Good day, welcome to the Texas Instruments third quarter 2013 earnings conference call. At this time, I'd like to turn the conference over to Mr. Ron Slaymaker. Please go ahead, sir.
Good afternoon, thank you for joining our third quarter earnings conference call. As usual, Kevin March, TI's CFO, is with me today. For any of you who missed the release, you can find it and relevant non-GAAP reconciliations on our website at ti.com/ir. This call is being broadcast live over the web and can be accessed through TI's 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 safe harbor statement contained in the earnings release published today, as well as TI's most recent SEC filings for a more complete description. Our mid-quarter update to our outlook is scheduled this quarter for December 9th. At that time, we expect to adjust the revenue and earnings guidance ranges as appropriate.
This was a good quarter for TI. Revenue came in just above the midpoint of our guidance range, growing 6% sequentially, and growing 10% if you exclude legacy wireless revenue, which declined to less than 2% of TI revenue in the quarter. Analog and Embedded Processing increased to 80% of TI revenue. The improved quality of our revenue is reflected in record gross margin in the quarter. Kevin will discuss this more in a few minutes. Revenue growth in the quarter was supported by some of the vertical markets that were especially weak in the second quarter, including computing, game consoles, and handset revenue outside of our legacy wireless revenue. Revenue from the communications infrastructure market continued to grow in the quarter. These areas were complemented by continued strength in our automotive and industrial revenue. From a year ago, revenue declined 4% due to the decrease in legacy wireless revenue.
Excluding this revenue, we grew 3% from the year-ago quarter. On this basis, this was the first quarter of year-on-year growth since the third quarter of 2012. Again, excluding legacy wireless, we expect growth to accelerate to 8% in the fourth quarter at the middle of our guidance range. The strength of our business model not only provides strong profitability, it also gives us confidence that we can sustainably generate $0.20-$0.25 of free cash flow for every dollar of revenue. This metric is especially important to our shareholders, as our capital management strategy is to return all of our free cash flows to them, except what is needed to repay debt. When measured over the trailing 12 months, free cash flow was 24% of revenue. Over that same period, we returned 133% of free cash flow to shareholders.
In the third quarter alone, we returned $1 billion to shareholders through a combination of dividends and stock repurchases. Earnings in the third quarter were above our expected range, a result of better-than-expected revenue and gross profit, tight operating expense control, and some help from discrete tax items. Let me walk through a few details of revenue. Analog revenue grew 11% sequentially. All four Analog product lines contributed to this growth, although power management was up the most, followed by high volume analog and logic. These areas benefited from sequential growth across many markets, including those vertical markets that were impacted by inventory reductions in the second quarter. Embedded Processing revenue grew 8% sequentially with processors up the most, followed by growth in microcontrollers and connectivity. Processors were driven by applications processor sales into consumer and automotive applications and DSP sales into industrial applications.
Microcontrollers were lifted mostly by sales of MSP430 products into industrial applications, as well as sales of microcontrollers into automotive safety applications. In our other segment, sequential growth in most product areas was offset by decline in legacy wireless. Legacy wireless revenue fell by $91 million to $57 million, as we had expected. This decline was partially offset by growth in calculators and in custom ASIC revenue, which grew as a result of communications infrastructure. Royalties also grew, DLP product revenue was about even. Turning to distribution, resales grew 9% sequentially. Distributors' inventory levels declined by about a day to just over five weeks. Kevin will review profitability and our outlook.
Thanks, Ron, and good afternoon, everyone. This quarter, gross profit was $1.78 billion, up 13% sequentially. Gross margin was a record 54.8% of revenue and expanded 330 basis points sequentially. I think it is useful to compare this quarter's gross margin to that of the third quarter of 2010, when our last high-water mark gross margin was set at 54.5%. In that earlier quarter, revenue was $3.74 billion, up 15% higher than our revenue this past quarter. Factory utilization in that earlier quarter was eight points higher, and our manufacturing capacity was lower since we had not yet brought online our cost-efficient 300-millimeter Analog wafer fab in Richardson, Texas, or our yet-to-be-acquired wafer fab in Chengdu, China We were also still in the early stages of ramping our recently acquired Aizu-Wakamatsu, Japan wafer fab.
In addition, Analog and Embedded Processing were also much smaller then, contributing 60% of our revenue at that time. The conclusion one can draw from this is that due to the structural changes that we've made at TI over the past few years, the quality of our revenue is much higher today. It is more diverse, more profitable, and less capital intensive, and we remain better positioned to support future growth from a manufacturing capacity standpoint. Continuing to operating expense, combined R&D and SG&A expense of $833 million was reduced by $27 million sequentially. Acquisition charges were $86 million, unchanged from last quarter. Almost all of this amount is the ongoing amortization of intangibles, which is a non-cash expense. The restructuring charges and other line of our income statement transitioned from a $282 million gain last quarter to a $16 million charge this quarter.
As a reminder, last quarter's gain was due to the transfer of wireless connectivity technology to a customer. This quarter's charge is associated with the shutdown costs from the previously announced factory closings in Houston and Hiji, Japan. Operating profit was $844 million, or 26% of revenue. Our tax rate in the quarter was 23%, a point below our 24% annual effective tax due to discrete items that were included in the third quarter. Our annual effective tax rate is unchanged, and 24% is the rate you should use in your models for the fourth quarter. Net income in the third quarter was $629 million, or $0.56 per share, which includes $0.01 of discrete tax benefits. Let me now comment on our capital management strategy, starting with cash generation. Cash flow from operations was $1.15 billion in the quarter.
We increased our inventory by $6 million compared with the prior quarter. Inventory days increased by one day to 106 days, consistent with our model of 105 to 115 days. Capital expenditures were $124 million in the quarter, and free cash flow was $1.03 billion. On a trailing 12-month basis, cash flow from operations was $3.27 billion, about the same as a year ago. Trailing 12 months capital expenditures were $402 million, down 27% from a year ago. As a result, free cash flow was $2.87 billion, up 4% from a year ago. Free cash flow was 24% of revenue for the trailing 12-month period, within our expected range of 20%-25% of revenue. In the year ago trailing 12 months period, free cash flow was 21% of revenue. Capital expenditures for the past 12 months were 3% of revenue.
Our continued low capital spending level is a direct result of the strong capacity position that we have built with our strategic investments over the past few years. The cash flow that will result as we continue to fill up this capacity should be strong in the years ahead. I'll note that depreciation expense for the past 12 months exceeded our capital expenditures by $496 million. As a percent of revenue, depreciation was more than 400 basis points higher than our capital expenditures. This is one of the reasons why our free cash flow has been trending higher than our net income. Of course, as depreciation declines to the rate of capital spending over the next few years, gross margin will benefit.
Another reason why free cash flow has trended higher than net income is the non-cash amortization expense of $80 million to $85 million per quarter that will remain on our income statement for another six years. For the past 12 months, amortization expense was $339 million, or about 3% of revenue. As we've said, strong cash flow, particularly free cash flow, means that we can continue to provide significant cash returns to our shareholders. In the third quarter, TI paid $308 million in dividends and repurchased $734 million of our stock. Our capital management strategy is to return all of our free cash flow to shareholders, except for what we need to repay debt. In the last year, we reduced our debt level by $500 million. Free cash flow was $2.87 billion, and we returned a total of $3.82 billion to shareholders, or 133% of free cash flow.
We've been able to return more than our free cash flow because proceeds from exercises of employee stock options totaling $1.28 billion over the past 12 months have also been an additional source of cash for the company. To break out the cash return, in the past 12 months, we repurchased $2.7 billion of our stock, or 95% of free cash flow. Similarly, we paid $1.08 billion in dividends, or 38% of our free cash flow. Fundamental to our cash return strategy and our cash management and our tax practices, we ended the third quarter with $3.59 billion of cash and short-term investments, with 82% of that amount owned by TI's U.S. entities. Because our cash is largely on shore, it's readily available for a variety of uses, including paying dividends and repurchasing our stock.
TI orders were about even sequentially. Our book-to-bill ratio was 0.97, consistent with seasonal declines in the fourth quarter. We expect TI revenue in the range of $2.86 billion to $3.10 billion in the fourth quarter. At the middle of this range, revenue would decline 8% sequentially, with about half of that decline coming from the seasonal drop in calculators. The remainder of the decline is consistent with the semiconductor industry's pattern over the past three years, 2010, 2011, and 2012, as well as our own history over that same period when legacy wireless revenue is excluded. In the fourth quarter, legacy wireless products should decline to about $50 million. We continue to expect that revenue from these products will essentially be gone as we enter next year. We expect earnings per share to be in the range of $0.42 to $0.50.
In summary, we believe the third quarter provides a preview of what TI is capable of producing as a company focused on Analog and Embedded Processing. The improved quality of our revenue is evident from the higher gross margin and free cash flow generation compared with our past. We also believe our top-line performance and potential will become more evident without the steady headwind of declining wireless revenue over the past few years. Our manufacturing capacity position remains strong, and we have a good opportunity for continued growth while maintaining our capital expenditures at low levels in the years ahead. With that, let me turn it back to Ron.
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 your questions, please limit yourself to a single question. After our response, we will provide you an opportunity for an additional follow-up. Operator?
Thank you. Ladies and gentlemen, to ask a question, please press star then one on your telephone keypad at this time. Please ensure that your mute function is turned off to allow your signal to reach our equipment. Once again, it is star one at this time. We'll go first to John Pitzer with Credit Suisse.
Good guys. Thanks for letting me ask the question. Ron, I guess my first question, just relative to that consumer bucket of computing, gaming, and handsets, can you just give us a sense of how much that represented of the overall business in the September quarter? How much that was up versus the rest of the businesses sequentially?
John, probably the best I could do is help you understand the weight of each of those segments. I don't have the growth broken out by each of those areas for you. Let me just go through real quickly the weighting of our end market exposure, which I can let you then play with to make some assumptions on. Comms in total were 27%, and this is first half 2013 data. Computing, 23%, industrial, 23%, consumer, 11%, automotive, 13%, and education, 3%. Again, the reason we called out those three areas was because I think in the second quarter, we called them out as especially weak. In fact, we also said that in our outlook for the third quarter, we expected that they would recover.
Keep in mind, the reason they were weak in the second quarter was because customers in those areas were depleting inventories ahead of anticipated new product launches in the second half, which we are now well into. Those new product launches were a factor in those areas that was driving growth. Again, comms infrastructure was already strong and growing in the second quarter. It continued in the third quarter, as was automotive and industrial. Strong second quarter continued as well into the third quarter. Do you have a follow-on, John?
I do. This is for Kevin, just on the incremental gross margin. Kevin, if you look at both the June and September quarter, your incremental gross margin both quarters were over 100%. I guess I'm just curious, to what extent was that just the wireless business declining? Do I think about a 30% type gross margin there? More importantly, when the wireless business is down to zero, how should I think about incremental gross margin from there?
Yeah, John, I'll answer the last part first. The incremental gross margin from our analysis in the past, over the course of a cycle, averages about 75%, up and down. In any one quarter, it rarely actually works out to that number. It tends to be pretty noisy in any one quarter, but over time, about 75%, that's what we do for internal planning purposes. As it relates to the most recent quarter, you're right, the fall-through was about 106%, it was certainly a pretty rich fall-through. Part of that was mix, part of that also was improved factory utilization, as we had higher revenues this quarter than the prior quarter. Consequently, we had a slight utilization benefit as well.
Okay. Thank you, John. Let's move to the next caller, please.
Next we'll go to Blayne Curtis with Barclays.
Thanks. Maybe from a high level, if you could just talk about, you definitely showed increasing year-over-year growth. September was up, then December accelerating, although off of a pretty easy compare. The full-year Analog growth seems more like a few %. If you could just talk about now that we have a full picture for the year, do you think there's anything going on in the Analog bucket, or do you think that this is kind of representative of what the overall market is growing?
Blayne, I think the year is not done, and right now there's kind of a lot of point spreads being made for our fourth quarter, but we really need to wait till we get into January to understand what the final score is. I think we feel pretty confident that 2013, when the final scores are in, will show market share gains once again for TI. Again, we have to push through the fourth quarter, but we believe, just as you've seen over the past few years, once again, we'll be gaining share in 2013 for Analog overall. Do you have a follow-on?
Thanks. Just if you could provide some color as within the segments in Analog, if there's any material difference between the sub-segments. It looked like power was up a bunch, if I have my numbers right in September, if you could talk about the growth driver there, and then where you see the biggest declines into December. Thanks.
Okay. I think in the case of power had a strong quarter. Also used the term earlier, easy compare. It was one that's probably more impacted by weakness or the inventory reductions in some of those vertical areas in the second quarter. In the third quarter, its growth was very broad-based, but it also benefited from recovery in some of those vertical markets. I would say the same for HVAL. If you look at third quarter, probably because those vertical markets layered on growth to what was already a reasonable base of industrial and automotive, power and HVAL both grew faster than the catalog spaces of HPA and SVA. Yet, I think if you go back to the first half of the year, and especially second quarter, you probably saw more relative strength in SVA and HPA.
I guess to put it all together, they're all pulling well in terms of growth in 2013 thus far. We'll see where it takes, again, where we go through fourth quarter. Some quarters you've had more strength out of SVA or HPA, and other quarters, as in third quarter, more coming out of power and HVAL. Blayne, thank you for your questions, and we'll move to the next caller.
Next, we'll go to Doug Freedman with RBC Capital Markets.
Great. Thanks for taking my question, guys. Kevin, can you give us an idea of what we should be thinking for the other operating expenses? That line seems to be moving around. I do know you had the asset sale last quarter, but ex the asset sale, still a reasonable bit of spending there. Can you give us some idea of what we should be budgeting for next year in terms of that line on the income statement?
Doug, I'm going to try to make sure I got what you're asking for. OpEx overall for the company, our model is to run between 20%-25% of revenue, depending upon where we're at in overall revenue demand. Most of that spending is actually inside the Analog and the Embedded Processing segments. There's relatively small levels of ongoing support spending inside the other segment, a little bit of R&D inside the other segment for DLP technologies. Other than that, pretty small. The larger numbers you'll see passing through the other as we go forward will be the continued amortization of intangibles, which will run about $85 million a quarter as we go through 2014. That'll eventually decline to about $80 million a quarter. Keep in mind, that'll stay there for six years until that finally winds down.
In addition, that's also where we tend to record any restructuring charges. The restructuring charges that we just had this past quarter were $16 million, and those also are recorded in the other segment. As we look into the fourth quarter, I'd expect that you'll see acquisition charges continue at about the $85 million, $86 million level, and you'll see probably final restructuring charges associated with the closure of our Hiji and Houston factories of about $20 million in the fourth quarter.
Doug, I think if you look at the income statement for the company, for the entire period of comparison here over the last four quarters, all of the acquisition charges and restructuring other lines are directly from that other segment. With that, you can then, if you want to extract that out to see what R&D and SG&A trends have been in that other segment, you'll see both. You won't see them independently, but the combination of R&D and SG&A has gone down, whether you look at it year-over-year or whether you look at it sequentially. Again, that's largely driven by the reductions in legacy wireless, as Kevin noted. All right, Doug, do you have a follow-on question?
Yeah. For my follow-on, if we look at the Embedded Processing business, that business growing nicely but not necessarily delivering the operating margins of the corporate average. Can you talk about what you're doing as far as the operating margin leverage to that business?
Yeah, Doug, as you may recall, we spent a couple of years significantly increasing our investments in that segment, particularly as it relates to the microcontroller portion of that particular segment. We've taken that spending up to a level now that doesn't need to increase anymore. Really what that business unit is focused on is the top-line growth. They get rather attractive gross margins off their parts when they ship them. It's really about diversifying the revenue base and getting more overall top-line growth to fall through to the bottom line. That's really what you just saw quarter-over-quarter, is that top-line growth came through rather nicely to the bottom line.
Okay, Doug, thank you for your questions. We'll move to the next caller, please.
Next, we'll go to Stephen Chin with UBS.
Hi, thanks for taking my question. First one's on operating expenses. In the quarter, it looks like you guys saw OpEx decline by 3% versus the original guide for down 2%. I'm wondering if there's any timing or how you guys were able to do better than your original guidance and what the expectation for OpEx going forward?
Yeah. Stephen, OpEx, we did guide to down about 2%, but we were able to keep a pretty tight handle on spending in the quarter. It came in a little bit lower than we had originally planned, which frankly just dropped right through the bottom line and produced extra earnings per share for our shareholders. As we look into the fourth quarter, we typically have some seasonality in the fourth quarter, what with the holidays for Thanksgiving and during the Christmas break. We'd expect to see OpEx probably decline somewhere in the 3%-5% range as we move into the fourth quarter.
Okay. Jeff Vaughn , Stephen?
Yes. One question on the Embedded Processing division, particularly microprocessors. With the MSP430 that you mentioned, I was wondering, is the consumer or wireless segment an area where you're looking to be more aggressive on expanding the sales of your MCU products? Thanks.
I think probably the way to think about it, Steven, is probably the biggest driver of growth for microcontrollers will be industrial. These are catalog devices that go into lots and lots and lots of different applications. That being said, if there's growth in other end markets or growth opportunities in other end markets, we'll certainly pursue them. Our expectation is that industrial will likely be the largest driver going forward. That's for the microcontrollers. For the processor side of the business, I would say, we would expect comms infrastructure certainly is part of that revenue and an important end market for us. Similarly, we sell DSP products into lots of different applications, and those broader applications are going to be just as important, maybe even more important, than communications infrastructure going forward. That's not anything negative about communications infrastructure.
It's just we have great market share, very high market share already in communications infrastructure, where there's still a lot of opportunity in some of these other areas for us to experience not only the rapid growth of that end market, but also rapid market share gains as well. Okay, Steven, thank you for your question. One of the things I should further delineate is when I'm talking about industrial likely being the fastest grower for microcontrollers, let me also more broadly include automotive in that as well, because that is a great opportunity that we're very well-positioned with those products there. Okay, Steven, thank you. Let's move to next caller.
Next we'll go to Tristan Gerra with Baird.
Hi, good afternoon. Your R&D spending as a % of revenue is fairly low relative to some of your high-value peers. What leads you to believe that you can maintain the growth, gain market share, and the mix? Are you saying that there is a possibility of raising your R&D over time?
Yeah, Tristan Gerra, let me just again remind everybody of the R&D that you're seeing for TI. The R&D spend that we've had in the Analog segment has been at a pretty consistent level versus the revenues of that business for quite a few years now. The fact of the matter is, the reality is, that R&D spend is being converted fairly efficiently into revenue growth to the point where that business continues to outgrow its competitors and gain market share each year. I would offer to you that that's actually a very efficient organization when it comes to its R&D spend. I can't explain why some of their competitors have to spend as much as they do.
Similarly, in Embedded Processing, we've increased the spending over there, as I mentioned to the last question, over the last couple of years, largely to expand the opportunities that we can pursue in the microcontroller space. After bringing up that level of spending for a couple of years now, I'd say it's high enough at this point, and we now need to see it start converting into real revenue growth, which we're starting to see early signs of. When you look at total TI R&D, it is coming down quite sharply, and that's purely a function of us winding down the legacy wireless products, including the announcement that we had that we're closing some of our operations and sites as it relates to that, and that will take our total R&D spend down as we go forward.
On the core products of Embedded Processing and Analog, I have no hesitation of being comfortable with the level of R&D spend that we have in those areas today.
Tristan, let me also just offer, in the Analog space, I would argue we're probably spending more than any of our competitors. You're looking at it as a % of revenue, but if you look at absolute dollars, yes, we're bigger in terms of revenue, but we're spending more than any competitor that I'm aware of out in that space. It's not like we're not investing in that business. We're investing quite heavily. Do you have a follow on, Tristan?
Sure, it's very useful. The part of your business that's non-consigned business, so distribution, is your Q4 revenue guidance embedding shipments about in line with end demand or any variance that you can mention relative to end demand and sell-in?
Tristan, I don't know that we have a real breakout in terms of the outlook between consigned versus non-consigned. We're really looking at it just in terms of what we shipped in or what we're shipping into that channel. I don't specifically know the answer to that. I will point, if you just back to third quarter anyway, as a reference point, I think we said TI total revenue, excluding the legacy wireless, grew 10%, and our resales or sales out of the distribution channel grew 9%. You saw a very close alignment in third quarter. I just don't have any specific data with respect to fourth quarter. The other thing I'll say, even with third quarter, we had a decline of a day of distribution inventory level. All right, Tristan, thank you. We'll move to the next caller.
Next we'll hear from Stacy Rasgon with Sanford C. Bernstein.
Hi, guys. Thanks for taking my questions. Had one quick one on OpEx. Your R&D guide for the year still puts you about $1.5 billion, which you said OpEx would be down like 4%. At the midpoint of that would imply R&D down like 12%. Are you just rounding down to that $1.5 billion? Maybe could you give us a little more color on your expectations for the R&D and the assumptions around the OpEx?
Yeah, Stacy, we do round to the nearest hundred million on all of these guidance numbers we give you, be it R&D or CapEx or any of the others. In fact, there will be some rounding inside there. Keep in mind also, the OpEx guidance I gave includes SG&A as well, and we will see that decline in the fourth quarter, really on the seasonal patterns I mentioned a moment ago with the holiday periods going on.
Do you have a follow on, Stacy?
I do. Just around, again, gross margins. Sounds like you got a bit more restructuring left over from the six-inch fab closures in Q4. Have the benefits from those closures fully impacted gross margins? I think in total it was supposed to be about 75 basis points. If they haven't fully impacted yet, can you give us some feeling for the trajectory of that impact on the margins?
Yeah, Stacy, what we talked about when we announced that we were closing those two older six-inch factories is that once each one of them was closed, we should see about $50 million per factory of cost savings on an annualized basis. One of those factories did close at the end of second quarter, and the other one should close here in the third quarter. We're beginning to see, in the third quarter, already partial benefit from that, and we should begin to see the benefit from that second factory closure materialize in the fourth quarter. Going into next year, we'll begin to see that $100 million of annualized cost savings as a result of closing those two older factories.
Okay, Stacy, thank you. Let's move to the next caller.
Next we'll go to Vivek Arya with Bank of America Merrill Lynch.
Thanks for taking my question. First one, just on the Q4 outlook, I just wanted to make sure if there's been any downshift or conservatism regarding Q4 outlook, because I think heading into the quarter, there was an expectation that, look, macro trends are improving, industrial businesses are doing better. There is this comms infrastructure build out. The broader Analog industrial players would be able to give above-seasonal guidance. Maybe it was optimism, but since then we have just seen sort of seasonal or slightly below seasonal outlook from Linear, from Cypress, and now you're also guiding somewhat below street expectations. My first question is, has there been any conservatism regarding Q4 outlook, and specifically, what segments do you think will be at seasonal patterns versus below seasonal patterns for Q4?
Vivek, I'll go ahead and make some comments on that and let Ron add some additional color. If you take a look at our guidance, the midpoint of our guidance is down about 8% third quarter going into fourth. About half of that is just the seasonal decline that we normally get with the roll-off of the calculator business. The other half is pretty much attributable to our underlying semiconductor business. We actually believe that that's probably quite seasonal and in fact quite consistent. If you take a look, as I mentioned in my prepared remarks, 2010, 2011, 2012, the semi industry as a whole, excluding memory, in fact, has seen fourth quarter declines from third quarter in each of those past three years.
Based upon the guidance that we're hearing from some of the other competitors, ourselves included, it looks like we're setting up for a fourth year of fourth Q declines. In fact, if you take a look at TI and extract out the wireless revenue, you'll see that our revenues have gone down in the fourth quarter of each of those same years, similar to the industry. In fact, when you just do the simple math during those last three years for the industry, it's down about 4%. In fact, I don't think that we're suggesting that there's some changed outlook or changed future. In fact, I would offer to you that fourth quarter appears to be shaping up quite normally to us. When we take a look at our third quarter orders, they were actually up about 1% sequentially.
Our distribution inventories, as Ron mentioned, were down about a day, and they remain quite lean with our distributors as well as our other customers. Our cancellations remain very low, and we've had no unusual requests for backlog pushouts. When we put all that together, that suggests to us that the fourth quarter is shaping up to be a normal demand environment, similar to what we've seen for certainly the last three years.
Vivek, I think the other thing I would just add is even when we talked to you at the mid-quarter update, somebody asked a question about order trends, and I remember saying at that point that orders quarter to date had been strong. Book-to-bill was above one, that we expected that to slow as we approach the end of the quarter, basically for a couple factors. One, just seasonality in general, both calculators as well as semiconductors. Was there a slowing through the end of the quarter? Yes. Also we expected that slowing would occur, and it pretty much happens almost every year and certainly has happened, as Kevin just pointed out, each of the last three years. Do you have a follow on, Vivek?
Yes, thanks, Ron. Very helpful clarification. As my follow-up, I just wanted to make sure, Kevin, did you say that tax rate for Q4 will be 24% or for the full year would be 24%? The real question is on dividends and buybacks. I think you mentioned that they have been at 133% of free cash flow over the last year. How long can they stay at those level? Should we just expect them to trend back to whatever, closer to the 100% target that you have outlined as a longer-term target? Thank you.
Okay, Vivek. The 24% effective tax rate is what you should use for the fourth quarter. With the various things occurring during the year, the entire year will be a little bit different than that. For planning purposes, use 24% for the fourth quarter. I would remind you that that tax rate includes the R&D tax credit, which is set to expire at the end of this year, and it's unclear whether or not that will be renewed by Congress in a timely fashion as we get into 2014. For us, that provides about a 2% benefit to our taxes. Absent that R&D tax credit, we'd be closer to a 26% tax rate as opposed to the 24% that we're suggesting that you use for fourth quarter.
As it relates to buybacks, I would just mention that, again, the reason it's been as strong as it has been, the buybacks and dividends at 133% of our free cash flow is a function of not only improving free cash flow, which we've certainly seen, but also the benefit the company receives when employees exercise their non-qual stock options. Given the increase in stock prices this past year for our company, we've seen an increase in employee stock option exercises. Just to give you a few numbers to summarize that, to kind of recharacterize what I've mentioned in my prepared remarks, our free cash flow for the last 12 months was $2.87 billion. Proceeds from employee stock options was $1.28 billion. That means we had sources of cash over the last 12 months of $4.15 billion.
We used or returned to shareholders in the form of dividends and buybacks, $3.82 billion. We paid off debt earlier this year of $500 million. We've used a total of $4.32 billion. We've virtually used all that we have generated this past year. We do that because one of the things we mentioned in our capital management strategy is that we have a model for how much cash that we carry. To the extent that we generate cash beyond that model level, we're quite comfortable with returning that to shareholders as we have been doing. As to the ability of that to continue to run at those high levels, that's a really hard call. I would say that stock options do expire over time, and these are generally 10-year stock option grants that are granted to employees.
Quite a few of these options are quite old, and they get used up. I would expect that as we move into the future, we're apt to see less extra cash coming in from stock option exercises. I think it's almost impossible to try to forecast the magnitude of that change.
Okay, Vivek, thank you. Let's move to the next caller.
Next, we'll go to Steven Smigie with Raymond James.
Great. Thanks a lot. I was hoping you could comment a little bit on Silicon Valley Analog here in terms of where we are in your three-year plan. If you don't have it available this quarter, would it be possible maybe on the mid-quarter update to talk about where the revenue is there, just so we can see where it was relative to where you guys acquired it?
Yeah, Steve, I would say with Silicon Valley Analog, first off, on the point of revenues, we disclose revenues at the segment level. We don't disclose revenues below that, so we won't be offering the actual revenue figures for SVA. If I were to talk about where are we at with respect to our expectation for SVA versus when we acquired National Semiconductor, I would say that the overall revenues are probably a little bit lower than we'd initially expected at this point in time, given the overall market environment that we've been faced with these past few years. However, the rate of progress on share gain has been quite a bit faster. As we mentioned during the last call, that those product lines are actually gaining share about a year faster than we expected. In addition, we've had more cost synergies than we modeled into the acquisition.
We've had lower interest expense on very low-cost debt we've been able to issue. We've had substantial benefits from taxes as a result of the acquisition. When you put it all together, the profit that we're achieving on the National Semiconductor portfolio today is very close to what we've modeled at this point in time. I'd say we're reasonably satisfied with that bottom-line effect.
Hey, Steve, even though we don't break out revenue quarterly, I can give you a kind of a rough breakout for the sub-segments of Analog for 2013 year to date. HVAL, 35%, Power, 25%, HPA and SVA are each about 20% of total Analog revenue. Again, that's year to date. Do you have a follow on, Steve?
Yeah, great. Thanks a lot for that clarity there. Just if we could follow on in terms of overall acquisition strategy in general, if you look at the microcontroller business, overall it's still fairly fragmented out there. Also on the Analog side, you guys have talked a lot about gaining share and perhaps accelerating that through acquisition. Is there anything substantial you guys might do on either side there to change your position, in any sort of substantial way at this point? Or is it more organic growth from here?
Yeah, Steve, I think that most of our growth for the foreseeable future is going to be organic to the extent that we do continue looking at acquisitions, given the valuations that we presently see with many companies out there that might be an attractive addition to our portfolio. It's difficult for us to look at what we might have to pay for some of those acquisitions and actually get a reasonable return on the investment for our shareholders. That suggests to me that to the extent that we are doing acquisitions, they're probably going to be smaller tuck-in ones, but probably come with reasonably good product lines to fill the holes that we have, or talented engineers to help with our R&D efforts. They're probably going to be biased towards Analog.
One of the problems when you go into the microcontroller side is you start getting different architectures into your family of products, that can add complexity that may not be easily leverageable. That's not to say that we wouldn't look there, that reduces the odds that you'd see a microcontroller tuck-in kind of acquisition.
Okay, Steve, thank you for your questions. Let's go to the next caller.
Next, we'll go to David Wong with Wells Fargo.
Thanks very much. Your guidance for EPS and revenue next quarter, does it assume the gross margin will be flat to slightly down with the lower revenues, or does gross margin keep rising?
Yeah, David, I think inherently with the revenue decline that we're expecting in the fourth quarter, we'll be reducing the loadings in our factories a little bit. You'd expect underutilization charges will probably increase a bit and have a corresponding impact on gross margins decline in the fourth quarter.
Okay, great. A minor administrative thing. You noted that your cash is largely onshore, you have no problem returning to shareholders. How do you get your tax rate down to 24%, so far below U.S. tax rate, still generate primarily U.S. cash?
I would first just comment that I don't really think our tax rate is all that low. 24% is an awful lot of money that we're having to send to the tax collectors. That aside, taxes inherently are pretty complex, and there's a lot of things that can affect tax rates, and especially a company's somewhat unique global footprint. Our footprint on a global basis, including the benefit that we got with the National acquisition, is allowing us to have a relatively effective current tax rate and further enabling us a pretty high repatriation of cash. As you noted, we bring most of our cash home, and we just ended this past quarter with 82% of our cash on shore.
For planning purposes, as we look into the future, we assume that the U.S. tax rate stays about the same, incrementally at 35%, and we just model that on a go-forward basis from a repat standpoint. Our objective, again, is to maintain 80% or more of our cash on shore so that we have it available to use for our shareholders for dividends and stock buybacks.
David, thank you for your questions. Let's go to the next caller.
We'll go to Tore Svanberg with Stifel.
Yes, thank you. First one for Kevin. You have about $4.2 billion in long-term debt, I think you have about $1 billion in a note payable. Can you just remind us what your plans are right now as far as paying back that debt, especially timing?
Yeah, Tore, we do have, in short-term debt, $1 billion that comes due on May 15th of 2014. We had talks during our capital management strategy discussion back in February that debt clearly is going to be a portion or part of our capital structure for quite a long time, especially given that we've got debt that goes out as far as 10 years out. That, in fact, we would continue including debt in our capital structure or rolling over portions of our debt if the economics made sense. To us, the definition of the economics making sense is if we can issue new debt at rates that are either below our impression of what the inflation rate will be or below our dividend yield.
With that in mind, I can't forecast exactly what we'll wind up doing when May comes around, but clearly, we'll be prepared to pay off that billion if we need to. If the economics make sense, we may very well decide to go ahead and just replace that with new debt.
Follow on, Tore?
Yeah, I follow on for you, Ron. I don't know if you'll answer it or not, but can you maybe comment on which end markets will do on a relative basis in the December quarter?
No, I will give you a consolation and talk about what they did in the third quarter. In the third quarter, this is not a fourth quarter comment, industrial, as we pointed out, grew and continued to be strong. Automotive grew, for us, being driven mainly by infotainment and advanced driver-assist systems such as blind spot detection, parking assist, a lot of those cool things you see on car commercials these days, and aspire one day to have a car that actually does that. Computing, notebooks were weak and tablets grew. I heard some competitors talking about weakness in tablets. Don't know where that came from. They were strong for us. Comms infrastructure, the big word was China. In third quarter, operators, in fact, started deployment, released contracts. Two-thirds of those contracts went to China OEMs, one-third to European.
I'll note we sell to both and maybe different products to different players, we expect to continue to enjoy participation in that China business. Small cell trials are continuing. What I would say is first deployments we expect to happen probably mid to second half of next year. Handsets, probably the good news there is now sub 2% of our revenue is legacy wireless. Keeps going down. Consumer, good recovery in revenue there after some of the inventory reductions in second quarter. It's probably a game console comment, mostly. Digital television continues weak. Again, sorry I didn't answer the question, but hopefully there was some value there. Tore, thank you for your question, and we'll move to the next caller. Next, we'll hear from Patrick Wang with Evercore.
Yes, thanks a lot. Just one question for me. Can you help us understand the underutilization charge in Q3, what you see for utilizations in the fourth quarter, I guess, in terms of that charge? Also what you plan to exit the year with in terms of inventory, both on the balance sheet and in the channel. Thanks.
Yeah, Patrick, the underutilization charge in the third quarter was about $70 million. That's down from about $100 million in the prior quarter. As I mentioned, with reductions in our factory loadings expected during the fourth quarter, I'd expect that charge to go up, but I won't give you a forecast with an exact number just yet. That'll be a function of exactly what our loads turn out to be by the end of the quarter. As it relates to inventory, our model for inventory is to carry between 105 and 115 days of inventory. We just closed the current quarter at 106 days, so we've got room that we probably want to take those days up a little bit more.
With the slowdown in demand in fourth quarter, that seems like a ripe opportunity for us to go ahead and bring those inventory days up a little bit as we get ready to enter 2014. Patrick, I think I understood you to say basically you got all your questions in there. To get to more callers, we'll move on to the next caller, please. Next, we'll go to Ambrish Srivastava from BMO Capital Markets.
Hi. Thank you, Ron. I apologize, I don't want to put you on the spot, but I'm getting a lot of emails. I'm confused, investors are confused. Going back to the mid-quarter update, I think you did say that the long-term seasonality is eight. You also said the variability is a lot, so the right way to think about it is the decline of 4% for the last four years or the last four-year average. Is it that we misinterpreted that and walked away thinking, not that I was modeling for down 4%, but the streak was clearly higher than me, and hence Vivek also asked that question. Did we misinterpret your comments, or did something change in the quarter?
I think the key to looking at that, Ambrish, all of that is exactly true. If you look at five-year average, it's down eight. If you look at the last four years and take, I think it was fourth quarter 2008 out, it's down four. I think what you hear us talking about here are if you look over the last three years, I think in the case of the industry and TI, both were, excuse me, it's something like down 9%. I guess the only real difference in terms of how we're discussing our own data is, I think, a meaningful change to say if you take legacy wireless out, and that was obviously three years ago, four years ago, a very significant part of our revenue mix. Today, it's an insignificant part of the mix.
Any impact that has on seasonality going forward, frankly, what it did in the past is irrelevant to our future. That's really the distinction that we're making at this point. Do you have a follow-on, Ambrish?
I did. Just to look forward now that the legacy business will be gone when we go into next year, what should we be thinking about the normal seasonal patterns? I'm sure you'll answer it, and we'll come back and ask you again for the four quarters.
The math I've done is I think if you look at first quarter and you take legacy wireless out and you look at the last three years, it's flattish. What I will encourage you to do, I think we have all of that data out on our website, and that would be a great project for you to work on tonight and publish in your note tomorrow. There'd probably be a lot of value added in that. Again, I'll refer you to that data and let you do the math.
I'll play with the data, Ron. Thank you.
You'll enjoy that. Ambrish, thank you. We'll move to the next caller, please. We'll go to James Covello with Goldman Sachs.
Great. Thanks so much for taking the question. I'll stick with the theme of a couple of callers ago and just stick to one question. There's been a lot of activity in the Analog segment lately relative to sensors. There's been some M&A activity and some of the Analog peers this year that have grown in excess of the market have grown because they've gained some share or had some opportunities in the sensor market. How do you feel about TI's position in that market? Where do you see you guys going there? Thanks a lot.
Probably the one thing I would say, Jim, that in that sensing space. Obviously there are lots of different types of sensors and they're different. I think sometime in the last month anyway, a few weeks, we did an announcement really of a breakthrough product category in inductive sensing, which is basically an entirely new class of data converters. That truly is innovative. That truly is breakthrough. That being said, when you look at the size of TI and our portfolio of 100,000 products, it can be a breakaway success, and it's not going to have a big impact on TI in terms of moving the needle. I think it's a great example of the kind of innovation that's been taking place at TI over the last decade, and why you see just consistent share gains from our company.
A lot of people talk about innovation. A lot of times that's spun up in technical terms. We really simplify it and measure it by are we producing and innovating and developing products that customers want. We really let our market share tell the story back, provide the feedback of whether we're providing innovation that is valued by our customers. That's one example. I went a little broader than your sensing question, absolutely a great space and we're right in the middle of it. Thanks, Jim. We'll move to the next caller.
We'll go to Glen Yeung with Citigroup.
Thanks. Hey, Ron, when you look at the activities that either cancellations or pull-ins, things you might not normally expect to see in the context of a set of orders, are you seeing anything there that's unusual, either one way or the other from your customers?
No. Cancellations remain at what we would call very low levels by historical standards, That's part of the reason why we really feel this is a pretty seasonal fourth quarter. We're not seeing anything on the cancellation front, for example, that would indicate it's anything weaker than normal. I know a lot of people maybe were worried about, well, what's the effect of the government shutdown and is that showing up in customer demand or anything like that? I'm not sure we have perfect visibility there, but we sure didn't see backlog being pushed out by customers. We didn't see a higher rate of cancellations, which we think are both meaningful indicators. Do you have a follow-up, Glen?
Yeah, thanks, Ron. As a follow-up, as I look at the reported gross margins, they're almost around 150 basis points more than what consensus is modeling for next year's gross margins. I recognize that movements in revenues and hence utilization will affect that. As we think about the puts and takes, assuming that next year is even a modest growth year for semiconductors, if we think about the puts and takes on gross margins, what could the consensus see right, i.e., gross margins down 150 basis points from current levels, or is perhaps consensus seeing things wrong?
Glen, I guess I would just remind people as they're putting their models together of a couple of things. As we look at what's going to drive our gross margins going forward, you need to keep in mind, as we fill these very cost-effective factories that we acquired over the last couple of years, and especially as we fill them with a higher mix of industrial and automotive end market type products, we're going to see margins continue to be pretty healthy. Another important point to keep in mind is that depreciation today is running about 400 basis points ahead of capital expenditures. By definition, that means depreciation will start to work itself down over the next couple of years, and that by itself will be a boost to gross margins.
Those are probably the three things I would really point to, and that is, again, as we grow and fill up those factories, as we improve our portfolio mix, especially industrial and automotive, and as depreciation declines over the next few years on these lower levels of CapEx, those together should provide nice tailwinds to our gross margins as we go forward.
To the scenario that Glen talked about, if 2014 has any kind of reasonable growth or any growth at all, given those tailwinds, it doesn't sound like the Street, if they're modeling on down 150 basis points on gross margin, that does not sound consistent with what you just described. Is that right?
It sounds inconsistent, and it would also suggest that you're missing just exactly how much free cash flow we'll generate next year.
Okay. Thank you, Kevin. Okay, I think, operator, we're going to try to squeeze in one final caller, then we'll wrap it up.
Okay. Our final question will come from Ross Seymore with Deutsche Bank.
Hey, Ron, thanks for squeezing me in. I'll stick with the one-question theme. Kevin, you just talked about some of the puts and takes for gross margin for next year. From the $800 million roughly level you were for kind of core OpEx in the fourth quarter of this year, X charges, restructuring, et cetera, what sort of OpEx puts and takes should we think of heading into 2014?
Ross, I think the simplest thing I'd suggest to you is that clearly the people who are in R&D and SG&A, the OpEx buckets, as we call them, are correctly anticipating that they'll see pay and benefit increases in 2014. We would certainly see the effect of that. I would probably also remind you that in first quarter, our OpEx typically is seasonally up from fourth quarter. That's a function of two things. One is the absence of the holiday period that we see in the fourth quarter. The second is the annual pay and benefit increases that begin to take effect in the first quarter and take full effect in the second quarter. Again, I would expect the spend rate as we go into next year begin to increase coming off of a low fourth quarter.
Year-over-year, though, given the wind down of legacy wireless, probably still year-over-year attractive within our OpEx model in the 20%-25% range, probably 20%-30% range of revenue, excuse me, being at the lower end of that range.
Ross, just maybe even shorter term, fourth to first, OpEx typically goes up because of the pay and benefit increase that Kevin described that kicks in in February, but also just fewer holiday and vacation days taken in first quarter compared with fourth. Fourth, we see the benefit from those holidays and vacation days, but sequentially then it tends to work against us in the first quarter. Okay, Ross, thanks for your questions. In general, we're going to stop here. Thank you for joining us. A replay of this call is available on our website. Good evening.
That does conclude today's call. Thank you all for your participation.