manufacturers in Q1. Fortunately, it was mid-quarter, and we had time to recover. Had the shutdown happened later in the quarter, the recovery could have shifted into the following quarter. It highlights the risk that will persist during the global COVID pandemic. In Q2, there is more demand to ship if we could align supply. Issues range from chip shortages to COVID-related delivery shutdowns. We're taking numerous actions to harden our supply chain, including expanding our production operations geographically, adding new geographically diverse suppliers, redesigning products to use available components, and many other actions. However, even with these actions, we expect supply line constraints to remain an issue through at least the first half of 2023. Shifting to the balance sheet and cash flow. Our cash and marketable securities at the end of the quarter totaled $1.2 billion. We had $37 million in negative free cash flow in the quarter. Cash burn in the first quarter is typical for us as we pay variable compensation and profit sharing in the quarter. We spent $201 million and $18 million on buybacks and dividends, respectively, in Q1. As noted in January, we expect to purchase a minimum of $750 million of shares in 2022. Regarding debt, to date, $384 million of convertible bonds have early converted in Q1. We paid $21 million to bondholders in the quarter. Now to our outlook for Q2. As you've heard this morning, customer demand is strong in all parts of the business. Our guidance range is wider than normal as the number of material shortages continues to expand. While we've been successful to date in addressing these issues, the margin for error continues to narrow, and the guidance assumes we won't see extended shutdowns of our production facilities due to COVID. With that said, sales in Q2 are expected to be between $780 million and $870 million, with non-GAAP EPS in a range of $1 to $1.29 on 170 million diluted shares. The second quarter guidance excludes the amortization of acquired intangibles. Second quarter gross margins are estimated at 60% to 61%. OpEx is expected to run at 31% to 34% of second quarter sales. The non-GAAP operating profit rate at the midpoint of the second quarter guidance is 28%. Turning to the impact of inflation on our P&L. First, gross margins. We are seeing increased costs primarily in components and labor, both internally and at our partners. We're mitigating those increases through a variety of measures, including product pricing, operational efficiencies, multi-sourcing, and seeing the benefit of our prior investments to reduce our bill of materials. Our gross margin performance in Q1 and outlook for Q2 reflects our success to date with these measures. But I remind you, this is a very dynamic environment. On the OpEx front, we have planned for increased costs tied to inflation and we still plan on growing OpEx 11% to 13% over 2021, as noted in January. Looking at our revenue profile for the year, the first half is shaping up a little bit better than expected with very tight supply. The second half is looking to be slightly up versus last year's second half, reasonably consistent with our expectations at the beginning of the year, even though test demand is incrementally higher. Supply issues may push some of that incremental demand into 2023. We expect third quarter revenues will grow high single digits from the Q2 level, and Q4 will grow sequentially as well. Our current view is that the first half revenues will be 46% to 47% and second half revenues will be 53% to 54%. To sum up, demand across the company remains strong and growing supply constraints are moderating our shipments more than at any other time since the pandemic began. We're expanding our playbook to address these issues, but it's a very dynamic playing field and we're calling a lot of audibles along the way. Our global teams in go-to-market operations, engineering and support functions continue to execute and collaborate in a challenging and dynamic environment to meet customer demand and deliver shareholder value. The passion is enjoyable to observe. Well done, team.

Thanks, Sanjay. Chris, we'd now like to take some questions. And as a reminder, please limit yourself to one question and a follow-up.

I guess first question, regarding your largest customer heading into 2023. That's a long time from now. But I'm curious, as you think about the move to 3-nanometer, how are you thinking about test demand as it relates to units versus complexity? I guess the worry in the marketplace is that the adoption of 3 could be spread out over 2 years as opposed to 1, and that would reduce the demand for testers from you guys. So I'd love to hear how you're thinking about that. And obviously, the caveat is it's very early to have a great view there.

Yes, C.J., it's really early on that. But first of all, I think there's two parts to your question. Overall, we're looking at 3-nanometer in 2023 as mostly a complexity growth equation versus a unit growth driver in the cellphone device arena. The other part of your question is, well, what if the adoption is partial for 3-nanometer? Certain devices go to 3, and other ones sort of lag. I think that is something we've seen before in iPads and other kinds of products in the market where they bifurcate migration. If and when that ever occurs, there could be a 1-year depressed demand for tests. But then in subsequent years, those bifurcated product lines kind of then move in lockstep again north; maybe one's waterfall back a year, but the complexity growth resumes thereafter. So it's all hypothetical, but something like that would have an impact.

Very helpful. I guess for my second question, just trying to think through your stronger second half versus first half. Obviously, the robotics piece is helping on that. But this is the first time I've seen you guys have a stronger outlook for Semi Test into the back half; normally, Q2 is your strongest quarter seasonally. Is that all a factor of supply constraints, or are there other drivers to that change in trend?

There are three main factors at play. First, Q2 typically sees high activity due to our largest customer experiencing a peak in tooling during that quarter and the next, but that is not happening this time. Second, customer visibility has improved because of extended lead times, leading customers to order in advance for capacity. As a result, we have better insights into demand for the latter part of this year and early next year. This aligns with the growing complexity related to new nodes that we're encountering. This year stands out because we do not have the typical summer surge in mobility, combined with the transition to new nodes next year, resulting in a more gradual increase in demand throughout the year.

I want to go back to your outlook, especially looking into next year. You made a reference to WFE, which is very positive given all the investment this year. But it also appears that the investment for trailing edge is growing much faster, particularly a higher level of investment in China. How are you positioned in China, especially with OSAT houses? Is there any incremental opportunity for share gain? And I have a follow-up.

Yes, I believe we are well positioned in China regarding legacy non-leading edge technology. The only area where we're lacking is with Huawei and its associated companies, which are striving to move towards more leading-edge positions. However, as China continues to grow in trailing edge, our share in the non-Huawei segment of China is approximately the same as it is outside the country. Therefore, it's neither a significant boost nor a burden. I expect we will fluctuate there based on investment and demand.

Okay. So nothing out of the ordinary. If they are indeed investing more, it should trickle down to the back end and you as one of the two. Okay. And then moving on to Industrial Automation. And thanks for the color on ADAS incremental opportunity. It seems to me that in order for you to fully realize the synergies between your Semi Test and Industrial Automation for EV and ADAS applications, you would need to have more software content. I'm just curious if I'm thinking about this the right way and whether you would have to go to the market and acquire software IP so you can have a more comprehensive portfolio for these growth end markets?

No, I don't see that, Mehdi. I think where we are positioned now, we're completely aligned to the technology trends in EV and ADAS and everything else there. So no, I don't see the need for that.

So Mark, you were saying that the SOC TAM is trending toward the high end of the $4.6 billion to $5 billion. I'm wondering if you can give us some idea of what end markets are driving that high end. I think last quarter, you gave us some breakdown. Compute, I think you said would be $1.3 billion. Mobility would be about $1.8 billion. Autos would be about $500 million. Can you give us sort of what end market is driving that toward the high end? It sounds like autos might be a little better.

Sure. It's Sanjay here. I'll take that one. We are seeing incremental strength in compute and in auto that's driving us to the high end.

Okay. And then I guess a question just around China. Are you seeing any impact or any pushouts or delays in shipment timelines from Chinese OSAT customers as a result of the lockdowns and maybe just as a result, also of weak demand in the consumer segment? I mean, obviously, demand is still very strong if you look holistically. But if you look at the consumer markets and you look specifically in China, are you seeing any examples of customers delaying shipments of testers?

No, we're not seeing any meaningful delays at all yet in China. We have seen in Industrial Automation business, the robotics business, some slowdown in China. Sanjay referred to the Q1 performance there as a bit weak, and we're watching that. So that's more immediate. And as factories shut down, we feel that. But I think on the test side, the headlights and the investments needed along the trend line are kind of obvious and compelling, and there's no blips yet.

I guess the first one focused on test. Given the supply challenges that you're discussing, do you think you're baking in enough conservatism on the '22 TAM outlook? And I ask in part because I think despite strong order rates, your competitor Advantest widened their 2022 TAM outlook, but at the lower, not higher end of the range. So I'm curious about that. And also, beyond your largest customer, have you seen any incremental signs of capacity digestion in the broader SOC - mobile SOC market?

Just a quick comment from me first, and I'll let Sanjay talk a little more about it. Compared to Advantest, we're in a very different supply chain situation. You look at their backlog out at almost a year. They have a very limited ability to see a road to incremental revenue this year, whereas we're looking at months and less versus 1 year. We do have capacity growth capability in the back half. So that's the difference, but I'll let Sanjay address the broader question.

Yes, I think overall, supply is getting tighter. As I've said on prior calls, I thought with substrate and fab capacity coming online in 2022, we'd be a little bit more in balance with supply and demand, but that hasn't been the case. Our current view is that we see tightening continuing into the first half of 2023, with a constraining element of supply to meet overall demand in semiconductor parts, FPGA linear logic, memory, and other ASICs.

Got it. And then, as I missed the part, I think I asked about whether you're seeing any signs of digestion in the broader SOC mobile market?

No, not really. I think demand is still strong, and the request for shipments isn't pushing out. So I don't see any signs yet in testing of sort of slacking demand.

Yes. So in terms of Industrial Automation, our guide is still to grow 35% plus this year. It's progressing as we'd expect. However, as Mark noted, China, which has traditionally been a very high-growth environment for IA, was slightly down year-on-year. We're watching it closely.

I do think the China piece is roughly 10% of our overall IA business. So it is flatlined in Q1 on growth. But part of how we get to 35 plus through the backend of the year is assuming some of this COVID stuff goes away in China and Q3 and Q4, we presume it persists through the second quarter.

Mark, you talked about weakness in the DRAM market. Can you provide some more color? Is this because of DDR5 adoption pushed out because of the supply rapid delay? Or are you seeing something more?

No, it's exactly that. DDR5 is moving slower than planned due to the server Sapphire Rapids delay. Nothing fundamentally. The three elements of the memory market between DRAM wafer test final test and then flash wafer test final test are roughly 25% each of the total $1 billion TAM. There is a bit more tooling this year for DRAM wafer test, but that's noise, not signal. As I mentioned, we're seeing some strengthening in the flash side, making up more than the difference on the DDR5 weakness. The other uplifting aspect about memory that's very encouraging for us is our market share being about 40%, low 40s globally. In China, which is still investing heavily and growing quite rapidly in memory, it's moving closer to the high 50s. So we've got good potential here with DDR5 hopefully returning next year.

Great. And as my follow-up, I want to go back to C.J.'s earlier question about mobile compute adoption potentially spread maybe over a couple of years. That phone maker has publicly stated that they are facing physical limitations in creating a larger die to accommodate more transistors at 5-nanometer, which I don't believe was the driver in iPad-type decisions historically. Is it fair to assume that the device maker could adopt 3-nanometer more widely moving forward given die side limitations?

Yes, I think that's more likely. It was a hypothetical question that C.J. was asking. So just to walk it back a little, this year we have seen a slackening in tester demand because the complexity growth is low in phone silicon, highly correlated with the fact that we're in our third year of 5-nanometer production. Yes, you can't afford to grow the die size; it doesn't make economic sense. With that, it's hard to get more transistors unless you do a node transition. There's a backlog and pent-up demand to get to this next turn to the next node of 3-nanometer. So this all depends on the yields of what 3-nanometer will look like. But on the other side, you could see a larger portfolio of devices driving the move to 3-nanometer next year, and people will be competing for fab capacity. We hope we are positioned with the customers who are at the front of the line on that fab capacity. I think we are. But I would subscribe more to the fact that there's pent-up demand, and it could swing the other way.

I had two as well. Mark, you talked about the hyperscale opportunity in your prepared remarks. I think you noted that some of the customer activity seems to be accelerating and that you've been pleased with your run rate. I was hoping you could elaborate on that a little bit and to the extent possible, kind of speak to the timing of a potential inflection in that business. I realize that's sort of a medium- to long-term opportunity for you. But curious to hear any new developments there?

Yes, that's tough to speculate on because there's two classes of hyperscale silicon development occurring. One is around new consumer products. That is incredibly difficult for us to forecast due to unprecedented-type product introductions to see if the market will develop or if they will even be introduced. They’re active, we're highly engaged, they're wildcards. Then there's another class that I would describe as more compute, cloud infrastructure. Chips going into the cloud for specialized use, could be AI applications going into the edge. Those have more precedented applications, and we have long-term plans that I expect to start ramping next year. I think some of these are going to ramp. Now, how material will they be? I think we're still talking from a tester point of view, tens of millions versus hundreds of millions of dollars of incremental revenue. So we're still going to be in early innings, but that gives you some sense.

You guys seem to be executing really well from a gross margin perspective despite everything that's going on. I think you beat the midpoint of your guidance in Q1 by more than 100 basis points. You're guiding to another higher than 60% quarter for Q2 despite everything that's going on. Is this primarily a function of your largest customer being absent, or at least being a small percentage of your revenue than typically driving a better mix? Or is there anything sustainable that could drive margins above your long-term model? Yes, in Q1, it was really a function of product mix and operational efficiencies that drove us beyond what we guided. Our growth in IA and LitePoint has higher than corporate average margins. We do have plans that are challenging those to make them higher, but we are also realistic about some of the headwinds influencing the environment.

In terms of the supply constraints that you have, can you kind of frame that competitively? I feel like last year you had sort of a better supply situation than Advantest did? Is that still the case? And any ramifications for market share in the more transactional parts of the business?

We're not inside Advantest to know a lot. But from the outside, their backlog is close to a year. Their customers are forced to order in advance, around a year. Apparently, whatever is going on in terms of supply chain strategy has been slightly more favorable on our side. We've worked this since 2019 as part of a business continuity planning exercise around a whole series of things that could have gone wrong in the world. We never anticipated COVID but generally built up inventory resilience and backups that have fared well for us for 2 years. So we've been hoping to get through all of this without any significant bottlenecks. We're seeing our first headwind here in testing at about a $40 million bottleneck in Q2, which is roughly 5% of our revenue. We anticipate being able to maintain revenue growth of that amount. Maybe that's going to be about the percentage through the year that we're going to be bottlenecked. We hope that we can keep it at that and keep lead times in sort of a 6-month range in testing.

I have just one question. Regarding your largest customer, are you still expecting them to account for 10% of revenue this year? Typically, you have good visibility into their performance around this time in the April to June period. Is there a lot of variability expected for the remainder of the year?

Yes. Sidney, our expectation is that we will not have that large customer be over 10%, consistent with what we thought in January.

For 2Q and 3Q, no, it's pretty locked in. Q4 is, frankly, a wildcard for us because it depends on factors such as the ramp in 2022, the early silicon prove-outs, and launching products. So fourth quarter is still opaque to us. If there's a potential for something in fourth quarter that we might not see yet, yes. But we'll be somewhat constrained there too based on the earlier supply discussions.

Well, for the first time in over 40 quarters, we're finished early. Thanks, everybody, for joining us. Look forward to talking to you in the days and weeks ahead. Bye-bye.