Good day, and welcome to the Texas Instruments Third Quarter 2014 Earnings Call. At this time, I
would like to turn the
conference over to Dave Paul. Please go ahead, sir.
Thank you. Good afternoon, and thank you for joining our Q3 2014 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 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 Safe Harbor statement contained in the earnings release published today as well as TI's most recent SEC filings for a more complete description. The Q3 was another quarter that marked strong progress. Our core businesses of analog and embedded processing grew again with combined revenue of 10% from a year ago.
Revenue of $3,500,000,000 came in solidly in the upper half of our expected range we had communicated to you in July. Earnings per share of $0.76 was at the top of our range as profitability was strong in the quarter. Our cash flow from operations was $1,400,000,000 We continue to believe that free cash flow growth is most important to maximizing shareholder value in the long term, especially on a per share basis. Free cash flow for the trailing 12 month period was almost 3 point $5,000,000,000 or 27 percent of revenue, consistent with our targeted range of 20% to 30% of revenue. This is a 300 basis point improvement from a year ago period.
We believe this reflects our improved product portfolio and the efficiencies of our manufacturing strategy, which includes our growing 300 millimeter output and purchasing assets ahead of demand at distressed prices. We also continue to believe that free cash flow will be valued only if it's returned to shareholders or productively invested in the business. Over the past 12 months, we've returned $4,200,000,000 of cash to investors through repurchases and dividends paid. In the Q3, TI revenue grew 8% from a year ago with growth in both analog and embedded processing. Analog revenue grew 11% from a year ago, led by power management.
High volume analog and logic, high performance analog and Silicon Valley analog also grew. Embedded processing grew 6% from a year ago due to microcontrollers, connectivity and processors, each of which grew by about the same amount. Embedded processing delivered its 8th quarter in a row of year on year growth. In our other segment, revenue was about even from a year ago as the decline in legacy wireless products was mostly offset by growth in DLP products. Turning to distribution.
Resales increased 10% from a year ago, consistent with our combined revenue growth in analog and embedded processing. Weeks of inventory were unchanged at a historically low level of just over 4.5 weeks. This level is low because we've structurally changed how inventory is managed in the distribution channel with our consignment program. This quarter, we continue to convert more of our distribution sales to consignment and now support about 55% of our distribution revenue on consignment, up about 10 percentage points from a year ago. With this program, inventory sits on TI's balance sheet and revenue is recognized when distributors pull products from our consignment inventory that is stored at distributors' locations.
This program minimizes changes in demand due to distribution inventory, channel inventory and most importantly allows us greater flexibility to meet customer demand. From an end market perspective, TI revenue growth from a year ago was due to communications equipment, industrial and automotive, each of which grew by about the same amount. Enterprise systems was also up, while revenue in personal electronics declined due to legacy wireless products. Now Kevin will review profitability, capital management and our outlook. Kevin?
Thanks, Dave, and good afternoon, everyone. Gross profit in the quarter was $2,040,000,000 or 58.4 percent of revenue. Gross profit increased 15% from the year ago quarter margin hit another new record. This gross profit reflects higher revenue and an improved product portfolio focused on analog and embedded processing that benefits from our efficient manufacturing strategy. Moving to operating expenses, combined R and D and SG and A expense of $795,000,000 was down $38,000,000 from a year ago.
The decline primarily reflects the targeted reductions in embedded processing in Japan that were previously announced as well as continued cost discipline across TI. Acquisition charges were $83,000,000 almost all of which were the ongoing amortization of intangibles, a non cash expense. Restructuring and other charges were a $9,000,000 benefit, primarily due to the gains from sales of assets. Operating profit was $1,180,000,000 or 33.6 percent of revenue. Operating profit was up 39% from the year ago quarter.
Net income in the 3rd quarter was $826,000,000 or $0.76 per share. Let me now comment on our capital management starting with our cash generation. Cash flow from operations was $1,380,000,000 in the quarter. Inventory days were 108, consistent with our model of 105 to 115 days. Capital expenditures were $103,000,000 in the quarter.
On a trailing 12 month basis, cash flow from operations was $3,820,000,000 up 17% from the same period a year ago. Trailing 12 month capital expenditures were $367,000,000 or 3% of revenue. As a reminder, our long term expectations for capital expenditures to be about 4% of revenue. Besides inexpensively adding capacity ahead of demand, we have focused on delivering higher levels of customer service. By combining this capacity with continued improvements in how we manage inventory, we're able to keep our lead times consistently low while continuing to deliver on time.
Free cash flow for the past 12 months was $3,450,000,000 or 27 percent of revenue. As we've said, we believe strong cash flow growth, particularly free cash flow growth is most important to maximizing shareholder value in the long term and will be valued only if it's returned to shareholders or productively reinvested in the business. In the Q3, TI paid $319,000,000 in dividends and repurchased $670,000,000 of our stock for total return of $989,000,000 As we've noted, our intent is to return all of our free cash flow plus any proceeds that we receive from the exercises of equity compensation minus net debt retirement. Total cash return in the past 12 months was $4,160,000,000 which was 9% higher than a year ago. In the quarter, we announced a 13% dividend increase in our quarterly cash dividend from $0.30 to $0.34 per share or $1.36 annualized.
This marked our 11th consecutive year of increasing dividends and our 52nd year of continuous dividend payments. Outstanding share count has reduced by 3.5% over the past 12 months and by 39% since the beginning of 2,005. These returns demonstrate our confidence in TI's 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 Q3 with $3,190,000,000 of cash in short term investments, up from $2,801,000,000 at the beginning of the quarter.
TI's U. S. Entities own 81 percent of our cash. Because our cash is largely onshore, it is readily available for a variety of uses, including paying dividends and repurchasing our stock. TI's orders in the quarter were $3,340,000,000 up 6% from a year ago and our book to bill ratio was 0.95.
This ratio would have been a bit higher except for the impact of the conversion to consignment of some products sold through distribution. While book to bill still would have been below 1, this is consistent with the pattern that we have seen for the past 4 years. Turning to our outlook, we expect TI revenue in the range of 3,130,000,000 dollars to $3,390,000,000 in the 4th quarter. At the middle of this range, revenue would increase 8% from a year ago. We expect 4th quarter earnings per share to be in the range of $0.64 to $0.74 Restructuring charges will be essentially nil.
Acquisition charges, which are non cash amortization charges, will remain about even and hold at about $80,000,000 to $85,000,000 per quarter for the next 5 years. Our expectation for our effective tax rate in 2014 remains about 28%. This is the tax rate you should use for the Q4. In summary, the Q3 demonstrates the growing strength of TI's business model. Our strategy is anchored in analog and embedded many markets.
We remain intent on excellence and execution, disciplined in allocating our capital and firm believers that free cash flow per indicator of shareholder value. 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 your questions, please limit yourself to a single question. After our response, we'll provide you an opportunity for additional follow-up. Danny?
Thank Our first question comes from John Pitzer with Credit Suisse. Please go ahead, sir.
Kevin, I guess my first question will be on OpEx. Clearly, you're well ahead of sort of your expected annualized savings of $130,000,000 with the restructuring actions. How do you think OpEx will trend sequentially into the December quarter? And if you look at R and D as a percent of rev now below 10%, is that just a structural change in the model? And how worried should I be at what scale level you should be investing on R and D for future growth?
I understand that you're still spending absolutely a lot more money than most of your peers, but tough to see someone with your gross margins and R and D as a percent of revenue below 10%?
John, I think there were a couple in there. So let me see if I can remember them all. Starting with OpEx and trends, clearly OpEx was down quarter over quarter largely as a result of some of the restructuring actions we previously announced as well as just the ongoing cost discipline in all the various business across TI. As you look into the Q4, if you just go back and take a look at a year ago, that's probably your best gauge to take a look at, Q4 and even the first quarter for that matter. We typically see that OpEx is lower in the Q4 due to holidays and vacations.
In addition, this Q4, we expect a further benefit of another $15,000,000 of cost to come out as a result of the restructuring that we announced on embedded processing. That's part of that $130,000,000 of annualized savings that you were mentioning John. And in fact, we're probably running ahead of that $130,000,000 of annualized savings right now. We're probably come out beyond that. From the total R and D spend as a percent of revenue, again, I think what's more important is not how much you spend, but how well you spend it.
And I think the best indicator of how well you're spending is what's happening to your market share as a result of the products you're introducing into the market. In our case, if you take a look at 'six through 'thirteen, the last full year periods that we've reported on, our total R and D spend is up 77% over that period. But most importantly, that R and D has resulted in products where we have continually gained share. So again, I think, John, it's less question of how much you spend it and a lot more how well you spend it and spend it on things that customers really care about. Good.
John, do you have a follow-up?
Yes, guys. Quickly, just I know you don't give segment guidance for revenue. But when I look at the embedded business in the September quarter, it was up only about 1.1% sequentially, which is a little bit weaker than I had been modeling. And I'm kind of curious as to what you think might have driven that. And given that some of your peers have talked about sort of a slowdown in the second half of September from a macro perspective, did any of that play into the embedded business in the September quarter?
Thank you.
Okay. So let me take that, John. I think if you look at the growth rates between analog and embedded processing, you can see a lot of quarters, they won't grow the same. And I think if you look at both of those businesses, they have different end market exposures. And really, it's those end market exposures that will explain most of the differences in those numbers.
So and the second part of your question was, yes, just did basically the overall environment impacting the growth that we've seen there. And I'd just say no, that really isn't in those numbers. Basically, that forecast is given off of the orders that we receive from our customers. So thank you, John. Can we
go to the
next caller, please?
Yes, sir. Our next caller is Ambrish Srivastava with BMO Capital Markets.
Hi, thank you very much. Just on the overall demand environment, there is a lot of confusion out there, but it seems like things have slowed down. So just wondered your perspective in the past you guys have given us certain things that you look at such as cancellations and so on and so forth. So and also within that frame of how would how would you compare it to somebody who claims that they are sell through, so they see a slowdown before a quarter or so before others? And then I had a very quick follow-up.
Okay. So a couple of points in there, Ambrish. The first is that I just point out that all of our distributors provide us with a point of sale reporting and we get that information on a very regular basis. So regardless of how the revenue is recognized, we see what's going on from a distributor resale standpoint. The second thing is, you touched on the current environment, And we really want to be guarded on trying to take a position for the industry on where that's going.
But we can really point to just what we see. So if I kind of step through those, our inventory and our books is at 108 days is within our target model. If you look at our channel inventories, those were essentially unchanged from last quarter. They're at a very lean level at just over 4 and a half weeks. Our lead times have been consistent and I'd say low.
Our cancellations and reschedules remain very low, so nothing changing on that front. And then probably, as we have grown over these last few quarters, we've been able to support that growth and keep delivering our products on time at very, very high service levels. So our operations continue to execute at that high level. So and then I'll get to the last part of your question was what percentage of our revenue goes through distribution and what percentage is supported by consignment. So the first one is about 55% of our revenues go through the distribution channel and about 55% of that is supported on consignment.
And as we pointed out in our prepared remarks, that has increased by about 10% from a year ago as we've had a couple of conversions go on this year. And I would say that we're not done with that. We actually expect that additional conversions will happen over the next year or so. So we expect that percentage will actually increase. So, Ambrish, do you have a follow-up?
Yes. A quick follow-up, Dave. On the embedded operating margins, you guys have had a steady increase. What's the right way to think about how do you get to north of 20%? Is it just a matter of focusing on the right product mix and getting that ramped up or is it something beyond that?
Thank you, Dave.
Sure. Yes, Ambrish, I'll take that one. On embedded processing, clearly the margins have been continuously improving over these last few quarters and we'd expect further improvement. Recall, I just mentioned earlier in the call that we'd expect another $15,000,000 or so of cost to come out of the Q4 and most of that will come out of embedded processing. I would add that from this point going forward and we've talked about this last couple of quarters, the amount of spend inside embedded processing is sufficient.
It doesn't need to increase anymore. What really the focus is revenue growth and that's what it's really all about. And that entire team has been very focused on that. I believe they just recorded their 8th quarter in a row of year over year revenue growth. So they clearly are working on the right metrics to really make that business start to deliver profitability at the levels we expect
it to at TI. Okay. Thank you, Ambrish. We'll go to the next caller please.
Our next question is from Stacy Rasgon with Sanford Bernstein.
Hi, guys. Thanks for taking my questions. First, I want to dig into the embedded processor margins again just a little bit. You had about $26,000,000 in OpEx come out from Q2 to Q3. You had revenue in the group up a little bit, mostly flat.
Margins were only up a little bit. Given all of that cost in theory, I would have expected to come out of embedded processing. Why weren't the embedded processing margins up more? It doesn't seem like all of those cost savings actually flowed through the margins this quarter.
Actually, they all flowed through the margin, Stacy, but not all through embedded processing. Recall those actions incorporate both embedded processing as well as actions in Japan, which flows through our solutions as well. Even with that at the EP level, you did see about a 72% fall through on that Delta revenue. So you're seeing it come through, but do keep in mind that the Japan stuff spreads around some of the other businesses as well.
Got it, got it. That's helpful. That's helpful. For my follow-up, if I can dig
into the gross margins a little bit, I think both for this quarter and next quarter.
So for this quarter, terms of mix or other cost savings that pushed it up? And then for next quarter, you came in a little bit high. I guess, was that only the revenue upside or was there something else in terms of mix or other cost savings that pushed it up? And then for next quarter, you came in a little bit higher. In terms of mix or other cost savings that push it up?
And then for next quarter, depending on how far you're guiding OpEx down, it feels to me like you're guiding gross margins implicitly flat to up in Q4 in a quarter where revenues are down and typically your gross margins would be down in Q4 as well. So if you could give us any sort of feeling for what the gross margin drivers for next quarter ought to be in terms of mix offsetting revenue declines, that'd be very helpful.
Sure Stacy. I think that clearly the extra revenue came through and it was quite rich. We also had a slightly better mix of overall products that were shipped in the quarter. So that combined gave us that new record gross profit margin of 58.4%. As we look into Q4, I would not characterize our guidance to suggest that margins go up.
We will be reducing loading some in the 4th quarter on a seasonal basis. And so you can expect that margins will be down a bit, not very much, but down just a little bit. So depending on how you build your model there, I wouldn't assume margins increasing in the Q4.
Got it. Thank you,
Stacy. And we'll go to the next caller.
Our next caller is Harlan Schuhr with JPMorgan.
Thank you for taking my question and congrats on the solid quarterly execution. The markets have been focused on Europe and China as sources of potential weakness. At the same time, the team has done a solid job on diversification and maybe that's what it is, the diversity of 4 gs, a more stable revenue profile. But at the margin, can you just talk about the overall health by geographies? Have you seen any signs of weakening above the normal seasonal trends that you would normally expect during Q3 or Q4?
Yes. Sure, Harlan. I'd say if you just look at what happened regionally, year over year, Asia and Europe were both up, while the U. S. Was even and Japan was down a little bit.
And sequentially, all the regions were up with Asia and Europe up the most. And if you look specifically inside of China and how resales had done there, We had looked into that and I just say that resales really just continued to be solid as we closed out the quarter. So that's basically what we saw. Do you have a follow-up?
Yes, absolutely. Thank you for that. You gave us some trends in OpEx into the Q4. How should we think about OpEx as we move into the first half of next year?
Hardin, I won't comment on first half, but I will give you some on Q1. Your best way to analyze that is just to take a look at what happened from Q4 to Q1 of the most recent year coming out of 2013, 2012. We do have seasonal increases in OpEx as we go from 4th to 1st, typically because of an absence of vacations that we have in 4th, and the partial implementation of our annual pay and benefit increases occurring first. So they will you should expect those to be up and you can take a look at again this past year to get a rough idea as
to about how much that increase is by in percentage
terms. Yes. And I'd just add Harlan that the as we indicated before, Kevin pointed out that we're expecting about $15,000,000 more of savings in Q4 at that point, will be essentially complete on those announced actions. So we'll basically be just have those seasonal impacts that Kevin had mentioned. All right, thank you.
We'll go to the next caller, please.
Next caller is Craig Ellis with B. Riley.
Thanks for taking the question. I'll start off with a bigger picture question, Kevin. The company has done a very good job of looking forward and being proactive, taking action to either prune lower margin revenue streams such as wireless or identifying opportunities for cost savings such as the optimization program this year. As you look ahead, how much further room is there for a similar such optimization efforts? Are we really at a point where the TI model is getting very close to being optimized?
Yes, Craig, I think that as we look forward from an optimization standpoint, I think the ones you've talked about so far have been more about where we might have trimmed in areas that had opportunities perhaps not quite as attractive as other areas. I think perhaps one of the areas least appreciated as people look at TI on a go forward basis is the impact that continued Delta revenue on 300 millimeter analog is going to for us. The cost structure to build on 300 millimeter is very attractive. And so incremental revenue going forward on 300 millimeter, we expect to produce not only higher margins, but also higher levels of total cash on that Delta revenue.
Do you have a follow on, Craig?
Yes, I do. Thank you. Dave, you mentioned the strength in analog power management on a year on year basis. Was it similarly strong sequentially? Sequential basis, if you look
at our analog business, sequential basis, if you look at our analog business, it was up primarily due to high volume analog and logic as well as power management. But we did see growth really from all 4 businesses, high performance analog and Silicon Valley also grew. So you see a little bit of difference there sequentially, but again power continue to be strong in both comparisons. Okay, thank you, Craig. We'll go to the next caller please.
Our next question comes from Stephen Chen with UBS.
Hi. Thanks for taking my questions. I just wanted to drill down a little deeper on the embedded processing segment. And be sure clear between microcontrollers and I guess DSPs and application processors, could you talk about some of the trends that you saw in Q3, especially from any, I guess, communications equipment exposure that you have?
Okay. So embedded processing, as you know, grew 6% from a year ago. That was driven by microcontrollers, connectivity and processors, each growing by about the same amount. And again, as Kevin pointed out, that was the 8th quarter in a row of year over year growth. Sequentially, revenue was a little better than even, connectivity was up and processors and micro controllers were about even sequentially.
So, do you have a follow on, Stephen?
Yes. In terms of gross margins, I was just wondering if this year there's an opportunity for you to take advantage of building any strategic inventory for certain products that you know have good turnover and hence maybe further optimize gross margins a little bit in the seasonally slower quarter? Thanks.
Yes, Stephen, we do that on an ongoing basis. That's one of the ways that we're able to keep the lead time for the overwhelming majority of our parts to 6 weeks or less because that in fact we build to stock not necessarily build to order meaning we maintain a stock of inventory that we can always have short lead times for our customers. So from a strategic standpoint that is done. It does have a little bit of ebb and flow. As you point out, we will have lower modems in certain quarters.
It gives us a chance to build up on a few extra parts on that. But I don't think there's a whole lot of opportunity to do that just to impact the gross margins.
I think we're really doing that to impact customer service.
Yes. And I'd also add, Stephen, that if you take the combined way that we're managing inventory now and we're being very intentional with specific plus the open capacity that we've got and a combination of those two things, and we've really been able to execute at very high levels and deliver our products on time to customers. And so we think not only having low lead times, but also delivering product when you make a commitment to a customer, combination of those things are what customers would like to have. Okay, thank you. We'll move on to the next caller, please.
Next, we have Joe Moore with Morgan Stanley.
Great, thank you. Can you talk about where your fab utilizations are now just qualitatively and if there's still an underutilization charge as part of these numbers?
Joe, on the underutilization charges, the answer is yes, there are, but we're not really talking about those anymore. And the reason is really kind of straightforward. We found that analysts and some investors were on a conclusion that because we had underutilization that increase our utilization would be the maximum amount by which we could increase our overall profitability if we put revenue into those factories and that's really an incomplete answer. I'll remind you that our stated strategy is to invest in our capacity ahead of demand. By definition, that means we will have open capacity and underutilization costs.
Our purpose though for buying ahead of demand is to purchase that capacity with as little cash as possible. So the utilization charge is mostly a non cash charge and it tends to miss the point that our objective is to be maximizing free cash flow. So for that reason, you haven't really heard us commenting on or quantifying underutilization charges these past few quarters and unlikely to in the future unless there's some material change that causes a better understanding of what's going on.
Yes. And Joe, I'll just add. The last time we had stated our end to end capacity, it was in the $18,000,000,000 range. So if you want to get a qualitative feel of where our capacity is, you could annualize our quarterly revenues and put it against that number and it will get you into the zip code. So you have a follow on, Joe?
Yes. Thank you for that. In terms could you give us an update on the base station market? And I know you said last quarter that continued to be strong. Just anything you're seeing from that end market?
Well, yes, if I just look at the communications market overall and I would from a year on year standpoint, it was up primarily due to wireless infrastructure, while sequentially, it was down. And basically, we just saw that in as a seasonal pause in the Q4. So thank you very much.
And we
can go to the next caller please.
Our next caller is Vivek Arya with Bank of America Merrill Lynch. Hi, Viva.
Hello. Thank you for taking my question. You mentioned a very interesting thing on the improving output from your 300 millimeter fab. And I was wondering if you could give us some more color around what percentage of your starts are moving to that fab? And more importantly, what that implies for gross margins?
For example, in the past, you have mentioned about 75% fall through on gross margin. So as you increase the loading on 100 millimeter, what does that do to that number over the next several quarters or years?
Vivek, I don't know that I'm prepared to give you a detailed quantification of that nature. What I would just say is that we've talked in the past about just from the geometries alone that the chip cost itself is about 30% lower on 300 millimeter than it is on 200. Not to mention that you also get better yield results and so on, so you get a number of different benefits. So what we're seeing happening is proportionally more and more new products that we're releasing are being released onto 300 millimeter, which means as we look out over the next year, 2, 3, 4 years, more and more of our revenue will be sourced off a 300 millimeter, which inherently means it's got lower cost, which should allow us to incrementally increase our gross margins as well as importantly our free cash flow. Beyond that, I don't have any specifics I'll give you at this point in time, but I do want to introduce that concept to people because I think it's being underestimated as to how important that is to TI's continued progress.
Got it.
Follow on Vivek?
Yes. Dave, the question is on the end market mix. I believe you mentioned comps was strong on a year on year basis. I was wondering if you could just give us your end market mix overall for Q3 and how do you see those trends playing out in Q4? Thank you.
Sure. Yes. So on a year on year basis, our revenue growth was due to comms equipment, industrial and automotive, as I mentioned before, each of which grew and contributed to that growth by about the same amount. Our enterprise systems was also up, while revenue in personal electronics declined due to legacy wireless products. So if you kind of go down one step in each side of inside of each one of those, Industrial, we've got about a dozen sectors inside of industrial.
And as we've talked to different investors over the last few months and as we're traveling around, it becomes clear to us that our definition of industrial probably isn't as well understood. And we've got years ago, we used to define it as what it wasn't, meaning it wasn't comms equipment or wasn't automotive or it wasn't other things. But now we've got very intentional definitions. We've got about a dozen different sectors that make that But now we've got very intentional definitions. We've got about a dozen different sectors that make that up.
So things like appliances and building automation and displays and point of sale products, factory automation, industrial transportation, medical healthcare, fitness, lighting, motor drives, power infrastructure, things like that. So very, very broad, but very specific. So when you look on a year on year basis, we had growth in nearly all of those sectors, with most of the sectors actually providing double digit growth. It was actually led by medical and healthcare fitness and factory automation. But again, we saw that growth very broad based.
Automotive, we've got 5 sectors inside of that. And I'll also just point that those sectors are actually out on our website if people are interested. But we've got double digit growth in all the sectors, but led by passive safety and infotainment. Tom's equipment, as I mentioned, was up due to wireless infrastructure. Personal electronics was down due to legacy wireless products.
And if you take that out, we did see growth in PC and Notebooks and mobile phones, and that was partially offset by some weakness that we saw in gaming. And then the enterprise systems, that's part of that is where our DLP products will sit, and that was up due to projectors. So, I'll stop there. So thank you, Vivek. We have a follow-up question?
That was my follow-up.
Yes. Thank you very much. We'll go to the next caller. Thanks so much, Vivek.
Our next caller is Mark Lipacis with Jefferies.
Thanks for taking my question. The first question is, historically, when you saw lead time stretch for some of your competitors, not even if your own lead time stayed stable, you would see not only double ordering at your competitors, but also double ordering at Texas Instruments as well. And I'm wondering if you think something has changed structurally with how you guys are running the business or how the industry operates that might have changed that dynamic?
Yes. Mark, I don't know if I'll speak for the industry, but I'll speak for TI and kind of go back to the comments I made earlier. As a result of us taking an approach whereby we intentionally invest in capacity before we need it, and therefore always have more what's needed for that quarter's business, it allows us to be able to build inventory in a much more thoughtful fashion than we have in the past. The consequence of that has been that not only have we been able to maintain lead times for the preponderance of our products at 6 weeks or less for several years now, we've also been able to maintain a very high level of on time delivery to our customers against those lead times. So when you put that together, what you really got for TI is just a different way whereby we are managing our manufacturing footprint and how we utilize that manufacturing footprint for purposes of managing inventory.
As it relates to double ordering, again, one of the biggest things that tends to be evident when double orderings begins to pop up is you start seeing cancellations or reschedulings beginning to ramp up. We do not see that and we have not seen that for some time now. So I think the strategy whereby we have capacity ahead of demand and we use that thoughtfully to make sure that we can maintain relatively low lead times is serving well to satisfy our customers, at least as it relates to demand for our products. You have a follow-up, Mark?
Yes, I do. And that's helpful, Kevin. And the follow-up is actually on that topic about the lead times. I think earlier in the script you said you're delivering you're focused on delivering higher levels of customer service. And then I think you said you followed that up with saying we're keeping lead times consistently low.
So I guess I'm just trying to reconcile when you say delivering higher levels of customer service, are you do you think that you have your lead times over maybe a longer period of time have come in that they're just structurally shorter? Or are you just delivering them more consistently at a shorter time. So I just try to reconcile when you said what is the higher level of customer service? What is the improvement been? Thank you.
Yes. Thanks for that clarification. I think it's a real good question. So when we refer to that, we look at a bunch of different customer service metrics. One is just how many line items are we shipping on time.
And we've had some very intentional initiatives that have been set up to be able to ensure that that is a good solid number and we're taking advantage of both the capacity and our inventory to ensure that we're consistently delivering on time. And so that's one of the metrics that we look at and that's what we mean by customer service. It's not really changing those lead times, but keeping them stable and then delivering products when we say that we'll deliver them. It's really the combination of those two things. So thanks, Mark.
And we'll go to the next caller, please.
Our next question comes from Blayne Curtis with Barclays.
Hey, good afternoon. Thanks for taking my question. Just going back, Kevin, on gross margin, if you could just go through what drove the upside versus your guidance? And then you're obviously sustaining that going forward. Is there anything that doesn't repeat in such that it sounds like you can grow it off this base.
Just wanted to make sure I heard that right.
Yes. Blayne, the guidance actually we're up a little from guidance, not a huge amount. But really what we had was a little bit more revenue come through. But importantly, we also had a slightly better mix of product that we ship to our customer. And so combined that gave us a slightly higher gross margin than previously expected.
And then on a go forward basis, again that mix continues to be in our favor as we go forward. And so consequently, as we look into Q4, our margins will continue to be quite strong. And importantly, because of the rather unique situation that we have with 300 millimeter analog manufacturing capacity, we have significant opportunity to continue to drive not just improve margins, but importantly improve cash flow. Again, I'll remind you that some of the things that we're doing, we've got 300 millimeter analog. I do want to make that point if I haven't yet.
We've got a lot more of our revenue coming from catalog parts, which inherently tend to have more attractive margin characteristics. We have increasing portions of our revenue coming from industrial and automotive spaces. And those three concepts are pretty important. Catalog parts, industrial, automotive tend to have very nice long revenue streams, which means we have plenty of time to figure out and maximize our cost from a production standpoint and maximize our revenue and cash flow offset. We've got increasing customer diversity, so our dependence on any one customer and therefore the risk to our loadings or our cash flow generation are diminished.
And perhaps also very importantly, we have a sales force that we estimate to be 3 to 4 times larger than our nearest competitors and those folks are becoming increasingly productive. And so we're beginning to get a lot more revenue per salesperson, which means we can grow revenue without having to grow OpEx as fast as revenue is growing. So those things combined not only will help our margins, but also definitely help our cash flow.
Okay. Blayne, do you have a follow-up question?
Yes, thanks. So just for a big picture level, I think you already answered this, but I just wanted to make sure I missed anything. We've had seen some companies reference pockets of weaknesses. I would say your results are normal, if not even better than normal. Are you seeing any areas that are weak and you're offsetting that with share gains or strength elsewhere?
Just curious your perspective or are you not seeing any weakness at all anywhere?
Well, yes, I'll let you draw the conclusions from the numbers. I think that obviously the results are solid. I think the outlook is consistent with the orders and our visibility into what customers are telling us they want from a consignment standpoint. And yes, so and from a share gain standpoint, one particular quarter that you're picking up share. But I think when you look back over a year, you can see that trend.
So we believe in our numbers, there's share gains, but I'd be careful on any 1 quarter to be able to point that out. So thank you, Blayne. And we'll go to the next caller, please.
Our next question comes from Ross Seymore with Deutsche Bank.
Hi, guys. Thanks for letting me ask a question. I guess the one segment nobody's asked about thus far is is other. It did a bit better than I had expected. Dave, you broke down what was the year over year and sequentials with the sub buckets in the other two segments.
Can you do that for the other segment as well? And if it is DLP, give us a little description on is there some seasonality in that that we need to
the decline in legacy wireless products and that was mostly offset by the growth in DLP products. On On a sequential basis, well, let me just finish year on year. If you look beyond that, calculators as well as the custom ASIC business were up and royalties were down just slightly. So sequentially, revenue was up due to growth in calculators and DLP. Royalties were flat and custom ASIC was down a little bit.
So I'd say that all those businesses are stronger in the back half of the year in Q3, but I would say that, that business is just executing well. When we look at that business longer term, I described that it's got several wildcard growth opportunities. If you look at most of the business today, it will be centered up in the front projectors that you'll typically see in offices and schools and government buildings. But from a broader standpoint, it's got some opportunities inside of automotive as well as some embedded opportunities there as well.
You mentioned the Pico as well.
Yes, the Pico projectors as well. So thanks, Ross. Do you have a follow-up?
I do. Kevin, you were helpful in giving us some bounds around the OpEx and looking at prior years and how things have dropped seasonally due to vacations, etcetera, in both the etcetera, in both the 4th and then the first quarter doing the other direction. Can you give us a little bit of a harder number on what that typical percentage change is quarter to quarter? Looking back in the past, you guys have had a number of restructuring programs that make it a little bit difficult to parse out what seasonality and what is restructuring driven in both the 4th and the Q1? Thank you.
Yes, Ross, I actually pointed to the last 4th to 1st transition with the idea that that's a good one for you to look at. Because in fact, the timing of the announcement for the embedded process in Japan restructuring, you may remember we announced that in January of this past year. So we hadn't even begun the activities yet inside the Q1. So the underlying cost in there is fairly clean from a comparative standpoint between 4th and 1st. So I'd use that as your figure of merit for trying to figure out where we go in 20 15.
Okay. Thanks, Ross. That was helpful clarification. And we'll go to the next caller, please.
Our next caller is Doug Freeman with RBC Capital.
Great. Thanks for taking my hi, guys. Thanks for taking my question. Congrats on the real If I could, Kevin, maybe attack a really high level one. You guys have been in the past very acquisitive, so much so that you really rebuilt your whole business model through acquisitions.
Can you maybe talk about your appetite to maybe add some debt onto your already leveraged balance sheet? And what it would take for you to look at another significant
deal? Okay, Doug. It sounds like maybe there's 2 ideas in your question there, 1 on M and A and 1 on debt. Clearly, on the M and A front, as you point out, we have been quite willing to take on M and A activities when it makes sense. And our definition of what makes sense is firstly has to be a strategic fit, meaning that it really has to make sense for where we want to take the company.
As we look at potential M and A in the future, it's most likely to be biased towards analog and probably towards items with catalog parts that service the industrial or automotive spaces. There may be others, but those are the likely areas that we would look at. If it turns out that it checks off that strategic fit standpoint, then of course, we have to take a look at the numbers and make sure that the price that we would pay that's what I mean by that is not necessarily what its market cap is, but the market we probably have to pay on top of that, that that total cost would be accretive to our WACC, our weighted average cost of capital within a 3 to 4 year timeframe. So there's several tests that we go through and we'll make a termination on acquisition. As it relates to debt, as you know, we have taken on debt to support acquisitions in the past.
We think having the balance sheet available to support those initiatives are important and a worthwhile thing for us to maintain for maximum flexibility as we move forward. And so as we look out in time and even if you look at the last couple of years, while we have been issuing new debt, we've also been retiring older debt and retiring a little bit faster than what we've been issuing so that we're taking our total debt levels down. And that's having a result of generally slowly opening the balance sheet back up to make itself available for any other opportunity that might present itself at some point in the future. So that's how we're looking at M and A and that's how debt kind of rolls inside that book.
Great. Thanks for
all that color. Yes, I guess if I could, can you guys pretty much along the same ideas on the balance sheet and how you look at cash return. I might have missed it. Did you happen to give out the share price at which you bought back shares this quarter? And would you moderate going forward, given the share price pullback that we just saw, how we should think about share count going forward?
Let's see. We bought back $670,000,000 worth of shares this quarter, total of 14,100,000 shares. So that's an average price of about $47.62 I would point out that over the last since the end of 2004, beginning of 2,005, we reduced our total share count by 39
percent. That
the average price of those shares that we bought over that time, I believe, were about $30.62 give or take. When we look to buy back shares as we've talked about in the past when the intrinsic value we believe is higher than what the market value is then it makes sense to buy back on a steady, if you will, dollar cost averaging kind of basis. That's what we've done for many years and that's what you've seen us do here recently. In fact, the recent pullback in price is just allowing us to buy back a few more shares with the same number of dollars. So that's kind of around the world look at share repurchase, if you will.
I answered your question on that, Doug.
Yes. Okay. Thank you, Doug. And we'll go to the next caller, please.
Following Doug, we have Will Stein with SunTrust.
Great. Thank you for squeezing me in. I'm hoping you can quantify a comment that you made earlier about the portion of sales on 300 millimeter today and that that could increase over time. Kevin, could you talk a little bit about where that number is now and where you'd see it going over the next couple of
years? Yes. Will, I think we're going to try to just hold back and say it is proportionately more than it has been each period that goes by proportionately more. But I think that we'll hold a little while longer till we can give a more comprehensive look and a better understanding for everybody on how to think about that 300 going forward. I just do think that it's again at the highest level, it's a fairly straightforward computation, 30% lower die cost, which means about 15% lower total ship total IC cost.
And as we get proportionately more revenue going on there that will just incrementally benefit our margins and our cash flows as we go forward in time. And that is something I've become aware of and talking to investors in the last couple of quarters that is underappreciated on what's going on inside the portfolio itself.
And as my follow-up, another comment that was made earlier was about, I think, the relative strength of analog versus the embedded segment this quarter. And I think there was a comment about end markets, the end market mix helping. I'm wondering if you can clarify that. Maybe help us understand a little bit better. I know analog is very diversified relative to embedded, but maybe talk about if there were maybe some end markets that were a bit more challenging that affected the embedded segment a bit more?
Yes. I think if you look at Embedded, it will have a high exposure to really 3 different markets. It's communications equipment, industrial and automotive. And like you said, our analog business has a much broader exposure, including exposure to personal electronics. It has obviously good exposure to industrial and automotive as well.
But it's really those types of differences that will explain that performance. If you look at embedded Processing and you look back at the growth rates between the two businesses, you'll note that oftentimes they will grow at different rates. But if you look over that 8 quarter period that embedded processing kind of works out that way over a longer period of time. Okay. And then, kind of works out that way over a longer period of time.
So thank you, Will. And we'll go to the next caller, please.
Next question comes from Tore
Sandberg. Yes. Good afternoon, Dave and Kevin. My first question is on sort of the general environment. Q4 is obviously a seasonally down quarter anyway, so it's hard to get a read on what's going on.
But as we sort of or as you talk to your customers and we compare now, let's say, versus 12 months ago, would you classify the forecast in the environment as better, weaker or about the same?
Tory, I guess I'd probably say that it's about the same. I wouldn't classify it as better or weaker. And I say it's about the same in that, again, we have had the same lead times for several years and customers are ordering consistent with that generally speaking. You get some that need to expedite from time to time because they've under ordered or have a surprise demand on their end, but that was true a year ago and it's probably true today. And we continue to see customers manage their inventories of our products.
I'd say very lean, as was mentioned earlier in the distribution channel, we're at all time lows of 4.5 weeks of inventory. That's the same last quarter and this quarter, that's very low. From a customer standpoint, we see them again ordering lead time and consistent ordering patterns. So that tells us their order they're keeping their inventories very low. So I think overall, I'd probably characterize it as relatively the same kind of customer demeanor today as what we saw a year ago.
Yes. Thanks, Kevin. That's very helpful. My follow-up question, Dave, if I have one. Sure, sure, please.
Yes. So your analog growth 10% year over year, is that sort of the goal internally to continue to grow the analog business 10% every year? Or was there anything unusual here in the last 12 months that kept that growth rate so high?
Well, I would say that internally, we have very aggressive goals to grow the businesses. And I've just described that as we want to outgrow the market significantly in analog. So we've got all the business units that are focused up on that. We do look at the performance of the product lines every quarter, and we really look at a 3 year compounded annual growth rate and stack that up against all the other businesses, both inside of TI as well as competitors externally. And that's how we measure it.
So again, as I said earlier, that 10% we believe would represent some share gains, but be real cautious to look at any 1 quarter and draw the conclusion or try to dice out what percentage of that growth would be from share gains. So, okay, thank you, Tori. Operator, I think we've got time for one more caller.
Yes, sir. And we have a C. J. Muse with ISI Group. J.
Muse:]
Yes. Hi. Thank you for taking my question. I guess first question and to a clarification. On the OpEx outlook for Q4 down $15,000,000 is that all restructuring and then there's gravy on top of that in terms of seasonal savings or is that $15,000,000 include the seasonal savings?
C. J, the thing you need to do is take a look at a year ago, 3rd quarter, to Q4, there was a seasonal decrease. On top of that will be another $15,000,000 as the last of the EP restructuring is completed.
Okay. That's helpful. And then I guess second question, I was hoping you could talk about inventories downstream, clearly relatively healthy on your books, but would love to hear what your thoughts are in terms of in particular at the disty level?
Yes. Our inventories really from a day standpoint had remained unchanged at what I would just describe historically low levels at just a little over 4, 4.5 weeks. Again, that number is low because of the consignment programs that we've put in place. Beyond our distributors and then to our customers, we do have a good percentage of our revenues on consignment overall. And in fact, if you combine what we ship through distribution as well as what we ship to OEMs on consignment, that represents about 50% or about half of our revenues where we actually there is no inventory sitting in front of us in that manufacturing line for those inventories.
So we know that inventory number and it is 0. So we have very good visibility into that portion. And we do have a balance of the business that's kind of the classic of describing as a book ship type business where they give us an order, it goes on to a backlog and we ship it at lead time. So and there, we haven't seen any reports of large inventory pockets from customers. We're going through and racking up this quarter's results and we'll see what happens on that front, but we're not aware of any big pockets that are downstream.
So, okay, well, thank you very much, CJ, for your questions. Thank you all for joining us today and a replay of this call will be available on our website. Good evening.
Ladies and gentlemen, this does conclude today's conference. We appreciate everyone's participation.