德州儀器 (TXN) 2016 Q3 法說會逐字稿

Good day, and welcome to the Texas Instruments' third-quarter 2016 earnings release conference call. At this time, I would like to turn the conference over to Dave Pahl. Please go ahead, sir.

Thank you, and good afternoon and thank you for joining our third-quarter 2016 earnings conference call. As usual Kevin March, TI's Chief Financial Officer, is with me today. For any of you who missed the release you can find it on our website at TI.com/IR.

This call is being broadcast live over the web and can be accessed through our website. A replay will be available through the web.

This call will include forward-looking statements that involve risks and uncertainties that could cause TI's results to differ materially from management's current expectations. We encourage you to review the notice regarding forward-looking statements contained in the earnings release published today as well as TI's most recent SEC filings for a more complete description.

I'll start with a quick summary. Revenue and earnings per share for the quarter were slightly above our expected range. Compared with the year ago, demand for our products continued to be strong in the automotive market and strengthened in the industrial market. Demand in the personal electronics market was about even from a year ago.

In our core businesses, Embedded Processing revenue grew 10% and Analog revenue grew 6% compared with the same quarter a year ago. Operating margins increased in both businesses. Earnings per share were $0.94.

With that backdrop, Kevin and I will move on to the details of our performance, which we believe continues to be representative of the ongoing strength of our business model.

In the third quarter, our cash flow from operations was $1.4 billion. We believe that free cash flow growth, especially on a per-share basis, is most important to maximizing shareholder value in the long term. Free cash flow for the trailing 12 month period was $3.9 billion, up 8% from a year ago. Free cash flow margin was 29.5% of revenue, up from 27.5% a year ago.

We continue to benefit from our improved product portfolio and the efficiency of our manufacturing strategy, the latter of which includes our growing 300mm Analog output and the opportunistic purchase of assets ahead of demand. We believe that free cash flow will be valued only if it's productively invested in the business or returned to shareholders. For the trailing 12 month period, we returned $3.8 billion of cash to investors through a combination of dividends and stock repurchases. Also today, we announced a 32% dividend increase that Kevin will provide more detail on in a moment.

From a year ago, Analog revenue grew 6% due to High Performance Analog, Silicon Valley Analog and Power Management. High Volume Analog & Logic was about even. Embedded Processing revenue increased by 10% from a year ago, due to growth in all three product lines, led by Processors. In our Other segment, revenue grew 7% from a year ago, primarily due to calculators and DLP products, and was partially offset by decreases in royalties and custom ASIC products.

Now, I will provide some insight into this quarter's revenue performance by end market versus a year ago. Automotive demand remained strong with most sectors growing double digits. Industrial demand improved and had broad-based growth with most sectors growing. Personal electronics was about even, despite continued year-over-year decline in demand from one customer. This decline was offset by growth elsewhere. Communications equipment grew from a year ago, and was even sequentially. Lastly, enterprise systems grew.

We continue to focus on making our company stronger through manufacturing and technology, the breadth of our product portfolio, the reach of our market channels, and our diverse and long-lived positions. These four attributes, taken together, are at the core of what puts TI in a unique class of companies capable of long-term free cash flow growth.

Kevin will now review profitability, capital management and our outlook.

Thanks, Dave, and good afternoon, everyone. Gross profit in the quarter of $2.28 billion was 62% of revenue. From a year ago, gross profit margin increased 380 basis points, primarily due to lower manufacturing costs. Operating expenses were $804 million or 21.9% of revenue. Over the last 12 months, we've invested $1.33 billion in R&D, an important element in our capital allocation.

Acquisition charges were $80 million, all of which where the ongoing amortization of intangibles, which is a non-cash expense. Operating profit was $1.4 billion or 38.0% of revenue. Operating profit was up 20% from the year-ago quarter. Operating margin for Analog was 40.9% and for Embedded Processing was 27.7%. Our focused investments on the best sustainable growth opportunities with differentiated positions enabled both businesses to continue to contribute nicely to free cash flow growth.

Net income in the third quarter was $968 million or $0.94 per share.

Let me now comment on our capital management results, starting with our cash generation. Cash flow from operations was $1.41 billion in the quarter. Inventory days were 117, consistent with our long-term model of 105 to 135 days. Capital expenditures were $139 million in the quarter. On a trailing 12 month basis, cash flow from operations was $4.46 billion, up 8% from the same period a year ago. Trailing 12 month capital expenditures were $585 million or 4% of revenue.

As a reminder, our long-term expectation is for capital expenditures, including the expansion of our 300mm Analog capacity, to be about 4% of revenue.

Free cash flow for the past 12 months was $3.87 billion or 29.5% of revenue. Free cash flow was 8% higher than a year ago. Our cash flow reflects the strength of our business model.

As we've said, we believe free cash flow growth, especially on a per-share basis, is most important to maximizing shareholder value in the long term and will be valued only if it's productively valued in the business or returned to shareholders. Our intent, over time, remains to return all of our free cash flow plus any proceeds we receive from the exercise of equity compensation minus the net debt retirement.

Consistent with this intent, we have announced a 32% or $0.12 increase in our quarterly dividend. This brings our quarterly dividend to $0.50 per share, or $2 annualized. I will also note that we have updated our long-term dividend model from our prior target of about 50% of trailing four-year average free cash flow to a range of 50% to 80%. This new range provides a more robust framework to adjust the allocation of our shareholder returns between dividend growth and share repurchases.

Today's announcement extends our string of dividend increases to 13 consecutive years, which we believe are an important element to shareholder returns. In the third quarter, we paid $382 million in dividends and repurchased $500 million of our stock for a total return of $882 million.

Total cash returned to shareholders in the past 12 months was $3.82 billion. Outstanding share count was reduced by 2% over the past 12 months and by 42% since the end of 2004 when we initiated a program designed to reduce our share count. These combined returns demonstrate our confidence in our business model and our commitment to return excess cash to our shareholders.

Fundamental to our commitment to return cash are our cash management and tax practices. We ended the third quarter with $3.14 billion of cash and short-term investments, with our U.S. entities owning about 80% of our cash. This on-shore cash is readily available for a variety of uses.

Our orders in the quarter were $3.64 billion, up 6% from a year ago.

Turning to our outlook: for the fourth quarter, we expect revenue in the range of $3.17 billion to $3.43 billion and earnings per share to be in the range of $0.76 to $0.86. Our expectation for our annual effective tax rate in 2016 is unchanged at about 30%. This is the tax rate you should use for the fourth quarter and for the year.

In summary, we believe our third-quarter results continue to demonstrate the strength of our business model.

With that, let me turn it back to Dave.

Thanks, Kevin. Operator, you can now open the lines up for questions. In order to provide as many of you as possible an opportunity to ask a question, please limit yourself to a single question. After our response, we'll provide you an opportunity for an additional follow-up. Operator?

Thank you.

(Operator Instructions)

Stacy Rasgon with Bernstein Research.

Hi, guys. Thanks for taking my questions. First, I want ask about OpEx. It was roughly flat quarter over quarter. R&D up and you talked about increasing R&D investments. Given that, what are your expectations for R&D and total OpEx as we go into Q4? I know seasonally they would be down, but given the increases that we're seeing in R&D how should we think about OpEx going into next quarter?

Stacy, I think as we talked about, R&D, what we tend to reallocate. And we'll continue to reallocate though the first half of next year, resources from other areas in the company into R&D. Overall, you should think about OpEx in 4Q pretty much being the same as you've seen in each of the last several years, and that is down because of a seasonal benefit there, just from the fact that there is going to be Thanksgiving and Christmas holidays going on. So total OpEx should continue to be seasonally down.

Any follow-up, Stacy?

Yes. Thank you for my follow up. For my follow up, I wanted to ask about the change in strategy between -- or apparent change in strategy -- between dividends and buybacks. What does that imply in terms of how you see the current return on your own stock versus the potential for how -- at least, how you view the stability and strength of your business model going forward? It seems like you're allocating more into dividends now.

Well -- and that is correct. We are planning to allocate more towards the dividends at this point in time. We've always said the buy backs will vary as a function of the stock price, and a dividend model that can vary from 50% to 80% we believe is a more thorough comprehensive of capital allocation. This new model provides a more robust framework to adjust the allocation of our shareholder returns between dividend growth and share repurchases, and that's really what we're talking about here.

Okay. Great. Thank you, Stacy, and next caller please.

CJ Muse with Evercore.

Good afternoon. Thank you for taking my question. I guess first question on the Analog out-margin side, I believe record levels for you, 40% plus year. I think the highest I've seen out there is high 40%s at Linear. Curious, do you think you could achieve that over time? As part of that thought process, what percentage of your mix would have to be on 300mm to get to that kind of level?

CJ, the possibility for improving margins in Analog are not anything that we would pull back your expectation on because you've of actually hit the essence of the formula there. That is we have more and more revenue being generated on 300mm, while we enjoy a significant cost-benefit versus any of our competitors. We'll see that just falling straight through not just to margin but all the way through to free cash flow, which is most important to our mind. Clearly, as you pointed out, this has been a new high for our Analog segment, and that business still has plenty of runway underneath as we go forward.

Yes. I'd just add, CJ, that when you look at our practice of buying capacity well ahead of demand. That puts us in a place of -- at the end of last year we had about $6 billion of open 300mm capacity. As revenue grows, if that was 200mm capacity that will help to grow free cash flow margin as well as operating margins in the businesses that use that. The fact that it's 300mm, of course, and the fact that it's a 40% lower cost will certainly help us as well. You have a follow up, CJ?

I do. A shorter term question. As you look to your implied guide of down 10%, and I know typical seasonality is kind of thrown out the door here. In the last five years, your average is roughly down 7%, so is there any one-time issues here, customer concentration, conservatism, how should we think about that?

I think if you look at the midpoint of the guidance, CJ, it would suggest a year-on-year growth of about 3%. In general, we just believe that's just very consistent with the modest overall macro environment, and that's really been the case for the last two years. Our view really remains that we're in a slow growth economic environment. We don't really see any catalyst that's really going to change that one direction or the other anytime soon.

Thank you. We'll go to the next caller, please.

Toshiya Hari with Goldman Sachs.

Thank you for taking my question. My first question is on the demand environment. I was hoping you could provide a little bit more detail on the different buckets within TI -- industrial, personal electronics, comms equipment, and automotive -- specifically what sort of trends you saw in the quarter? More importantly, what your internal expectations are heading into Q4.

Okay. Let me just talk about what drove the quarter. As I mentioned before, the automotive market remained strong. We've got five sectors inside of that market, and most of them had grown double digits. Industrial demand improved, and the encouraging part of that was it was very, very broad based. We had most of the sectors growing.

Inside of personal electronics that was even, and that was despite continued year-over-year declines in demand from one customer. Speaking to the diversity of our product portfolio and our customer base, we were able to offset that with growth elsewhere inside of personal electronics. Then, communications equipment grew from a year ago, was even sequentially. Lastly, we did see growth in enterprise systems, as well.

When we look into the fourth quarter, we don't -- we'll call out anything that is moving around significantly. You've seen us do that in the past when we had some type of discontinuity to help explain the results, and that's not the case as we move into fourth quarter.

You have a follow-up, Toshiya?

I do. Thanks for that. My second question is on M&A. Clearly, TI has been very disciplined, especially with regards to valuation over the past couple of years, maybe even several years. Obviously, you've seen a lot of activity elsewhere in the industry. I was just hoping you could provide an update on -- as to how you approach M&A these days. Thank you.

Toshiya, our approach to M&A is really no different than we've talked about in the past. There are really four attributes that we think that we have that put us in a pretty unique position, versus our competitors, and actually quite difficult for others to duplicate. Those include our approach to manufacturing technology, the breadth of our product portfolio, our reach to our market channels, and the diversity and all the positions that our products enjoy in the marketplace.

When we look at the acquisitions, it's about trying to see if we can strengthen those four attributes. Really it starts out by looking at companies from the standpoint of their strategic match, and that means will that strategic match leverage, the strength and competitive advantages we have been building on for these last number of years? From a strategic standpoint that would probably include a company that's likely engaged in analog and probably has pretty good exposure to industrial and automotive, which is where we believe most of the growth for semiconductors is going to happen for quite a few years to come.

If a potential target passes a strategy test, then we've got to get past the numbers test. That really means the price that we pay for it. Will the return on that price, will the return on that investment in capital exceed our weighted average cost of capital in about a four-year time frame? If we can't get all that work, then we will remain disciplined and just stay on the sidelines.

As we look at what's been going on in the M&A landscape these last couple of years, it strikes us that there are probably really two things underway from most participants in that space. They're either trying to increase their scale or broaden their market opportunities because they have fairly narrow markets. In our case, we already enjoy advantages in both of those, so we don't feel compelled to have to move unless we can pass that strategy and math test I just talked about a moment ago.

Okay. Thank you, Toshiya. We'll go to the next caller please.

Chris Danely with Citi.

Hey. Thanks, guys. One more question on the Q4 revenue guidance. I'm just looking at the results of the last 10 or so years, and it's only been down double-digit, I think, once in eight years or something like that. I remember last year you guys got it down 7% sequentially, so did anything -- has anything changed recently? Have you seen any change in order rates, any change in the end markets, change in linearity, change in anything? I'm just wondering, why the down 10?

Again, Chris, no big changes on that. I think is it just consistency with the overall macro. If you look over the last five years, we've had a range from -- as high as down 1.3%. Back in 2012 I think we were down 12.1%, so certainly our range that we've given is certainly inside of those things from a sequential standpoint. That's kind of how we're looking at it. Do you have a follow-on?

Yes. Can you just talk about utilization rates? How much below capacity you guys are, and where we should expect utilization rates to trend in Q4, and how you feel about inventory as a result of that?

Chris, on utilization it really hasn't changed much in probably the last three or four quarters, and I don't expect it to change anytime in the near term. We load the factories in accordance with our expectation of demand, and because of our inventory strategy sometime that demand just goes beyond next quarter and may include the following quarter. From an inventory standpoint, recall that our model is to have 105 to 135 days. We had quite a bit more inventory last quarter, from a days standpoint. We had mentioned then that we had built that in anticipation of growing revenues in the third quarter. We have had those growing revenues and our inventory days dropped down to 117. I expect that we'll continue to operate within the model days that we talk about and our utilizations will continue to be relatively stable, as we've seen the last three or four quarters.

Great. Thank you, Chris. Next caller please.

Joe Moore with Morgan Stanley.

Great. Thank you. I have one short-term question and one long-term question. In the shorter term, you said personal electronics kind of flat year over year despite declines at one customer. Can you give us a sense for which areas are growing year over year? Is it other smartphones, is it PC, what are the growth components of that business right now?

At that level, we've had PCs come in year over year about even, which is an improvement from what we saw. I think that outside of that one customer, we've actually seen really broad-based growth. I think that all the classical areas inside of personal electronics are doing well.

You have a follow-on?

Sure. Thank you. Longer-term side, as you look at your automotive business, obviously, that's a focus for you guys. Can you talk a little bit about how do you plan for that business and how much of your effort is sort of aimed at traditional building block catalog type products, and how when you think about markets like autonomous driving, where there might be systems-level solutions, how much are you thinking about the need to do something that's a little bit more revolutionary rather than evolutionary in that part of the market?

Sure. Yes. We have got five sectors that make up the automotive market for us that we look at. They're infotainment, safety, advanced driver assist systems, or ADAS, body electronics and lighting, and then powertrain systems, which include hybrid and electric systems. We're -- have a very intentional focus to invest across all of those five sectors. We're seeing that in the results, with revenues in most of those sectors growing double digits.

We have an approach inside of automotive that is similar, really, across our other markets, where we look to leverage our four competitive advantages, and I think that that's helping to turn in those results. We are favoring what I would say catalog and application-specific solutions. I wouldn't put that into the category of full turnkey system-level solutions where we're doing what we would consider to be the work that our customers do. We are providing them building blocks that have different levels of integration, but we're not looking to provide a full ADAS system, as an example. We'll enable those systems, and I think that that's pretty consistent with what you see across all the markets and sectors that we serve.

Okay. Thanks, Joe. We'll go to the next caller, please.

Chris Caso with CLSA.

Yes. Thank you. Good evening. First question would be regarding your comments on the industrial segment. You talked about some broad improvement. Could you talk about your level of visibility there and the likelihood of that sustaining into the end of the year and moving into next year?

Can you just asked that once again, make sure I got the question right, Chris?

Sorry, it was regarding the industrial segment and your comments on that. You talked about broad improvement, if you can give some more color on that and your view on the potential for that to sustain into year end and into next year?

Yes. As we talked about, last quarter we did see an improvement in industrial demand. It actually had been flat for several quarters, year on year, so we were encouraged by that. This quarter, we built on that a little bit, and like last quarter it was very, very broad based. I won't try to go in and try to predict how long that will last or carry forward, but I definitely would say that we were encouraged as we saw most of those sectors growing.

You have a follow-on, Chris?

I do. The follow-on is regarding inventory in both the channel at your customers. Given the stronger growth in Q3, was any part of that growth perhaps attributed to some increase in inventory? I understand most of your inventory in the distribution channel is on consignment now, but what's your view of customer inventory levels right now?

If we look into the distribution channel, inventory actually decreased by about a half a week, and it's currently at very, very low levels at about four weeks. That's pretty similar to what we saw a year ago. Again, we believe that that inventory level just reflects an environment of really good product availability. We've got good inventories on our books, available to be able to ship. We've continued to have very stable lead times. Those things kind of taken together just drive very high customer service metrics, so customers, in general, are getting products when they want to have them.

I'll just make the reminder, too, that that four weeks of inventory is just structurally lower because of the consignment programs that we have at our major distributors.

Thank you, Chris, and we'll go to the next caller, please.

Ambrish Srivastava with BMO.

Hi. Thank you. My first question is on the Embedded side. This segment has been growing really well for the last -- now the third quarter in a row, you're posting pretty good year-over-year growth. What's driving that, Dave? I know it has three major segments. Could you just help us understand the drivers behind the growth?

Yes. I think when you look at the embedded market, we've seen good growth there for a number of quarters. If you look at the end markets, they have a very good exposure to automotive, to industrial, and to the comms market. As we talked about the communications market -- well, really all three of those markets contributed to growth. The great thing about the growth is it's very broad based. It's not in one market. It's not in one sector. It's not driven by one technology. So we feel really good about the sustainability of that growth, and the great thing with that is it's fallen through to the operating profit, which you can see inside of that business. Most importantly, it's fallen through to free cash flow growth.

You have a follow-on, Ambrish?

Yes, I did. Could you talk a little bit about your -- there seems to be a change in the relationship in how you work for the major distribution partners, and our sense is this is nothing new. Could you just help us understand how you thought through the pros and cons of this change and whether it's just a matter of you having more in consignment or it's a trade-off between confidence of sourcing your own wins versus potential loss in future sales? Thank you.

Sure. Our relationship with distributors have been evolving and always evolve. I think any healthy business relationships evolve, but especially as the reach of our channel advantages continues to strengthen. You've heard Kevin and I both mention our four competitive advantages, and certainly the reach of our channel is one of them. That includes both the size and the skill of our direct sales and applications teams, but it also includes the growing capabilities of the web and TI.com.

As an example, three years ago we discontinued our incentive programs for demand creation on accounts that represented about 80% of the TAM, where we had resources calling on those accounts directly. We made that change because our team is just more effective and efficient in demand creation, and this has played out as we've continued to gain share every year since then. Then more recently, we've discontinued the incentives for the remaining balance of the accounts. We expect, as with all relationships, we'll continue to evolve, but we do believe distributors will continue to play an important role in order fulfillment for our customers.

Okay. Thank you, Ambrish, and we'll go to the next caller, please.

David Wong with Wells Fargo.

Thanks very much. TI and several other chip companies had really impressive September year-over-year growth that was many percent about guidance, and you're also giving December guidance, which is solid, even though the year-over-year growth in December looks like it's going to be below September. Can you give us some feel as to what's happening here? Did things weaken as September progressed, or is December a particularly tough comparison?

David, if you look at our normal seasonality, the one thing that we can conclude for sure is that second and third quarter are our seasonally strongest quarters, fourth and first are seasonally weakest. Certainly, that will play in effect as we go into fourth quarter, so it's not unusual for us to see a fourth quarter that's down and in fact why we point to the year-over-year guidance and how that is consistent with just -- our belief is that things in the macro really haven't changed too dramatically, so that has gone into the guidance.

Do you have a follow-on?

No. I'm good. Thank you.

Okay. Thank you, David.

Vivek Arya with Bank of America Merrill Lynch.

Thanks for taking my question and congratulations on the strong results and the dividend boost. I'll also try my hand on this Q4 outlook question, because I still don't know whether this is a top-down forecast or it's a bottoms-up forecast. Which specific end market is going down sequentially so much for your overall sales to go down 10%? I don't know whether it's the Other segment, which had very strong growth in Q3. I know, Dave, you mentioned on a year-on-year basis, but last year I believe your sales were down in Q4 so the comps should have been easier versus the last year. That's really my first question, that is this a top-down forecast or is this a bottoms-up forecast?

I think Dave has probably answered that same question about three or four times now, and the answer is still going to be the same. It is our best outlook right now based upon a continuing weak macro economy. We had a book-to-bill in 3Q of 0.99 compare that to 1.0 a year ago, so the visibility in to the quarter is a little bit reduced versus what it was a year ago. There is no one market, no one customer, no one product. It is our best estimate at this point in time, and I'll remind you it does include a range. That range probably comprehends to a high probability of where we'll land for fourth quarter.

I'd just add that, Vivek, that whenever we've had a big discontinuity that you're searching around for, we've been clear at calling those things out. We want to be able to provide that transparency when it's necessary.

Do you have a follow-on?

Yes. Thank you. Now, given the focus on free cash flow, why isn't TI better utilizing an asset like your balance sheet? Because you're operating -- if my numbers are right -- at about 0.15 net debt to trailing EBITDA, and most of your peers are comfortable operating at 3 to 4 times. So given this low growth environment and sort of negative returns for holding cash, why is TI holding so much cash, and why not buy back more of your stock if you think M&A looks expensive or not attractive right now? Thank you.

I take it from that you would consider our stock quite undervalued at the moment in time, so I appreciate that observation. We have been buying back shares, every quarter, since the fourth quarter of 2004. Our method for buying back shares is to be disciplined and consistent in how we go about doing it. More importantly, what we've talked about is that we will buy back more shares when the price is lower and probably fewer shares when the price is higher.

To that end, we recently -- as with today's announcement -- have updated our allocation budget, if you will, for what we allocate from free cash flow to dividends, from 50% of our trailing four years average free cash flow to 50% to 80%. We think this new range provides a more robust framework that allows us to adjust the allocation of our shareholder returns between dividend growth and share repurchases.

To as far as leveraging the balance sheet any further, right now, that balance sheet has been leveraged higher in the past. We have been slowly paying that debt back down over time, which means the debt is available should an important strategic opportunity present itself at the right moment in time. We believe that leveraging now, just for the sake of leveraging, would actually block the balance sheet from being available should an opportunity present itself from a strategic standpoint.

Okay. Great. Thank you, Vivek. We'll go to the next caller, please

Amit Daryanani with RBC Capital Markets.

Thanks a lot. I have a question and a follow up as well. Just to start on the dividend increase that you guys announced. Could you just talk about how -- when I think about the future dividend payout strategy that you guys have, is it going to be more a reflection of your free cash flow growth on an annualized basis, or will the desire be to increase dividends along with free cash flow growth plus increasing the payout to get to the midpoint of the 50% to 80% range? How do I think about this on an annual basis?

If you think about that what you have been seeing is probably what you will see. That is our free cash flow has been growing consistently for quite a few years now, and our formula design means that by definition our dividends are increasing consistent with that. What this new range does is just give us more framework to adjust the allocation between dividends and stock buybacks. Recall that we take all the free cash flow and we use it either for dividends or buybacks. This allows us to allocate more towards dividend growth as opposed to buybacks going forward.

If we reach a point where share price tends to be relatively lower at a point in time, then this allows us to slow down the dividend growth rate and adjust that back towards dividend repurchases. That's what this new range is designed to do, is to give us more flexibility to be responsive to the changing market valuations that we see out there.

Got it.

Do you have a follow-on?

If I could just follow up on -- your R&D sort of increased double digits for a few quarters, sounds like it will do so again. Does that suggest that you're kind of shifting a little bit from -- harvesting towards investing more in your business? If that's the case, should we recalibrate expectations? Maybe come to 2017 and 2018, that the share gain that you guys have typically had of your 30, 40, 50 basis points, could be somewhat higher giving all the investments you're making?

Certainly we're investing more into R&D with the intent of continuing our share gains and ideally accelerating them. Really what we're doing on this stuff is we're being very disciplined on R&D as well. We're not spending just for the sake of spending, but we're spending because we have a higher degree of confidence in the ideas that are coming to us that we can actually turn that into profitable revenue growth and ideally market share. We're doing that by just reallocating our internal resources, as we mentioned in the beginning. We're not taking our total spending up, but we're taking people who may have been working in the manufacturing lines or people who may have been working in sales or general administrative areas and putting them into working on R&D projects as well. It's an overall reallocation internally, and I would expect that that will continue to support market share gains as we go forward.

I would add that we've gained, on average, 30 to 40 basis points of market share, and I think we've gained share each of the last six years. I think we're on track to be able to put a seventh year up, and we'll wait for the numbers to get posted before we declare that. Again, market share just doesn't move quickly in the markets that we've got. I think that that just speaks to the quality of the markets. I think these opportunities that Kevin's talking about just gives us confidence that we can continue to have share gains into the future.

Thanks, Amit. We'll go to the next caller, please.

Ian Ing with MKM Partners.

Yes. Thank you. In your prepared comments you talked about being opportunistic in terms of purchasing assets ahead of demand. Do you anticipate any greater opportunities or fire sales as the industry, I would say, rapidly consolidates? I mean, you've probably got enough wafer fabrication equipment, but maybe there's other areas like test and packaging. Thanks.

Ian, I would say that we are continuously on the lookout for buying assets that we don't need now but we believe we will need at some point in time or could be a strategic cost advantage to us at some point in time. We have been, in pretty much every quarter, picking up some piece of equipment here or there. Again we look at the total opportunity, meaning can we get it at the right price and can it be something that will contribute to our free cash flow as we go forward? So we look at it in that context when we considered these sort of things. I would expect that we'll continue to be out there making opportunistic acquisitions, some may rise to the level of an announcement, but most of them will probably just remain as they have been, just spot purchases we do from time to time as we look around the world.

I'll also point out we're making those purchases while keeping our CapEx at about 4% of revenues so continuing to strengthen that competitive advantage. Do you have a follow-up, Ian?

Yes. Over to Embedded Processing. Questions on expectations for DSP applications, should we think of it as largely communications infrastructure, 4G and 5G? I noticed there's DSP IP companies out there that are talking about using these chips as accelerators for things like deep learning, artificial intelligence, accelerating some applications.

When you look at our portfolio, the lines between what's a DSP and what's not a DSP blur to some degree. Like you said, in the comms equipment space you've got processors that have multi-cores of both ARMs and DSPs but also have a heavy component of hardware accelerators. Our OMAP products that are gaining traction in automotive and other applications inside of industrial have DSPs inside of them. If you look at our communication products, where we support over a dozen different standards that help machines be connected and be smarter, they will have DSPs inside of them. We kind of just look at that as an enabling technology.

The good news is that that business complements our Analog product portfolio very well. It isn't one type of product or technology that's building that business. We've got a very good diversity of customers that we're building over time, so we're encouraged by the confidence and the sustainability of that revenue. The wires underneath and the technology underneath are important on how we deliver it, but the business model is what's really encouraging for us.

Okay. We can go to -- have time for one more caller, please.

John Pitzer with Credit Suisse.

Yes, guys, thanks for squeezing me in here. Kevin, my first question is you talked about OpEx being down sequentially in December quarter. Depending upon how much down sequentially it is, you could still have gross margins kind of flat to up in the December quarter. Notwithstanding the revenue decline, you've had prior quarters were down revenue doesn't necessarily mean down gross margins. Curious how you think about gross margins going into the calendar fourth quarter, and what levers you can still pull other than just volume to keep moving gross margins up and to the right?

John, I think you know that we normally don't try to give that precise a forecast other than just top-line and bottom and earnings. I will just remind you that just like 3Q benefited from lower costs as we increased production in 300mm and depreciation is rolling off, depreciation will continue to roll off and 300mm will continue to become a bigger portion of our wafers that we start. So you've got a number of underlying cost things that will continue to give us good results on the overall gross margin for those who are interested in that number.

That's helpful. I appreciate it. Dave, as my follow-up, you talked about your largest customer in personal electronics being down year over year in the September quarter. Do you think that's an accurate reflection of their unit demand, or is there something going on with your content or pricing? Would you expect that customer to be down again year over year in the December quarter?

Well, I will say that I'm not going to make any comments on their demand. With most of our large customers -- I'll just generalize the comment -- most of our large customers will be on some type of a consignment or hub-based system. There's usually not a large delay in the demand. When it actually gets pulled and when it is used, there's always exceptions to that overall.

From a content standpoint, I think we've mentioned before that we sell hundreds of products into our largest customers, including our largest one, and we continue to win new products. We're confident in the design wins that we've got and the revenue that that will drive overall.

Okay. Thank you, John, for you call. Thank everyone for listening tonight. A replay of this call will be available on our website. Good evening.

That does conclude today's call. We appreciate your participation.