Good morning, ladies and gentlemen, and welcome to the Q2 2020 NXP Semiconductors earnings conference call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session, and instructions will follow at that time. If anyone should require assistance during the conference, please press star then zero on your touch-tone telephone. As a reminder, this conference call is being recorded. I would now like to turn the conference over to Mr. Jeff Palmer. Thank you. Please go ahead.
Thank you, Jerome, and good morning, good afternoon, everyone. We hope you're all safe and healthy. Welcome to the NXP Semiconductors second quarter 2020 earnings call. With me on the call today is Kurt Sievers, NXP's CEO and President, and Peter Kelly, our CFO. The call today is being recorded and will be available for replay from our corporate website. The call will include forward-looking statements that involve risks and uncertainties that could cause NXP's results to differ materially from management's current expectations. These risks and uncertainties include, but are not limited to, statements regarding the continued impact of the COVID-19 pandemic on our business, the macroeconomic impact on the specific end markets in which we operate, the sale of new and existing products, and our expectations for the financial results for the third quarter of 2020.
Please be reminded that NXP undertakes no obligation to revise or update publicly any forward-looking statements. For full disclosure on forward-looking statements, please refer to our press release. Additionally, we will refer to non-GAAP financial measures, which are driven primarily by discrete events that management does not consider to be directly related to NXP's underlying core operating performance. Pursuant to Regulation G, NXP has provided the reconciliations of the non-GAAP financial measures to the most directly comparable GAAP measures in our second quarter 2020 earnings press release, which will be furnished to the SEC on Form 8-K and is available on NXP's website in the investor relations section at nxp.com. I'd now like to turn the call over to Kurt.
Thanks very much, Jeff, good morning or good afternoon, everyone. We really appreciate you joining the call today. Let me turn to our results. Our Q2 revenue was modestly better than the midpoint of our original guidance. Our automotive business was significantly impacted by the COVID-19-caused factory shutdowns of our customers. We did experience better than anticipated trends and sequential growth in all of our other end markets. We are encouraged by the positive trends we experienced in China, the sales out of our distribution channel did improve sequentially. Taken together, NXP delivered revenue of $1.82 billion, $17 million above the midpoint of our original guidance range. Non-GAAP operating margin was 20.7%, about 170 basis points above the midpoint of guidance. We experienced slightly higher revenue with a higher proportion of distribution channel sales.
We did deliver better gross margin because of a positive product sales mix, all of this combined with tight control of our operating expenses, resulted in better than expected operating profitability. Turning to the specific trends in our focused end markets. In automotive, revenue was $674 million, down 35% versus the year-ago period, and showing a 32% sequential decline. In industrial and IoT, revenue was $435 million, up 12% versus the year-ago period and up 16% sequentially. In mobile, revenue was $255 million, down 14% versus the year-ago period, up 3% sequentially. Please note, the year-on-year comparison in mobile was impacted by the sale of the voice and audio business. Lastly, communication infrastructure and other revenue was $453 million, down 9% year-on-year and up 12% sequentially.
Now, before I'm turning to the specifics of our Q3 expectations, I'd like to make a few important comments. Our revenue guidance range for Q3 is again wider than normal. However, we believe the setup is gradually more positive heading into the second half of the year. This is thanks to customer traction with NXP specific drivers, including automotive radar, wireless connectivity, our crossover processors, our secure ultra-widebands, just to name a few. Furthermore, we do see an ongoing stabilization of our end markets. However, at the same time, we want to balance our enthusiasm as we continue to view the broader demand environment as fluid given the COVID-19 pandemic. It is too early to make a broad statement regarding a complete return to normalized demand or the specifics of the shape of the recovery.
As an example, several of the ultimate end customers of our products, especially the automotive OEM customers in Europe, North America, and Japan, are still running production at below pre-pandemic levels. Therefore, we view the best course of action is to continue to focus on those aspects of our business which we can directly control. This certainly includes stringent discipline of our distributor channel inventory to maintain our target channel inventory at 2.4 months of supply. Exactly in that light, we held back about $145 million of shipments to distributors during the past quarter. Additionally, we continued to run our internal factories significantly below normal operating levels, avoiding building excess inventory. With that preamble, we are guiding Q3 revenue at $2 billion, down about 12% versus Q3 2019. From a sequential perspective, this represents an increase of about 10% at the midpoint versus the prior quarter.
At the midpoint, we anticipate the following trends in our businesses. Automotive is expected to be down in the low 20% range versus Q3 2019 and up about 20% versus Q2 2020. Industrial IoT is expected to be up in the mid-teens range versus Q3 2019 and is expected to be up in the mid-teens range versus Q2 2020. Mobile is expected to be down in the low teens range versus Q3 2019. Again, the year-on-year trends are being impacted by the sale of the voice and audio business. On a sequential basis, mobile is expected to be up about 10% versus Q2 2020. Finally, communication infrastructure and other is expected to be down in the low teens range versus Q3 2019 and down in the upper single-digits range versus Q2 2020. Now let me summarize.
We will be laser focused on what we can control, and we will continue to navigate an uncertain demand environment. Clearly, our number one priority is to assure the health and safety of all of our NXP team members, while at the same time facilitating the best possible business continuity with a customer focus on supply chain and R&D execution. We don't have any unique insights as to when this challenging period will subside, but we continue to have ample financial liquidity and strength to weather the current environment. We maintain all the critical investments in areas that will assure NXP's long-term success in its chosen strategy, while we actively and continuously review all areas of discretionary spending.
Our focused investments in leading-edge new products and deep customer engagements in fast-growing segments such as automotive ADAS and electrification, secure connected edge processing for the IoT, and our secure ultra-wideband are all very, very durable, we do continue to enjoy significant design win traction. We are committed to the consistent execution of our long-term strategy and continue to be deeply engaged and sharply focused on enabling our customers' success. Now I would like to pass the call to Peter for a review of our financial performance before we turn to all of your questions. Peter?
Thanks, Kurt. Good morning to everyone on today's call. As Kurt has already covered the drivers of the revenue during the quarter and provided our revenue outlook for the third quarter, I'll move to the financial highlights. In summary, our second quarter revenue performance in total was a little better than planned. Our industrial and IoT end market, along with common infrastructure, showed significant strength. Our shipments into the mobile end market were about where we planned, and our automotive revenue was significantly weaker than planned, as OEMs and Tier 1 suppliers in Europe, North America, and Japan closed their factory for extended periods. You'll note our proportion of sales through distribution versus direct was significantly higher, reaching a record 58% given the strength in China and the weakness in automotive, which is more of a direct business end market. Moving to the details of the second quarter.
Total revenue was $1.82 billion, down 18% year-over-year and $17 million above the midpoint of our guidance. We generated $892 million in non-GAAP gross profit and reported a non-GAAP gross margin of 49.1%, down 420 basis points year-over-year and 110 basis points above the midpoint of guidance. Gross margins were better than expected because of the mix swing towards distribution and an overall richer product mix. Total non-GAAP operating expenses were $516 million, down $25 million year-over-year, and better by $29 million from the first quarter. This was $7 million better than the midpoint of our guidance because of lower payroll expense. From a total operating profit perspective, non-GAAP operating profit was $376 million and non-GAAP operating margin was 20.7%, down about 820 basis points year-over-year, but 170 basis points higher than guidance due to better gross margin and lower operating expense.
Non-GAAP financial expense was $92 million, which was $10 million higher than guidance because of the new $2 billion debt issuance we undertook during the quarter. Cash taxes for ongoing operations were $16 million and non-controlling interests were $5 million, slightly better on a combined basis than our guidance. Stock-based compensation, which is not included in our non-GAAP earnings, was $105 million. Now I'd like to turn to the changes in our cash and debt. Our total debt at the end of the second quarter was $9.35 billion, up about $2 billion sequentially, and our ending cash position was $3.27 billion, up $2.2 billion because of the debt issuance and cash generation during the quarter. Net debt was slightly better at $6.09 billion, and we exited the quarter with a trailing 12-month a of $2.8 billion.
Our ratio of net debt to trailing 12-month a at the end of the second quarter was 2.2 x, and our non-GAAP trailing 12-month a net interest coverage was 8.5 x. We continue to have a strong balance sheet and excellent liquidity. The response to the recent debt issuance was truly phenomenal. The offering was structured in three tranches of a $500 million five-year note, a $500 million seven-year note, and a $1 billion 10-year green bond. The offering was 11 times oversubscribed, and we were very excited that we were one of a very small number of tech companies to have successfully made a green offering. During the second quarter, we paid $105 million in cash dividends, and as we noted last quarter, until our leverage returns to our two times target, we have temporarily suspended our buybacks, though we will maintain our quarterly dividend.
Turning to working capital metrics, days of inventory was 120 days, an increase of seven days sequentially as revenue levels declined. Though on a dollar basis, inventory was flat sequentially. We continue to closely manage our distribution channel with inventory in the channel at 2.4 months, well within our long-term targets, and we held back about $145 million of orders into distribution to ensure our channel inventory metrics remained within our target range. I'm very proud at how the team has managed both owned and channel inventory. Days receivable were 24 days, down four days sequentially. Days payable were 71 days, a decrease of 12 days versus the prior quarter. Taken together, our cash conversion cycle was 73 days, an increase of 15 days versus the prior quarter.
Cash flow from operations was $414 million, net CapEx was $74 million, resulting in non-GAAP free cash flow of $340 million. A testament to the strong cash flow generating capability of the business, even in a challenging period. Turning to our expectations for the third quarter, as Kurt mentioned, we anticipate Q3 revenue to be about $2 billion, ± $100 million. A wider range than normal, considering the uncertain environment we are navigating. At the midpoint, this is down about 12% year-on-year, while up 10% sequentially. We expect non-GAAP gross margin to be about 49%, ± 100 basis points. Operating expenses are expected to be about $535 million, ± $10 million. Taken together, we see non-GAAP operating margin to be about 22%, ± 180 basis points.
We estimate non-GAAP financial expense to be about $98 million and anticipate cash tax related to ongoing operations to be about $34 million. Non-controlling interest will be about $3 million. I have a few closing comments I'd like to make. Our gross margin guidance for Q3 remains flat on Q2. With revenue up 10% sequentially, we see some benefit from the additional volume, but this is offset by our product mix, which will be less robust than in Q3 as compared to Q2. Secondly, the continued effects of a very low factory utilization as we manage our inventory levels. As the global economy starts to correct itself and our revenue re-accelerates, we see no reason why we cannot hit our 55% gross margin target at the $2.4 billion of quarterly revenue level.
In terms of OpEx, the guidance of 535 is not a new normal, but it's a constrained number reflecting the stringent expense controls we've imposed on the organization. The actions taken include the elimination of annual merit increases and incentives and effective hiring freeze, including replacements, as well as salary cuts for executives. Although these are the right things to do in the short term, they're not sustainable in the medium term, and you should assume a more normal level of OpEx in 2021 to be about $575 million a quarter, depending on seasonal influences. When our revenue returns to a more normal level, we would expect our operating expenses to reflect a long-term model of 16% R&D and 7% SG&A. Lastly, we are proactively driving down our internal inventory levels.
Our long-term target is 95 days and aim to achieve about 100 day level exiting the third quarter. This will clearly impact our factory utilization. Finally, these are very difficult times. Kurt and I would like to thank all of our colleagues around the world for their commitment to NXP and for doing the right thing for our customers. The current period is unprecedented. It is extremely difficult, but over the long run, NXP has the right strategy, is in the right markets, and has the right products to continue to win. Now I'd like to turn to our questions. Operator? Hello?
Ladies and gentlemen, if you have a question at this time, please press star, then the number one key on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Please wait for a caller's queue for their question. Your first question comes from the line of John Pitzer with Credit Suisse. You may now ask your question.
Yeah, guys. Thanks for letting me ask the questions. Congratulations on the solid results given the challenging environment. Kurt, I guess as we look at the company, as you mentioned in your preamble, there's a lot of company-specific drivers, and I guess I wanted to try to get a better understanding. When you look at the auto growth you expect in the calendar third quarter, to what extent is that coming from the growth areas of the business versus the more mature parts of the auto business? A similar question on the industrial IoT. Great sequential growth in the June quarter. You're guiding for that to sustain into the September quarter. I'm just kind of curious, to what extent is this the benefit you're getting of taking that Marvell asset and running it through your stronger distribution channel?
Thanks, John. Good morning, first of all. A lot of good questions. Let me maybe start with saying, because it really holds both for auto and Industrial IoT, that indeed it all looks like that Q2 was the trough, and now we are moving up from here. The moving up is indeed a mix of company-specific growth, in both segments, by the way, in auto and in Industrial IoT, and obviously also a recovery of the market. In Industrial and IoT, very clearly, and I think we did also a press release on this, the Wi-Fi 6 portfolio from Marvell, which we launched in April, is actually helping. We have also now successfully integrated the Wi-Fi portfolio into our microcontroller and applications processor software development kit, which makes it really easy for customers. Yes, we do see early traction from that combination.
It is a factor in the continued very nice growth in industrial IoT. Clearly, industrial IoT, John, has another strong factor, which is China. The good growth in Q2 in industrial and IoT has been carried largely from a Chinese footprint perspective, and that also continues into Q3. In auto, things are a little bit different. In auto, clearly Q2 was completely abnormal since we had these factory shutdowns, specifically from the OEMs in Europe and the U.S. As we explained in the last call, we really wanted to make sure that we wouldn't create too much excess inventory at our customers, not only at distribution, but also at the direct customers. That's actually the reason why the business has decreased more than we would have anticipated, but we just didn't want to follow any excess inventory building.
Now this is nicely returning. We see pretty good momentum in auto actually into Q3. It is indeed a mix between OEMs and Tier 1s getting back to production. We assume that by the end of Q3, production levels across the world in automotive should be back to 80% of the pre-pandemic period. That's a pretty solid return through Q3. It's also that radar and our cluster business just gets steam again with new design wins which are taking traction.
That's helpful. As a quick follow-up for Peter, just on the gross margin line. You're guiding for good sequential growth in the calendar third quarter, but you also mentioned that you're trying to keep a cap on inventory build. I'm kind of curious as to what that means for utilization Q2 into Q3. Just remind us, most of what you make or ship this quarter, you made last quarter. Is the utilization impact that if you get a benefit in Q3, is that really going to show up in Q3 gross margin? Or how do we think about Q4 gross margin?
Okay. Several questions there. First of all, utilization in Q3 is about, on average, about 400 basis points lower than Q2. I'm trying to take about, to go from 120 to 100 days, I think it's about, I don't know, $70 million of inventory out of the system. One of the reasons utilization is down is we're shipping from inventory to keep it under control. Utilization is running about, I think about 50%, maybe a little less. We do have this rule, accounting rule that when it's a little bit complicated because we do it by factory and it's six months trailing utilization. When utilization is below 70%, we accelerate the fixed cost write-down. Both Q2 and Q3 suffer from an accelerated fixed cost write-down in that you don't carry it forward through inventory.
On the other hand, assuming Q4 is okay and utilization starts to come up a little bit, even if it doesn't get to the 70% level, which it probably won't. We'd expect to see some small benefit. As time goes on, we'll be able to put more and more of the fixed costs into inventory. Typically, utilization in the current quarter impacts the next quarter. In the current environment, because utilization is so low, you take a hit. Not all of it, but you take most of it in the current quarter.
Your next question comes from the line of Vivek Arya with Bank of America Securities. You may now ask your question.
Thanks for taking my question. Kurt, just one follow-up on the autos segment. I think you mentioned at the end of Q3, production levels will be back to 80%, I believe you said, of their normal trend. When I look at your automotive sales at the $800 million guidance, I think that'll be up to 70% of its prior peak. I understand these things are not always coincident, but I'm curious, how do you look at the unit and the content recovery from here? Because this year, we all understand it's tough. When I look at some of the IHS forecasts for next year, they are looking at auto units perhaps being up double- digit, and I know the visibility is low.
If they are up double- digits, what does that conceptually say about your autos business, given the trend that you have seen so far play out this year?
Hi, Vivek. Let me first of all comment to the current quarter, I would say, the past two quarters and the next quarter. That is actually all pretty much in check. According to IHS, the car production in Q1 was down year-on-year 22%. Our business was only down 4%. In Q2, car production, according to IHS, was down year-on-year 45%. We were down, as just announced, 35%. We've been doing really a lot better in these first two quarters. I don't claim this is all market share gains or something, but part of this is indeed some inventory they've been building in the first half, like always, which will come down in the second half, but actually not too much, which is why I think latest mid through end Q3, we should be totally in balance again.
Such that the 80% car production at the end of Q3 is, I think that has a reasonable fit with our revenues. If you think about the fact that in Q1 and Q2, we've actually grown well ahead of the car production. A little bit more bigger picture, Vivek. First of all, yes, I definitely believe the algorithm which we've spoken about that the semiconductor auto market should be like 3%-4% ahead of SAAR. We think that absolutely holds also through this pandemic and out of the pandemic. Our target to outgrow that by one and a half times also stands. I don't know what the car production next year is exactly going to do. IHS actually, I think they have something like 13% or 14% growth.
Yeah, with the algorithm, Vivek, we should be then nicely growing ahead of this in our business. It doesn't work by the quarter, but over a year or two, it absolutely stands. I also haven't really seen a lot of massive platform delays or something. I believe the new wins which we have in our growth areas are all intact and also come on time. I have, say, a relatively solid portion of optimism when I think about automotive for the second half of this year, but then certainly going into next year.
Got it. Very helpful. Kurt, for my follow-up. Comms infrastructure is interestingly now your second largest business after autos. I imagine one of your more profitable or perhaps the most profitable business, which is why I think, Peter, you mentioned about that mix effect in going into Q3. I'm curious, how do you think about the growth prospects and leverage to 5G? When I recall back to the Analyst Day, I think this was supposed to be a segment with kind of more modest growth prospects versus others. How do you think about the leverage to 5G? When will you start to see those benefits? We are seeing very strong global deployment of 5G. Just talk to us about comms infrastructure, just because it's now such a large segment for you and obviously an important contributor to your gross margins. Thank you.
It's certainly a large and important segment, but the relative size to the others, obviously in this abnormal period, is really shifting every quarter. Q2 certainly wasn't a normal quarter from a revenue perspective. Anyhow, going forward, clearly, the 5G deployment is a key factor. Not everything is shiny, Vivek, to be fair. The one thing is that clearly, there is one large Chinese company, which is a big carrier of the Chinese deployments. Nobody really knows what the export control regulations will do relative to this customer. That's at least one factor on the horizon where we always have to be a bit, say, cautious instead of being overly optimistic. The other one is that more recently.
We are working on our gallium nitride product, as you know, which has higher output powers, which is actually very attractive across the customer base. The only issue is that we are late relative to the demand. I wished we had a faster expansion of our capacity. We have a little bit of a delay here against the demand, which causes us, I say, some delay against the opportunity. Overall, I'm absolutely with you. We have a very strong position here across LDMOS, all the way through gallium nitride to silicon germanium in that space. The market is bumpy. It has always been bumpy. It remains bumpy for the factors I've just quoted.
Your next question comes from the line of Stacy Rasgon with Bernstein Research. You may now ask your question.
Hi, guys. Thanks for taking my question. I wanted to ask about the commentary in your release where it said expectation for improved sales trends through the second half of the year. I want to verify, does that actually imply that you do see Q4 growing sequentially off of Q3? If so, can you give us some feeling for what end markets might be driving that?
Well, typically, Stacy, good morning. Clearly, we only guide the next quarter, and that next quarter is a 10% sequential growth. Yes, I made that statement. What I would say is, we see no reason why Q4 should not be growing over Q3. The one I would call out, which is probably pulling this the most, is going to be automotive.
Got it. Thank you. For my follow-up, I wanted to ask about the industrial strength. A lot of companies have been seeing relatively strong traction in this market in the wake of the pandemic, but there's also been some concerns around potential customer overbuilds and pull forward. I know you've been trying to be cautious, not just with your disti channel, but also with some of your customers. Maybe that was more in auto, but maybe also in industrial to try to reduce their own demand forecasts to try to control that. I guess, what are you doing along those lines, and how confident are you that the industrial upside that we're seeing right now actually is sustainable versus just being pulled forward?
Yeah. Thanks, Stacy. That's a good question indeed, because I believe, especially in the current environment, this is a very important part of the controls which we can execute on our business. The distribution inventory is indeed the most relevant factor in our industrial and IoT business because a large portion of that business is actually going through distribution. As we've spoken about a lot of times, we remain super disciplined on the two and a half months. Actually, we had a score again of 2.4 months of inventory in the last quarter. I think I said in my prepared remarks that we could have shipped 145 million more in this last quarter. A solid part of that would probably have been in industrial and IoT. This is our way to make sure that we don't overship into this market.
To anything we can see, which I know I say through the distribution controls on the inventory, which we have in hand, we have a pretty good handle on this. There isn't that much direct business in industrial for us.
Your next question comes from the line of Craig Hettenbach with Morgan Stanley. You may now ask your question.
Yes. Thank you. Question for Kurt, just on ultra-wideband. Can you talk about the breadth of design activity and how you see kind of your exposure the next couple of quarters in autos versus smartphones?
Yeah. Thanks, Craig. I'm glad you are asking. I definitely believe we are very much on track with building the ecosystem in ultra-wideband across mobile and auto. I think we had kind of signposted now for a year already, that this year in 2020, we would get started in a more material way in mobile. That is indeed one of the factors which is driving our sequential growth in mobile into Q3. Let me say that much. It's a flagship product in an Android space company, which is ramping with our ultra-wideband. From a revenue and unit perspective, clearly mobile is now outpacing automotive to start with, but it's been planned that way because we first have to roll out the mobile ecosystem, and then the mobile secure car access application is going to follow in the next year.
We are shipping already a very small amount of ultra-wideband into automotive today, but it's still coming in the more conventional or traditional form factor of a key fob. It's a more secure way of doing car access. With Android now going out in a larger scale in mobile, and we'll see this also with auto OEMs next year in car access.
Got it. Thanks. Just as a follow-up on industrial, I know the Marvell connectivity business had a lot of industrial exposure. Can you just talk to some of the strength you're seeing in industrial? How much of that is perhaps the legacy NXP business versus the impact of just Marvell starting to kind of ramp with you?
If Peter or Jeff, I think you can dissect, at least in a very rough way, the Marvell from our original industrial business. Let me first of all say, Craig, it is really the combination which makes the difference, because we do leverage our leading position in apps, processors, and MCUs to pull through the connectivity. It becomes more and more difficult actually to talk about the two things in a separate way because they are just getting more and more combined going forward.
Yeah. Kurt, I guess I'll take that. Craig, as you know, we're not going to break out Marvell individually every quarter, but I will say in Q2, the Marvell Wi-Fi business was a not quite double-digit percentage of the overall industrial business, but it was a very high kind of single, low double-digit percentage of the overall industrial and IoT business, just within that one end market. We're not going to break out Marvell in aggregate.
Your next question comes from the line of Ross Famer with DB. You may now ask your question.
Thanks, guys. Let me ask the question. Wanted to ask the first one on the inventory side of things, and this might go into the spirit of leaving no good deed unpunished. I just wanted to think about how you look at inventory strategically and how your customers are considering it. The real question is, some of your competitors are seemingly going the exact opposite way of you, running much higher inventory, making sure channel inventory is ready for a rebound, et cetera. I guess, what are we going to have to see to get you to let that either $150 million in the first quarter or $145 million in the second quarter in bookings actually flow through?
I guess, related to that, internally, once you hit that 100 days, is that when the utilization will start to creep up and your inventory burn internally will stop?
Hey, Ross. Good morning. Let me take the more strategic perspective first. I think competitors who you are quoting, who would tell you the exact opposite, have a different business model. This is more about catalog product companies, where the majority of our product is really application and/or customer deal specific. Which means in our case, we just don't need to hold more inventory because we have a fairly good visibility into the specific applications with our customers and the associated run rates. We continue to believe that the two and a half months of inventory and distribution is about the right number for us. We are very sure we will not miss a beat or not miss any uptick by following that strategy. Now, Peter, maybe over to you relative to the hundreds or better 95 days in future.
Yeah. Just to add a little bit to that, Ross, we'll ship the 150 when the distributors start shipping their product to their end customers. At the moment, the reason we don't ship 150 is they order more than they can ship, and we're not willing to let them have more than two and a half months because we have pretty sophisticated models by distri, by type of product. In terms of the internal inventory, our goal is to go down to 95 days. Utilization will start to increase when revenue starts to increase. On flat revenue, flat inventory, you'd see utilization be relatively flat. Flat inventory, increasing revenue, you'd see utilization start to increase. It's pretty straightforward. It's all about the more we ship, the better the utilization gets.
Okay, thanks for that. I guess as my follow-up on the OpEx side, I know there's specific math we can do on the $575 and what percentage of revenues that would be. To the extent you have that framework, Peter, about the gross margin being 55% at $2.4 billion, your commentary about the OpEx getting up to kind of a $575 level on a quarterly basis, any sort of framework about the relationship on that to your visibility on revenues, your assumptions, et cetera, and how much you might turn that up or down relative to how the revenue trajectory improves or doesn't at that time?
Yeah, that's a great question there, Ross. One of the things that Kurt was saying before is we think Q2 was the weakest revenue number for 2020. We think the second half will be stronger than the first half. We think 2021 will be better than 2020. Within that context, we think we have the right products, we're making the right investments, and we're going to win in our markets. We have no plans to reduce our investment. This year we've taken some short-term actions that I mentioned, like basically eliminating pay rises and incentive payments and cutting executive salaries. At least the first two of those are not maintainable as you go into next year, so it will add to our cost.
Our assumption is maybe not completely in the first and second quarter, but we can afford this level of OpEx because of what we think the revenue will do. If there was a huge second wave or something, then maybe we'd change our view on that. Although we didn't see a V-shaped recovery from Q2 into Q3, we don't think 2021 is a complete write-off. Is that helpful? Does that give you the context?
Your next question comes from the line of CJ Muse with Evercore. You may now ask your question.
Yeah. Good morning. Good afternoon. Thank you for taking my question. I guess a follow-up question on the gross margin side. You're effectively telling us that between $2 billion-$2.4 billion, you should see 85% incremental gross margin. Curious, should that be ratable as you progress up?
No. We don't say that. What we've said is you have a model that we think is reasonable of a 5% in change in revenue gives you about 200 basis points of margin. Sorry, gives you 100 basis points of margin. That's at the kind of plus or minus the $2.3 billion level. There's a couple of other things that go on. Our fixed costs at that level are about 35%. If you do the math, that's about, I don't know, $400 million, maybe a little bit more, a quarter. At a much lower level of revenue, your fixed costs are higher, so you do get some higher fall throughs at very low levels, both directions. You have to be able to do the calculation, and it's certainly not ratable.
Okay. Would you expect it to be concurrent with utilization increase, or would it be one quarter delayed?
Well, it's a great question. Up to 70%, until utilization is 70%, any improvement in utilization won't benefit us really in the current quarter. You see a big benefit once you go above 70% because you start to carry some of the fixed costs forward. It's done by factory, so it's kind of hard to give you a real line when we're operating the way we are at the moment.
Okay, that's helpful. As my follow-up, in your prepared remarks, you talked about being encouraged by some of your wins, radar, wireless, crossover processors, secure UWB. If you had to really highlight what is driving your conviction on growth in the back half, what would it be within that construct? Clearly, with the addition of macro improvements as well.
Yeah, I'd say if I had to pick two for the second half, then it's probably the automotive radar and the wireless connectivity.
Great. Thank you.
Your next question comes from the line of William Stein with SunTrust. You may now ask your question.
Great. Thanks for taking my questions. Good morning, everyone. First, Kurt, there's a couple areas in automotive that you haven't highlighted that one of which I think is already ramping, and another one is more of a future. I think we'd love to hear an update on the battery management systems progress and also the S32G network processor. Any design win traction to talk of there? I do have a follow-up if I can.
Yeah. Thanks, Will. Absolutely. By the way, by not mentioning them doesn't mean that they don't do what they should do. The battery management is actually very nicely on track, especially this year. In the meantime, I think it's fair to say it all looks like that the pandemic, as unfortunate as it is, but it seems to further push the share of electrification. Just from a run rate perspective, from a new model launch perspective, it all looks like that electric drivetrains are benefiting from the pandemic over the next period of time. Given our very strong position with the leading European OEM, as you know, and I would say very nicely growing position with a lot of different customers especially in China, we are enjoying that very much. I'd say very much on track.
I was actually surprised you didn't ask one question in that context, which is the combination of ADI and Maxim as a competitor. Let me just mention it here because it all looks like that our superior value proposition of a system approach, including the micro, is again, something which isn't matched by that deal as it looks to us. Very nicely on track. The other one you mentioned is the S32G. That's the 16nm FinFET based gateway processor. On track, launching, and ramping in production next year. I just have to hold your horses a little bit. Very likely in the next maybe three or four months, we're going to come out with a press release with a pretty prominent customer, which is going to give you then further evidence there and in which model and with which volume. This solution is going to ramp.
It's a much broader basis, but there is one very prominent one, which I hope we can actually announce in the coming period.
Great. I appreciate that. If I can get a follow-up, perhaps with Peter. A lot of companies raised capital to improve liquidity, in sort of the March, April timeframe, perhaps as NXP did. Some of them have since sort of reverted back and repaid some debt to perhaps consider that maybe that level of liquidity is no longer needed. Is NXP contemplating this, or are you planning to run at the elevated liquidity level for a while? Thank you.
To be honest, irrespective of the COVID-19 crisis, I would've gone out and taken that debt anyway because it's to pay down the 2021, it's about $1.4 billion note. In the coming months when we think it's the right time, we'll pay it down. It was always planned to use it for that and a little bit of the 2022. That's what we'll use it for.
Got it. Thank you.
Your next question comes from the line of Toshiya Hari with Goldman Sachs. You may now ask your question.
Good morning, and thanks very much for taking the question. I wanted to ask on the mobile business. Obviously, you guys are operating in a fairly challenging environment with smartphone units declining strong double digits. You do seem to be outperforming the market. I guess if you can speak to what you're seeing from a mobile wallet adoption perspective, that would be very helpful. If you can speak to your opportunity set as it relates to ultra-wideband going into the second half and more importantly into 2021, that would be helpful as well. I've got a quick follow-up. Thank you.
Yeah. Thanks for the question. Great question. Indeed, we also see and believe we are outgrowing, and indeed, this is very much about content and attachment rates rather than mobile unit run rates. The two key drivers for growth going into the third quarter and the second half in mobile is indeed ultra-wideband, as I briefly mentioned before. It's also inside the mobile wallet, with one of our very leading customers in that space, we have actually a change of the system architecture. That change impacts the silicon dollar content, which we are shipping into the solution. It's actually nicely growing with that change. The sequential growth which you are seeing is the new addition of ultra-wideband, and it is a higher content of silicon in the mobile wallet. Thirdly, it is the attach rate of mobile wallets per se, which is going up.
That was the other part of your question. The mobile wallet attach rates are on track to the 50% mark in the next year. Maybe a little bit more anecdotally at this point, but it feels to us that the pandemic is giving also a boost to contactless payments in those countries on the globe where there has been a much lower attachment rate so far because people haven't really accepted it yet. It is actually another one which seems to be, and it's a bit early to make that call too firm, but they seem to be also helped by the pandemic. Three drivers in mobile, higher silicon content in the solution for the mobile wallet, the attach rate of the wallet going up, and thirdly, ultra-wideband.
Great. As my follow-up, I wanted to ask on the competitive landscape. Kurt, I think you guys have in the past talked about NXP potentially being a beneficiary of the U.S.-China trade tensions. I realize share gains take a long time, probably take years in your business, from a customer engagement perspective, are you seeing any change for the better for NXP? Thank you.
Yeah, it keeps being a door opener very clearly. We are seen as a European company. We are a European company in doing business with China. That is definitely a positive lever into engagements. As you said, even in fast-moving markets, design wins take time, but it is clearly a positive for us. Yes.
Your next question comes from the line of Blayne Curtis with Barclays. You may now ask your question.
Hey, guys. This is Tom O'Malley. I'm for Blayne Curtis. I just wanted to ask quickly on competition in UWB. Apple has their own in Android. Do you see anyone entering the market or any increasing competition there, or is this really just a discussion of attach rates going forward?
Hey, Tom. Great question. I think ultra-wideband across ecosystems is very much of a deep system play. It is not about this one RFIC. It is actually about the combination of the RFIC, the secure element. In all the applications which we are starting to ship now, it always comes together with a hardware secure element and the associated software.
That's a triangle combination of software, hardware-secure elements, and RFIC. That is actually what gives us the differentiation, and which I believe is also a pretty high hurdle for competitors. If you only look at it from an RFIC perspective, it's probably not such a big deal over many years to make a competitive product. For the whole solution to be shipped, it's a big deal, and that's where we have the lead.
Great. That's helpful. Just a broader one. Following up on John's question earlier, and really in your prepared remarks, you called out China specifically. Obviously, I assume a portion of that is the automotive coming back, and some industrial as well, but could you kind of size where that benefit came in? Was it more on the auto side or more on the industrial IoT side?
Well, are you asking for Q2 or for the guidance into Q3?
Both would be helpful, but you made the comment, I think, on Q2 and going into Q3, so whatever you can give on both.
Yeah. Okay.
I'll take what I can get, right?
Yeah. Clearly on Q2, China was a big factor across the board, and that's simply because the pandemic, it has had a phase shift. China has seen the biggest impact from the pandemic in Q1 already. Then a pretty good recovery in Q2, while Europe and the U.S. have been essentially shut down for the earlier part of Q2. Now, if you go into Q3, it is actually more broad-based. The auto recovery, for example, the 20% quarter-on-quarter, that's not just China. This is really across the board. We do see auto recovering, especially, I would say, from a sequential perspective in U.S., in Europe, and also starting in Japan, because those are the places where the shutdowns were in Q2.
Thanks a lot. Operator, we have time for probably one more question today.
All right, your last question comes from the line of Chris Caso with Raymond James. You may now ask your question.
Yes, thank you. Good morning. Question is with regard to the buyback. Perhaps you could talk about what the plans are there and what would be the criteria for doing some resumption.
Shall I take that?
Peter, I guess that's for you.
Yeah. We'll restart the buyback when we get to a ratio of two times net debt to trailing 12 months EBITDA. We ran 2.2 in Q2. We'll be, I don't know, 2.3, 2.4 in Q3. It's gonna be a while before we see the buybacks restart again. It's when our net debt levels have returned to two times.
Got it. That's clear. Thank you. Just a follow-up. With regard to some of the comments you made about production and getting the utilization back up and getting to your inventory target, do you have a timeframe in mind when, and obviously this is dependent on demand, but absent a inflection in demand, how long does it take at these utilization rates to get to your inventory targets internally?
Well, we get down to 100 in Q3, and we're trying to get to 95. At these utilization levels, assuming revenue is okay, it's pretty easy at this point to get down to 95.
Right. By the end of the year for sure.
Yeah, I would think so, yeah.
Got it. It's helpful, thank you.
Okay. I would now like to turn the conference back to the company.
Great. Thank you, Jerome. Thank you everyone for your interest and your time today. I'm not sure, Kurt, if you had any last-moment remarks you'd like to make before we sign off today.
Thanks, Jeff. Let me thank everybody for your attention today. I think it's indeed a good moment now because it all looks like Q2 was the trough. We do see growth going forward across our businesses, across the regions. I feel good about this because it is clearly a mix of a recovery of the markets, but at the same time, playing out of our company's specific drivers, which is most important for us to win market share going forward. At the same time, we still treat this very cautiously. Which means from a OpEx spend perspective, from an inventory perspective, from all the factors which are under our direct control, we're gonna stay very vigilant, as we've done over the past period. With that, I thank you all and speak to you next time. Thank you.
Great. Thank you all.
Well, ladies and gentlemen, this concludes today's conference. Thank you for your participation and have a wonderful day. You may all disconnect.