Good day, and welcome to the Texas Instruments Capital Management Strategy conference call. At this time, I would like to turn the conference over to Dave Pahl. Please go ahead, sir.
Thank you. Good morning, and thank you all for joining our conference call and allowing us to share an update on our capital management strategy with you. Kevin March, TI CFO, is with me today and provide details and to answer your questions. This call is being broadcast live over the web and can be accessed through our website at ti.com/ir. From the website, you'll be able to see our presentation slides today. A replay will be available through the web, as well as any relevant non-GAAP reconciliations. 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 presentation slides, as well as TI's most recent SEC filings for a more complete description. Today, we'll provide you an update of our capital management strategy.
Our goal is to provide insight into a few key areas, as well as review our 2014 performance. We had strong results in 2014, and it was a year that we believe was a preview of the performance we think TI can deliver in the years ahead. We gained share in our core markets of analog and embedded, delivered solid revenue growth, expanded our free cash flow margin to 27%, and returned more than 100% of our free cash flow to our shareholders. We believe that our performance is best viewed through the lens of our capital management strategy. The strategic elements of this strategy, taken together, contribute to our ability to grow, generate, and return cash to shareholders. With that as a backdrop, let me turn it over to Kevin to share the details.
Thanks, Dave, and good morning to everyone. We really appreciate you making the time to join us on this call today. Now, let me go ahead and get started. Today, I'm going to summarize TI's capital management strategy and our 2014 results. I'll outline our plans for expanded 300 millimeter analog manufacturing capacity, and then I'll review our free cash flow growth as well as our outlook for continued free cash flow growth. With that, let me begin with our capital management strategy and our 2014 results. Many of you are very familiar with TI, but I think it's worth noting again that we believe TI is in a unique class of companies that's able to grow, able to generate, and importantly, able to return significant cash to our shareholders for a very long time to come. Our business model has been carefully designed around several significant competitive advantages.
We quite simply have the broadest portfolio of analog and embedded products in the industry. We have firmly anchored ourselves with a low-cost manufacturing strategy that includes highly differentiated manufacturing technology. We have the broadest sales channel in the semiconductor industry today. All of which, when combined, results in diverse and long-lived market positions to deliver meaningful, high terminal value to our shareholders. As to our capital management strategy, it reflects our belief that free cash flow growth, particularly on a per share basis, is the most important performance measure to maximizing shareholder value in the long term. That free cash flow will only be valued if it's returned to shareholders or productively reinvested in the business, and that good execution and disciplined capital allocation are the most important responsibilities of business leaders today. I think our capital management strategy is nicely summed up by this graphic.
At its core, we enjoy a great business model because we focus on analog and embedded processing, what we believe to be the most attractive markets in semiconductors. Rather than just generate a lot of cash, we believe it's important to have that cash readily available for our business needs and our shareholders. To that end, we've chosen to employ an effective tax strategy that enables us to bring the majority of our cash back home. This results in us having a very strong balance sheet with well-funded pension plans and other liabilities, in turn maximizing our access to debt when the economics make sense. This further allows us to deploy cash in support of our competitive advantages, including investments in technology, manufacturing, extending our channel advantages, working capital, and when available, acquisitions.
Once these investment needs have been met, we return our excess cash to our shareholders in the form of dividends and stock repurchases, as well as periodically repaying debt. It's been a couple of years now since we originally introduced our capital management strategy and a year since we last updated it. Before I move on to the other topics I have to discuss this morning, let me just take a few minutes to provide an update on how we performed in 2014 against our capital management metrics. In the area of free cash flow generation, our target model is to generate or convert between 20% and 30% of our revenue into free cash flow. For 2014, the score was 27%.
As to cash on hand, our model is to keep 10% of our trailing 12 months revenue, plus our next 12 months expected dividends and debt repayment as cash on hand. At the end of the year, we had 95% of that target metric of cash on hand. As to cash on shore, our model is to keep at least 80% of our cash on shore. In fact, we ended the year with 82% of our cash on shore. For our pension obligations, our target model is to keep our pensions fully funded on a tax-efficient basis, and our global pension plans, when you take a look at year-end, were funded to the 97% level. Debt is, in fact, a part of our capital structure, and we expect it to continue to be when the economics make sense.
For 2014, we ended the year with $4.6 billion of total debt with an average coupon of 2.15%. It's interesting to note that we actually assumed some debt with the National acquisition, and in fact, of the debt that TI has actually issued itself, we've issued about $4 billion of debt of that 4.6 with an average coupon of 1.62%. In capital expenditures, our target is to keep or to average about 4% of revenue to be spent on capital each year, and last year we spent about 3% of revenue on capital. Our inventory model for customer services, for services to have about 105-115 days of inventory on hand, and we ended the year just a little bit above that with 117 days.
Our cash return target is to return 100% of our free cash flow, plus the proceeds from exercises, minus any cash that's used to retire debt. In 2014, we returned 115% of that target. On dividends, our model is to allocate about 50% of our trailing four years average free cash flow as a dividend budget to follow when we use dividends. In fact, in 2014, we worked out to about 52% of that amount. From repurchases, it's simply to take our cash return target minus what we use for dividends and return the balance of that in the form of repurchases, and that worked out to 123% in 2014. Speaking of dividends and repurchases, since 2004, we have steadily increased our dividend, with 2014 marking the 11th consecutive year of increases.
As I mentioned, our budget for dividends is to use about 50% of our trailing four years average free cash flow. Because we believe regular dividend increases are important to many shareholders, in years when this math may not suggest a dividend increase, we would still expect to recommend one. The reason is because this formula makes our dividend quite affordable. In fact, this formula resulted in our 2014 dividends equating to about 38% of our 2014 free cash flow, leaving us plenty of headroom to meet this obligation. On the subject of repurchases, 2014 marks the 11th year in a row of steadily declining shares outstanding as a result of our stock repurchases, with our total share count down 39% over this period. For 2014 itself, total shares outstanding dropped by 3.3%.
As I've discussed in the past, our plan is to steadily repurchase shares when we believe the company's intrinsic value exceeds its market value. The amount of spend that we allocate annually for repurchases is our free cash flow, plus proceeds from stock option exercises, minus amounts to be used for dividends and debt repayment in that year. As of the year-end 2014, we had a remaining stock repurchase authorization from our board of $3.2 billion. Shifting gears now, I'd like to update you on our 300-millimeter analog manufacturing plans. Late last year, we began implementing what we refer to as our 300-millimeter $8 billion analog plan. This plan begins with RFAB, which most of you are already familiar with. To remind you, when fully built out, RFAB will be able to support $5 billion in annual revenue. We already have the majority of equipment on hand.
In 2014, it became our largest revenue-generating factory, turning out $2 billion of analog revenue. The second element of our plan entails DMOS6. This factory was originally built to support our wireless products, but until recently, has only been supporting our embedded products. We are now in the process of converting capacity in this factory that will be able to support another $3 billion of 300-millimeter analog revenue. Qualification is already underway for initial production to begin at the end of 2015. When complete, the two factories combined will be able to support $8 billion of 300-millimeter analog revenue, which means when this capacity is fully used, over half of our analog business will be supported by internal 300-millimeter wafer production. Importantly, this plan fits comfortably within our 4% of revenue CapEx guidance. These are photos of the two factories that I'm talking about.
RFAB was opened in 2009 as the world's first dedicated 300-millimeter analog wafer fab. This is a big factory with 220,000 sq ft of clean room space, currently about 40% utilized, generating $2 billion of revenue in 2014 out of a fully loaded capability of $5 billion in analog revenue. Just a few miles down the road from RFAB, DMOS6 was opened eight years earlier in 2001 as our 300-millimeter fab supporting our former wireless products. It's only a bit smaller than RFAB, with about 85% of the clean room space or about 190,000 sq ft of clean room. Currently, about a quarter of its capacity is presently being used to support embedded, and as I mentioned, analog qualification is underway. When fully loaded, we expect it will generate an additional $3 billion of analog revenue.
In addition, we expect the close proximity of these two wafer fabs to one another will enable us to benefit from rapid learning cycles and allow a significant leverage of resources between the two factories, meaning bringing up our second 300-millimeter analog wafer fab should occur even more efficiently than our first one. Let me take you through the math of why 300-millimeter wafers really matter. This chart shows the relative sizes of the wafers to one another. A 300-millimeter wafer has about 2.25 times the surface area of a 200-millimeter wafer, and given the average size of an analog chip, allows us to produce about 2.3 times as many chips per wafer. To process the larger 300-millimeter wafer, it only costs about 1.4 times what it costs to process a 200-millimeter wafer.
When you do the math, this means the same chip processed on a 300-millimeter wafer costs about 61% as much as a chip processed on a 200-millimeter wafer, or about 40% less cost per chip. What's that mean to gross margin? If, for example, you take a device selling for $1 that is currently generating 60% gross margin, here's the 200-millimeter versus 300-millimeter math that results. The sales price is the same, no matter if it's sourced from 300 or 200 millimeter. The die cost on 300 millimeter declines 40%, from $0.20 to $0.12. Assembly, test, and other costs are unchanged at $0.20. The total cost of goods in this example declines from $0.40 to $0.32, driving gross profit margin up by eight points to 68%.
In 2014, we generated over $8 billion of analog revenue, with $2 billion of that, or about 25% of that revenue, coming from our fab. While we expect some incremental revenue growth in our existing 200-millimeter wafer fabs, the preponderance of our growth will come from 300-millimeter. When fully implemented, our new plan will produce an additional $6 billion of 300-millimeter analog revenue, for a total of $8 billion of internally sourced 300-millimeter analog revenue, meaning that more than half of our analog revenue will come from our own 300-millimeter factories. As I have previously mentioned, we expect this capacity improvement will all fit comfortably within our 4% of revenue CapEx model. Now let me switch topics to move onto a look at our free cash flow growth over the past few years and what our outlook is for free cash flow growth going forward.
Looking back over the last 10 years or so, we have seen essentially zero top-line growth as we have worked to reposition the portfolio into one made up of analog and embedded from one that was previously dominated by wireless. While analog and embedded were growing, given its size, the unwind of wireless completely dampened that growth, such that total TI revenue has been essentially flat throughout this period. Yet despite that flatlining of the top line, the product mix shift to analog and embedded, supported by the production shift to 300-millimeter and the significantly lower CapEx that we now enjoy as we engage in our opportunistic purchase of capacity, have all combined to drive our free cash flow as a percent of revenue from the mid-teens 10 years ago to 27% in the most recent year.
To say it another way, despite no top-line growth, but with substantial margin improvements and benefits from our CapEx strategy, we've been able to actually grow free cash flow at a 7% compound annual growth rate over the last 10 years. In addition to our portfolio and manufacturing shifts, as I mentioned earlier in the call, we employ a tax strategy that results in the preponderance of our cash being readily available, not only to invest in the business, but to return to our shareholders that cash that we don't need. As a result of consistent accretive buybacks over the past 10 years, we've reduced our shares outstanding by 39% over that same period of flatline revenue growth. The result has been to add another five points to free cash flow per share growth.
When you put it all together, despite no top-line growth, our product mix and CapEx strategies, combined with our share repurchases, have resulted in our shareholders enjoying a 13% compound annual growth in free cash flow per share. As we look into the future, what we get excited about is that the transition to analog and embedded is now complete. In fact, it completed in 2013, although 2013 itself still included meaningful amounts of wireless revenue, somewhat dampening our 2014 year-on-year growth comparison to 7%. Nonetheless, 2014 represents our first year-on-year growth since 2010. As we move into 2015, our year-over-year comparisons will no longer be dragged down by this legacy business. The real strength of our product portfolio of analog and embedded will now be driving TI's top-line growth without any further drag from our legacy wireless unwind.
Given a 10-year track record of 9% compound annual growth for analog and embedded, our free cash flow growth going forward will now have the added benefit of top-line revenue growth shining through at the company level. Just to summarize, looking back, two of our three levers for expanding free cash flow per share have been working for us, resulting in a 13% compound annual growth rate to free cash flow per share. Our first lever has actually been neutral in that we've experienced no top-line growth as we shifted our portfolio to analog and embedded. With our second lever, we've expanded free cash flow margin through a product mix, 300-millimeter production, and opportunistic capacity additions resulting in very low CapEx spending. With our third lever, we have made steady accretive stock repurchases resulting in 39% fewer shares outstanding.
Looking forward, we expect all three levers to be working for us as we work to continue to expand our free cash flow per share. We now expect top-line growth to resume. We expect free cash flow margin to continue to improve to 30% of revenue on a sustainable basis on continued 300-millimeter expansion and model levels of CapEx. We expect to continue share repurchases as long as the economics continue to make sense. Just to wrap up, we believe TI is in a unique class of companies able to grow, generate, and return cash to shareholders for a very long time to come. Our business model is uniquely designed around significant competitive advantages, the broadest portfolio of analog and embedded products in the industry, soundly anchored in low-cost manufactured and differentiated technology, with the broadest sales channel of any of our competitors.
All of which results in diverse and long-lived market positions yielding high terminal value for our shareholders. Looking forward, we see continued growth of free cash flow as the top line starts to help, the 300-millimeter capacity becomes a growing advantage, and our free cash flow margin expands to 30% of revenue on a sustained basis in good markets. 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 the opportunity to ask a question, please limit yourself to a single question. After, our response will provide you an opportunity for an additional follow-up. Operator?
Certainly. At this time, if you would like to ask a question, please press star and one on your touch-tone phone. You may withdraw your question at any time by pressing the pound key. Once again, to ask a question, please press star and one on your touch-tone phone. We will take our first question from Harlan Sur from J.P. Morgan.
Good morning. Thank you for hosting this call. On the DMOS6 conversion and the continued move into RFAB, is the plan to layer in primarily new analog revenue opportunities into both fabs, or are you planning on moving some existing 200 millimeter high volume products into these fabs as well to take advantage of the cost reductions as you outlined? How does the team allocate manufacturing between DMOS6 and RFAB on a go-forward basis?
Okay, Harlan, thanks for those questions. On the first, the plan that we've put in place is to support growth. Really, the vast majority of our incremental revenue will be built on 300 millimeter. It's not designed to try to optimize existing products. It just takes a long time to be able to move those products from one fab to the other. The second question is that how we allocate. In RFAB today, we actually have products that run from all four of our product lines, so High Volume Analog & Logic, Silicon Valley Analog, High Performance Analog and power, all run inside of there. Essentially what we do is we target specific technologies, or we call them process flows, into a factory. As an example, today, we've got enough of those process technologies in RFAB to support $2 billion.
We'll do a similar type of thing with DMOS6. You have a follow-on?
Thanks for that, Dave. As it relates to driving your free cash flow margin to the 30% level and as the company continues to build scale in its target markets, how should we think about your OpEx trajectory relative to your revenue growth? I know it's trended towards the lower end of your target of 20%-30% of revenues, it seems that your increasing scale alone would imply continued leverage on the OpEx. Is it fair to assume that TI grows its OpEx lower than its revenue growth over a multi-year period? Any way to try to help us quantify that?
Harlan, I'd say that certainly would be an objective, is to keep OpEx well contained. As we've talked in the past, we expect OpEx to run in the range of between 20%-30% of revenue. In weak markets, that OpEx will probably run up to the higher end, in stronger markets, more towards the lower end. In 2014, we averaged about 25% during the year, if I recall, about 23% during the second half of 2014. As we continue to grow going forward, our growth objectives are to continue to gain market share, as we have been doing with analog and embedded for multiple years now, which mean we continue to expect to outgrow those markets on a continuing basis.
To that end, we expect those markets, the semi markets, to continue to grow at about 2x the global GDP, about 2 times the global GDP, that's really on the back of increasing semiconductor content into electronics, especially in industrial and automotive. When you put all that together, we would naturally expect our revenue to be growing faster than our OpEx, keeping ourselves highly profitable and generating a lot of free cash flow.
Thank you, Harlan. We'll go to the next caller, please.
We'll take the next question from Stacy Rasgon from Bernstein.
Hi, guys. Thanks for taking my questions. First, I just want to see about the equipment assumptions for DMOS6. Do you already have that equipment in-house? If you don't, where are you planning to acquire it?
Yeah. Stacy, DMOS6, as you know, was built and originally supported our wireless business. The vast majority of that equipment is in place and has actually been depreciated. The actual tools that we need to convert it is actually fairly small. We've been active in the used equipment market over the last few years. Essentially, all the equipment that we need to support the full plan to $8 billion on 300 millimeter analog is in place today.
Got it. Thank you. That's helpful. I guess for my follow-up, I just wanted to verify, it does sound like you are, I guess, raising your commitment on free cash flow margin. You were sitting at 20%-30%. This is now a commitment to actually deliver sustainable 30%. Just to verify if that is the case, and can you give us some feeling of timeframe?
Yeah. Stacy, what we're talking about here is that we just raised the top end of our free cash flow conversion range to 30% last year. It's 20% or 30%, as you noted, is our stated range. We delivered 27%, and a lot of that's coming on the back of increasing 300 millimeter revenue coming through our portfolio. What we've described here today is that we continue to see that margin expanding, and we should be able to sustain 30% of free cash flow conversion in strong markets and solid markets. That's really what we're trying to characterize here today. We're not changing the range or trying to up the range. We're simply describing that we see a clear path to get to that 30% and to sustain that 30% when we get there in solid markets.
Okay. Thank you, Stacy, and we'll go to our next caller, please.
We'll take the next question from Ross Seymore from Deutsche Bank.
Hi, guys. Just a follow-up on that last one. I know you're not changing the range, Kevin, but can you just walk us through a little bit about how you view getting from what you just delivered at 27% free cash flow margin to the 30%? What are some of the levers to create that incremental three points?
Going forward, Ross, we really have three drivers to be thinking about. We now have top-line growth working for us, which we didn't have in the past. We have margin expansion that will continue as we have more 300 millimeter production making up our total production, as well as overall product mix, and then just really continue our share buyback. From a free cash flow per share standpoint, we've got those three things going. Certainly, just taking a look at the expansion of 300 millimeter as a portion of our revenue, that alone is going to be a solid contributor, not only to our profitability, but also our free cash flow generation as we drive that up to a higher % of revenue.
You have a follow on, Ross?
Great. Sure. As my one follow-up, any sort of thought about how linear this move can be to that $8 billion number? I noticed you didn't put any timetables on when you would have that full-on capacity rolled out, is it something that will be linear, or are there big chunks that we should think about, if there's any depreciation impact within that would be helpful as well.
There's really not any big chunks. I think as we ramped our fab, we did that in an orderly fashion. Really, the reason why we're qualifying and going through that process now is it takes time to qualify, it takes time to release the products. It obviously takes time for those products to get qualified and used in our customers' systems. We're not expecting any big moves or shifts, but just a continual improvement, much like you saw with the ramp of our fab.
Yeah, I might add, just to remind everybody, that we've had our fab in operation for a little over four years now, and we've gone from zero to $2 billion of analog revenue in that space during that period of time. Again, it is a multi-year trajectory that we're on to fill these factories up, and it's really going to be a function of how well the market performs. Clearly, we're performing well inside that market as we continue to steadily gain market share against our competitors.
That's right. Okay. Thank you, Ross. We'll go to the next caller, please.
We'll take the next question from Mark Lipacis from Jefferies.
Thanks for hosting the call and for taking my question. Kevin, on one of the slides, you showed that after $8 billion of capacity with the two factories, you'll have over 50% of manufacturing of your analog manufactured internally, which I guess implies a healthy slug of analog manufactured outside. My question is, can you just review where you are on the split right now, internal versus external? If you have such a big cost advantage, why not just manufacture everything inside internally?
Yeah. Mark, I would say that slide number one is 50% on 300 millimeter. The balance of that we expect to manufacture on 200 millimeter. As Kevin indicated in the presentation, we actually still have incremental room on 200 millimeter, although that we'll still see most of the revenue growth come on 300 millimeter. We will continue to use some outsourcing, but for analog, the vast majority of our wafers are built internally. Embedded processing will need some advanced CMOS, and as you know, we use all of that external. I think if you look at our total mix in 2014, I think less than 20% of our total wafers were sourced through foundries. Do you have a follow-up?
Yes. Thank you for that clarification. You guys have made several large acquisitions, it seems, in support of your goal to generate a lot of free cash flow and return it. How should we think about the role of large acquisitions to continue to achieve those growth and free cash flow goals going forward? What's the appetite for large acquisitions going forward? Thank you.
Mark, I would say our appetite is unchanged from the past, meaning, it's more a function of how we look at acquisitions. As you pointed out, obviously, we look at them to generate copious amounts of free cash flow for us. The real metric that we use, first off, the preference that we'll have for acquisitions will be as it has been in the past, and that is focused on analog, for continuing to enrich our analog portfolio and strengthen our position in the analog market. Beyond that, as we take a look at acquisitions, the most important thing after it being an analog is does it strategically make sense for us, be in markets that are attractive that we want to be involved in, and then does it mathematically make sense? Meaning the numbers have to work.
By that, the measures that we use is that the acquisition has to be ROIC accretive. In other words, what we pay for it has to be accretive to our weighted average cost of capital within a four-year timeframe. We've got to be able to take a look at what we think we'd have to pay for an acquisition, run that math, and see if it's actually accretive from that standpoint. That's the metric that we use. That's the metric we'll keep on using going forward, and when we see an opportunity present itself, we'll make an appropriate decision at that time.
Okay. Thank you, Mark. We'll go to the next caller, please.
We'll take the next question from Vivek Arya from Bank of America Merrill Lynch.
Thank you for taking my call. You used the phrase "in good markets," and you mentioned return to top-line growth. Last year, your sales grew about 6.9% year-on-year. Is that the kind of top-line growth, Kevin, you need to sustain this 30% free cash flow to sales?
I think that rate of top-line growth will certainly get us there. Again, we have to continue to convert more of our revenue onto 300 millimeter as we grow, that'll come through. Certainly that kind of rate will help get us there. Underneath that, keep in mind, analog and embedded processing combined grew considerably faster than the total company. That's what we're counting on as we go forward, that we continue to benefit from the success of those two segments outgrowing their respective markets and driving top-line growth. Ideally, the answer is yes, that was a pretty strong market last year. You saw us step from 24% free cash flow in 2013 to 27% in 2014. If we continue to enjoy a strong market, we'll see ourselves continue to step up towards that 30% and then be able to sustain it.
Got it. As a follow-up, since you decided to exit a lot of the wireless exposure, gross margins have expanded by 600, 700 basis points or so, but that's also the timeframe in which you have started to use a lot more of your 300 millimeter capability. Is it possible to quantify how much gross margin benefit you have achieved from 300 millimeter already?
I think you could probably do some backwards calculations on the example I showed on that slide that had the math of the cost benefit we get from 200 millimeter. You can probably work that out. The fact of the matter is, as we see more and more 300 millimeter sourced analog revenues running through our portfolio, we will expect to see that gross margin continue to expand and expand to the 60% and beyond level.
Yep. It has been a nice driver. When you're doing that math, you'll see that. Thanks, Vivek, we'll go to the next caller, please.
We'll take the next question from William Stein from SunTrust.
Thanks for taking my question. A moment ago, you were asked about your appetite for acquisitions, and I'd like to ask it maybe in a bit more of a capital-focused way. Your cash flow generation is quite consistent. Even in the 2009 downturn, you generated a lot of cash. Yet your gross leverage is now below one turn, your net leverage is about 0.2 turns, your interest coverage is about 50 times. Why not consider taking a more active stance on the capital structure of the business, raise more debt to do something to accelerate earnings growth and cash flow growth?
Yeah, William, again, our focus is on what we can do to accelerate free cash flow growth. To that end, when it makes sense, we do use debt in the capital structure, and it certainly made sense for us to go ahead and take on debt. We previously had no debt, by the way. When we acquired National Semiconductor, we took on a significant level of debt for exactly that reason, and that is to incrementally accelerate our rate of free cash flow generation. To that end, as we look forward, when we find opportunities that will allow us to continue to accelerate free cash flow generation, we will certainly entertain the idea of including additional debt on the balance sheet in support of that objective.
Great. Thank you.
Follow on, Will?
Sure. Actually, I'd like to ask about end markets. I know it's not the topic of today's call, but I believe you have meaningful exposure to wireless infrastructure. Can you remind us approximately what that exposure is and whether you see that as a growth market this year for the company?
Yeah. Our communications equipment end market was 17% of our revenue. That would be inclusive of things like wireless infrastructure, which is the largest inside of there, but also telecom infrastructure, enterprise switching, and residential gateway. We saw strong growth in that market this past year, and we benefited from that. I'll also make the comment that we've actually got a very broad exposure of types of products that we sell into that market. In fact, we actually sell more analog product into that market than we do embedded processing. I think what we've heard from others in the market that there's a reason to believe that growth could continue into this year. I think that when we look at it, certainly it's always been a very choppy market, and they don't tell us ahead of time when things will roll over.
I think we'll stay optimistic and then just react when we see things that change. Okay, thanks, Will. We'll go to the next caller, please.
The next question we'll take from Ambrish Srivastava from BMO Capital Markets.
Thank you. Pretty cogent presentation, guys. I just had a couple of questions. Number one, on the gross margin incremental fall-through, if memory serves me correct, it used to be 75%. What is the right way to think about that fall-through rate now as 300 begins to ramp?
Yeah, Ambrish, we've gone back and checked that recently, it really hasn't changed that much. That's still probably a pretty good model over a multi-quarter period for you to be thinking about.
Okay. On the share gain side in analog, all the data is out, you guys gain share again, albeit small. On the embedded side, Dave, do you have a read on where shares stands for you versus the collective market? Thank you.
Yeah, Ambrish. We actually gained share in both analog and embedded processing for the year. I think that we gained about 40 to 50 basis points in analog, probably a little bit stronger than that in embedded. Our share is about 15% in the embedded market and just a little bit over 18% inside of analog. Maybe I'll just also just point out that those share gains have been consistent over periods of time, we really believe that those share gains are driven from the broad portfolio that we've got, the differentiated technology that we can build on our manufacturing assets that give us a cost advantage through our broad sales channels. I think all those things are working together to continue to drive share gains, collectively, those are very hard for our competition to replicate.
Thanks, Ambrish, and we'll go to the next caller, please.
I will take the next question from CJ Muse from Evercore ISI.
Great. Thanks for taking my question and hosting this call. I think we all understand the sustainable or multi-year delta between D&A and CapEx. What I'm trying to understand is how we should think about as gross margins move higher, what are the implications to free cash flow? I guess, put another way, how should we think about incremental free cash flow margins as you start removing some of these underutilization charges, et cetera?
Yeah, CJ, remember, the underutilization charges are really just an accounting charge, so I wouldn't go there. I'd go back to the thought process that Dave talked about, which is DMOS6 is now coming into play for us for 300-millimeter analog revenue growth. As I mentioned in my prepared remarks, we expect to have qualified by the end of this year with products coming out of that factory. As Dave commented, that's pretty much a fully depreciated factory. That equipment was first put in there at the beginning of 2001, and we depreciate pretty much on a straight-line five-year schedule. You can imagine there is not much depreciation cost that's going to be heard in GPM, so that's why GPM margins will go up.
Because we've already got almost all the equipment we need, there's relatively little CapEx and therefore free cash flow generation will be extremely robust coming off that. Rather than go through thinking about a free cash flow fall through, so to speak, I think it's more important to think about the free cash flow journey that we're on, working our way from the most recent 27% for 2014 to get into a sustained 30% as we move and grow revenue in the next couple of years.
That's helpful. I guess as a quick follow-up there, clearly there's still some cyclicality left in semis, I think. 30% in good markets, how do you think about free cash flow margin throughout a cycle? Is it still that kind of 20%-30% range? Is it now 25%-30%? How should we be thinking about that?
It's going to really come down to what you want to characterize as a cycle. If you go back to 2009, that was a pretty hard down cycle, and I think we slipped into the upper teens back then, some of them below 20. I'm certainly hoping we don't ever see one of those cycles again. I would expect that we can operate inside the 20%-30% range that we've been talking about. As we get more and more 300 millimeter, we begin to move that bar generally in an upward direction.
Even as the cycle may want to repeat itself, we should be able to sustainably generate higher levels of free cash flow versus what we might have done in prior cycles, just given the economics of the portfolio and the shift to 300-millimeter analog, our CapEx strategy, the fact that we could use this equipment for many years, all those things just provide us tailwind on our free cash flow generation. I'd just add to the robustness of the business model. I think in the fourth quarter of 2008, during the darkest days of that economic downturn, I think we still generated around $1 billion of free cash flow, most of that coming out of working capital. I think it's just a good highlight of how robust the business model is, even in very difficult times like that.
Thank you, CJ, and we'll go to the next caller, please.
I will take the next question from Blayne Curtis from Barclays.
Hey, guys. Thanks for holding this call. I just wanted to ask, in the scenario you have the 50% mix in some time frame in the future, obviously shows revenue growth. What are the opportunities, let's just say in a more modest growth environment, how much of the 200-millimeter products you have today could actually be switched over to 300-millimeter in a more modest growth scenario?
Blayne, our plans really aren't to shift it from 200 millimeter to 300. It's really there for growth purposes. Again, when you look at our portfolio and just the tens of thousands of products that we have and the diversity of our customer base, it just takes a lot of time and energy to move those products. The second thing I'll point out is when we release new products today, we usually multi-source them. We'll release them perhaps into a 300-millimeter fab as well as a 200-millimeter fab, and that just gives us flexibility for loading and ensuring that we can meet demand as it comes in. Again, the objective is to support growth. Do you have a follow-on?
I did. When you look, you're way ahead of the competition in terms of 300 millimeter, the cost savings that you're showing, the 68% gross margin. I know in the core analog markets, probably hard to win on price. When you look at some of the markets that you've sort of backed away from, maybe more price sensitive. Given this advantage, how do you look at the markets and maximizing actually just cash flow $ versus, it looks like the target you're zeroing in on is a 30% margin. Have you looked at some of these markets now with the advantage, and maybe could you gain some share in more volume markets?
Yeah. Blayne, I'd say that, we've been asked the same question, which is a very good one, in a lot of different ways. Essentially, our business model isn't changed because we've got the cost advantage. We really want to look and find places where we can find sustainable revenue growth, where we've got a differentiated position. If we don't have that and don't believe that we can do that, over time, we'll shift our R&D $ to places where we believe that we can. It goes back to the saying of, you're going to work really hard in good markets and bad markets, so choose good markets to operate in. That's where our focus has been. Thank you, and we'll go to the next caller, please.
We'll take the next question from Christopher Rolland from FBR Capital Markets.
Hey, guys. Thanks for the question. I think you guys have always said you had a $18 billion of revenue footprint, and I'm assuming this DMOS number was already in there. It's still at 18, or is it 21 now? Now that you guys have so much 300-millimeter capacity, are there opportunities to revisit your footprint and perhaps find some fabs to restructure some of the older facilities, and help accelerate gross margins?
Yeah. Chris, the DMOS6 was in those numbers, and essentially, the $18 billion number really hasn't changed that much over the last few years. What was the second part of your question?
Yeah. Now that you have so much 300-millimeter capacity, are there more opportunities to revisit your footprint and perhaps look at some older facilities that you guys have to transfer some of this stuff over?
I gotcha. Yeah. Again, we've got a fairly robust footprint of 200-millimeter factories. They're very cost-efficient. They produce a lot of free cash flow. We actually did close a couple of factories a couple of years ago. Just note that, I think one was around 35 years old. The other one was older than 40 years old. When we look at these assets, they really can be employed for decades, and produce a lot of cash. That's really how we look at it versus really just chasing the last profit dollar that we can out of the machine.
Yeah, let me just add to what Dave was saying. The illustration of how old those factories were is a good point. That means those products that were built in there were sold from those factories for a very long time, because once you qualify these analog parts into a factory, it really doesn't pay to move them to another factory. Eventually, demand for old parts will go away and loadings will drop to the point where maybe a factory doesn't generate the kind of free cash flow that we think we should be pursuing. That's more of what leads us to looking at our footprint as opposed to just trying to consolidate for the sake of consolidation.
Great. Also on the inventory side, you guys talked about being perhaps over your days target there. If you could remind us, how much more disti conversion do you have left to consignment, and where might these top out, and when might we get back into the range?
Yeah, Chris, I believe we are about 60% of our revenue now is going through distribution, and about 60% of that is now on what we call consignment inventory. It's that consignment inventory that is shifting the ownership of that inventory from where it was previously on the distributor's books to where it's now on our books. That is leading us to push up to the upper end of our model range of inventory days. As we talked about on the earnings release call, as this shift continues, and we expect it to continue over the next few years, meaning, a bigger percentage of revenue going through distribution and a larger portion of our inventory being in consignment arrangements, that is causing us to take a look at what is an ideal model for inventory days.
We'll review that over the next year as we watch this transition, and we'll probably refresh it at a later date. Right now, we're not going to be too hung up on that as we watch what's actually happening with customer behavior and their preference to move towards consignment inventory.
Okay, thanks, Chris, and we'll go to the next caller, please.
We'll take the next question from Tore Svanberg from Stifel.
Yes, thank you. The first question, let me just see if I get this right. Assume there's no recession, we should estimate that TI's free cash flow per share growth will be at least 13%. Then it's sort of up to us to determine the growth above and beyond that.
Yeah, I don't think you heard us give a forecast on free cash flow per share growth, Tore, but what you did hear us say is that over the last 10 years, our shareholders have been able to enjoy what's been a 13% CAGR without the benefit of top-line growth. There's been a lot of other changes going on inside that portfolio, as you know, as you've followed us for a long time. Clearly going forward, we'll have the benefit of top-line growth. Continued margin expansion as 300 millimeter becomes a bigger portion of our overall revenue generation capability. To the extent the economics continue to make sense, share buybacks. Those combined give us confidence that we'll continue to have meaningful free cash flow growth as we go into the future. I won't try to characterize exactly what that % might work out to.
I'm sure your model can help us get there, too.
Yeah.
Very good. As a follow-up, you mentioned 25% of analog at RFAB, and I think you said that all four product lines are now at RFAB. Is that distribution fairly even, or do some of the product lines have higher percentage than others? The reason I'm asking is because obviously there's a gross margin differential between, let's say, HPA and High Volume Analog.
Yeah, Tore. I don't actually have the percentage of each of the businesses. It really just will depend on the products that we put in there, and the decision on that is primarily driven by the process technologies that can be supported. Obviously, if we can build a 300 millimeter wafer versus a 200 millimeter wafer, we'll do that because of the cost advantage that we'll see. I'll also just add that it's not just determined on high volume products going into RFAB. We have some people that get confused because you can run high volumes, that that's the only thing that we put in there. In fact, we can run wafer lots with multiple products inside of them because of the automation of 300 millimeters. It makes it very efficient for even very small lot sizes.
We have a good diversity of products that run through there, and we can support quite a few different types of products. With that, I think we have time for one more caller.
Okay. We'll take our last question from Steve Smigie from Raymond James.
Great. Thanks a lot, gentlemen. Kevin, I was hoping you could address the tax strategy a little bit. Your tax rate is somewhere in the 20s. Other folks have it much lower, but you have much more onshore cash. In terms of figuring out what tax you want to go after in terms of tax holidays versus not tax holidays and having cash onshore, what are the decisions that you're making there in terms of those issues?
Okay, Steve, the decision really is pretty straightforward, and that is, we believe that, again, at the risk of sound like a broken record, the free cash flow per share is the best way to increase the value of the company, and that that free cash flow is only useful if you return it either to the business for investment or to your shareholders. To that end, our thinking starts right from how do we make sure we've got access to it to invest, and to the extent we have surplus cash, get it to our shareholders. That then informs us as to what our tax strategy should be. In this case, what it means is we wind up repatriating a sizable amount of our total earnings offshore.
As you model us going forward, because we have a strong preference to actually return surplus cash to our shareholders, you should model our future changes in profitability, assuming it's going to be brought back to the U.S. and taxed at a 35% marginal tax rate. Take our current tax rate, we've offered guidance for 2014 of our effective tax rate being about 30%. As you grow our profit in 2016 and beyond in your models, you should incrementally tax that profit at 35%, and you'll get a new weighted average effective tax rate as you pass through time. That's the best way to think about us from a tax standpoint.
Yeah. I'll point out that anyone that modeled that in the transition from 2014 to 2015 got right on top of that number, and there was no surprises there. Steve, you got a follow-on question?
Yeah, just real quick, on the stock buyback, I think you indicated that the way you determine to do that is when you feel the, I think the market cap falls below the intrinsic value. For the determination of intrinsic value, is that a discounted cash flow model, or how do you go about figuring out what you consider to be your intrinsic value?
Yeah, Steve, it's a discounted cash flow model, just to keep it real simple. When we do that, there's a sizable gap, which has kept us in the market buying back shares on behalf of our shareholders.
Okay, great. Well, thank you for joining us today. Before we go, I just wanted to highlight something that we've prepared for those of you who have not attended our previous capital management calls. On our website, we'll have posted later today a comprehensive summary of our capital management strategy. It will include the materials that we've covered today, as well as materials that we've covered in our previous calls, such as the summary of our business model, our tax strategy that we just touched on here, as well as multiple balance sheet items. Thank you again. A replay of this call and the slides that we have shown today will be available on our website. Good day.
This does conclude today's program. Thank you for your participation. You may