Texas Instruments - Earnings Call - Q3 2011

Transcript

Speaker 4

Welcome to the Texas Instruments Third Quarter 2011 Earnings Conference. Today's conference is being recorded. At this time, I would like to turn the conference over to Braun Slaymaker. Please go ahead, sir.

Speaker 0

Good afternoon, and thank you for joining a Third Quarter 2011 Earnings Conference call. As usual, Kevin March, TI's CFO, is with me today. For any of you who missed the release, you can find it on our website at ti.com/ir. This call is being broadcast live over the web and can be accessed through TI's website. The replay will be available through the web. This call will include forward-looking statements that involve risks and uncertainties that could cause TI's results to differ materially from management's current expectations. We encourage you to review the safe harbor statement contained in the earnings release published today, as well as TI's most recent SEC filings for a more complete description. Our mid-quarter update to our outlook is scheduled this quarter for December 8. At that time, we expect to adjust the revenue and earnings guidance ranges as appropriate.

This is a complex earnings report, primarily because we have a soft demand environment and significant acquisitions with accounting implications. We've taken extra effort to provide clarity and transparency. In today's call, we'll give you an update on our progress with the National acquisition and explain the associated costs and accounting. We'll cover the demand environment for the third and fourth quarters, and we'll provide detailed transitions for earnings per share. On our website, we also provide you additional details to assist in understanding the combination TI and National. We have closed the National acquisition on September 3. In addition to the usual activities of integration, the first leg of a world tour has been completed through Asia. Our lead manager, Greg Lowe, and his team of regional business leaders have already held many meetings with the TI sales, with distributors, and most importantly, with customers.

The power of the combined portfolios is being confirmed. Distributors are committing more dedicated resources to TI and the stronger analog portfolio. Customer feedback is almost universal. Where National's design and position have been shrinking over the past three years due to their commercial posture, customers nonetheless love their products and are ready to immediately re-engage. The world tour will continue over the next three weeks as we aggressively work to show customers the benefits of our combined portfolios. Let's move now to the demand environment, which was weak in the third quarter. Revenue was flat sequentially in the quarter when historically it has grown on average about 7%. That being said, we were encouraged that revenue landed above the high end of our range of expectations due to a stronger-than-expected month of September.

Our book-to-bill ratio is below one for the quarter, and we expect revenues to again decline more than seasonally in the fourth quarter. Even with this outlook, we see conditions that suggest the bottoming process has begun. After a sharp drop in July, the rate of decline in orders slowed considerably in August and September. Lead times are short, and inventories at distributors and OEMs are lean. On average, our semiconductor product revenue in a fourth quarter would be about flat initially, and our total revenue would decline about 3% due to the seasonal pattern in calculator sales. This fourth quarter, excluding National Semiconductor, we expect our semiconductor revenue to decline about 7% at the middle of our range, total revenue to be down about 10%.

When including a full quarter of National Semiconductor's revenue, semiconductor revenue should be about flat at mid-range and the company down about 3% because of calculator seasonality. Our acquisition of National Semiconductor was an important step in our strategy to further strengthen our position in the analog semiconductor market. National Semiconductor's 12,000 products enhance our portfolio. TI had 30,000 analog products before the acquisition and will introduce about 500 new products in 2011. These additions to our portfolio meaningfully accelerate our product strategy. Similarly, the addition of 5,000 plus employees, many of whom have deep and hard-to-find expertise in analog design, analog process technology, packaging, and manufacturing will enhance our capabilities.

Finally, National Semiconductor's three factories have lots of room for growth, which means we can focus our energies on customers and on driving growth from this portfolio instead of the unproductive task of moving products between factories that creates a lot of overhead expense and customer disruption. We have included a supplemental chart on our site that details how much the acquisition contributed to our financial results. We closed the acquisition on September 23rd and therefore had seven days of its results consolidated into TI's and our analog segment's third quarter. The revenue contribution over the seven-day period was $18 million, and there was $2 million of associated operating profit consolidated into our analog segment under the organizational name Silicon Valley Analog.

Silicon Valley Analog, or SVA as we abbreviate it, is now a fourth component of our analog segment along with High-Performance Analog, Power Management, and High-Volume Analog & Logic. Of course, the $2 million of operating profit does not include acquisition-related charges, which are included in our other segment. Kevin will review the acquisition-related charges and the impacts of the acquisition on some of our balance sheet lines in a few minutes. Our website also includes a chart that provides a historical perspective of the combined financials to the beginning of 2010 and a chart that provides National July to September results for 2011. We thought the latter would help you establish a baseline for our performance as we move into the fourth quarter as a combined operation. In future quarters, we will not break out details of Silicon Valley Analog performance separately from the analog set of financial results.

Let's shift to the details of our third quarter results. Revenue declined 7% from a year ago and was flat sequentially. Revenue from our core product lines declined 3% from a year ago and declined 4% sequentially. The reason our products underperformed TI revenue overall in the sequential comparison was due to a rebound in DLP revenue, which is in the other segment, as payments from our Miho factory resumed in the third quarter following the shutdown after the Japan earthquake. Also, baseband revenue grew following our customers' inventory reduction in the first half of the year. Analog revenue declined 2% from a year ago and 2% sequentially. The decline from a year ago was most driven by lower High-Performance Analog revenue, while High-Volume Analog & Logic and Power Management were each about even. Sequentially, both HPA and Power Management were lower, while HVAL grew a few percent.

In Embedded Processing, revenue declined 7% from a year ago and 10% sequentially. Catalog products were the reason for the decline from a year ago, while communications infrastructure products were about even, and automotive products grew double digits. In the sequential comparisons, most of the declines also came from logic products, although communications infrastructure products were down double digits and automotive products were down a few points. In our wireless segment, revenue grew 24% from a year ago and grew 4% sequentially. Baseband products were the biggest factor in both comparisons. Baseband revenue of $63 million in the third quarter fell 40% from a year ago and grew 16% from the prior quarter. Outside of baseband products, revenue from OMAP applications processors grew, and revenue from connectivity products declined in both comparisons.

Our OMAP revenue should continue to benefit in the quarters ahead as our latest customer programs continue to ramp into production. Recently announced programs that are based on our OMAP 4 platform include Samsung's Nexus smartphones, which will be our OMAP production program based on Google's Ice Cream Sandwich reference platform. Also, Samsung's Galaxy S2 smartphone, Motorola's Droid Bionic and Droid RAZR smartphones, LG's Thrill smartphone, which is the first to feature 3D playback and capture, and the recently announced Amazon Fire Tablet. A great lineup of customer products with more to come. Revenue from our other segment fell 3% from a year ago due mostly to the sale of our cable modem product line in the quarter of 2010, as well as declines across product areas.

Sequentially, revenue in this segment grew 10%, mostly due to higher shipments of DLP products, as noted earlier, as well as from earthquake-related insurance proceeds. Turning to distribution, sales distribution channel declined a couple of percentage points compared with the prior quarter. Despite falling resales, distributors reduced inventory levels even faster, resulting in the reduction of a day of inventory. Now Kevin will review profitability and our outlook.

Speaker 5

Thanks, Ron, and good afternoon, everyone. Let me start by walking through some acquisition-related costs. We will continue to report our results on a GAAP basis. However, we will make a survey for you on costs associated with the acquisition, both in the near term as well as on an ongoing basis. We have broken out a line on our income statement called acquisition cost, where most of these charges have been consolidated. This includes items such as restructuring costs, transaction costs, retention bonuses, and amortization of intangibles. In the quarter, we had $147 million of acquisition costs. An important consideration for the next quarter is how inventory is recognized under the acquisition method of accounting for business combinations. Inventory that we initially see from National must be at fair value, how you generally mean selling prices.

Therefore, included in cost of revenue is the expensing of the write-up of this inventory to fair value. Although we are unable to consolidate this into the acquisition cost line, we will identify for you how much this cost impacts cost of revenue. In the third quarter, it was $7 million due to the limited amount of time that Silicon Valley Analog was part of. It will be much more significant in the fourth quarter now that we have a full quarter of shipments. The near-term impact is that sales of this inventory will effectively occur with little to no gross profit. This should continue to be an issue only for the fourth quarter when we will fully expense the remaining write-up amount. All of the costs of the inventory write-up will be carried in our Other segment.

Our Analog segment will carry the cost of this inventory at normal standard costs, which generally reflect the cost of production. Our intent is for you to be able to easily see the ongoing performance of TI plus National in our Analog segment. Instead of carrying the acquisition-related costs in Analog, we'll carry it in our Other segment. Let me provide a summary of the $154 million in acquisition-related costs for the quarter. They include $147 million of acquisition costs as identified on the income statement, plus $7 million of inventory write-up to fair value that is included in cost of revenue. Additionally, there was a negative $10 million as tax impact. All of this sums to $0.09 per share of costs associated with the acquisition. Growth effort was essentially unchanged compared with the second quarter.

Factory underutilization expense was higher as we lowered our production in response to weak end. To a lesser extent, and not atypical for a downturn, we scrapped some inventory of custom products in response to certain customer program cancellations. This was partially offset by a positive net contribution associated with the quake. Operating expenses of $783 million included $13 million associated with Silicon Valley Analog. Even so, operating expenses were reduced 6% sequentially, or $52 million, primarily due to reductions in the variable components of our compensation plans in response to the weaker environment. As we discussed at our mid-quarter update, we have adjusted down our estimate for TI's annual effective tax rate for 2011 to 25% due to lower expected taxable income.

The third quarter included catch-up adjustments for taxes since we had accrued at a higher rate in the first half of the year, part of the $10 million tax discrete acquisition-related charge I mentioned earlier. Net income in the third quarter was $601 million, or $0.51 per share. Again, in the EPS calculation, please note that accounting rules require that we allocate a portion of net income to any unvested restricted stock units on which we pay dividend equivalents. In the third quarter, the amount of net income excluded from the EPS calculated was $10 million. If you don't make this adjustment, you'll likely calculate the EPS to be a penny higher than we have reported.

Given the additional complexity of this quarter, it might be helpful if I walk through the varied ins and outs of the transition from $0.56 of earnings per share that we reported in the second quarter to the $0.51 reported for this third quarter. First, we had $0.08 per share of higher costs associated with the acquisition. That's $0.09 in the third quarter compared to a penny in the second quarter. In addition, gross profit was negatively impacted by about $0.05 associated with miscellaneous items, including the lower utilization level, inventory scrap, and product mix. We also had $0.02 of negative impact in other income and expense, mostly relating to litigation or some of the divested business. We had an additional penny of interest expense this quarter. On the positive side, we saw a net $0.04 beneficial transition associated with the earthquake and our short insurance proceeds for that event.

We also benefited by $0.03 from operating expense reductions. Our lower tax rate and good associated catch-up adjustments provided a $0.03 benefit. Finally, our lower share count contributed a penny. I'll leave most cash flow and balance sheet items for you to review in the release. However, let me make just a few comments. The company ended the quarter with $2.62 billion in cash. During the quarter, we issued $1.2 billion in commercial paper, which remains outstanding. Cash flow from operations was $1.14 billion. This was up $511 million from last quarter, mostly reflecting the payment of income taxes in the prior quarter. Capital expenditures were $193 million in the quarter, down from $276 million in the prior quarter. We used $450 million in the quarter to repurchase 14.1 million shares of TI Common Stock and pay dividends of $148 million.

We increased our inventory by $203 million in the quarter. The increase is due to $215 million of inventory that came to us as a result of the acquisition. Please remember that this amount includes the effect of the write-up of the National inventory to fair value. A calculation of days of inventory yields 103 days. However, I would caution you that this is a misleading number. Not only does it include the write-up of National inventory to fair value, it also includes the entire inventory received in the acquisition, even though the revenue or cost of revenue includes only for the seven days that business was operating as part of TI. It is probably more meaningful to back out the inventory and cost of revenue associated with the acquisition completely.

With that, our days of inventory on an apples-to-apples basis declined from 93 days at the end of the second quarter to 92 days at the end of the third quarter. Receivables and days sales outstanding should similarly be adjusted for the acquisition, with the net effect being that DSOs of 46 days would be 43 days after the adjustment. Orders of $3.07 billion in the quarter were down 15% sequentially. TI's book-to-bill ratio was 0.89 in the quarter compared with 1.04 in the second quarter. Turning to our outlook, we expect TI revenue in the range of $3.26 billion to $3.54 billion in the fourth quarter, or down 6% to up 2% sequentially. As a reminder, this includes a full quarter of SVA revenue compared with the record for the third quarter that included only seven days.

Also, please remember that the fourth quarter is when we typically see about a $100 million sequential decline in calculator revenue following the back-to-school period. We expect earnings per share to be in the range of $0.28 to $0.36. EPS will be negatively impacted by about $0.15 associated with the acquisition-related costs in the fourth quarter. This will include about $160 million of acquisition costs and additionally about $100 million included in cost of revenue. Also, note that the fourth quarter will include a full quarter of operating expense associated with the acquisition compared to seven days in the third quarter. Operating expenses will also reflect the non-recurrence of cap adjustments to our compensation. Together, these items will result in operating expenses increasing by $160 to $170 million.

Again, in consideration of the additional complexity, let me walk through our EPS transition from the $0.51 that we reported in the third quarter to the $0.32 that is at the middle of our fourth quarter range. First, acquisition-related charges of $0.15 in the fourth quarter will be about $0.06 more than we had in the third quarter. Lower revenue in the fourth quarter will impact EPS by $0.02. Lower factory utilization will impact us by about a penny. As I mentioned previously, we will have a full quarter of SVA operating expense, which will impact us about $0.07. Outside of the acquisition, operating expenses will go up by about a penny due to the non-recurrence of the third quarter catch-up adjustments to our variable compensation expense. Finally, we will see about $0.02 of negative impact from taxes in the fourth quarter.

This impact is due to the third quarter including the catch-up estimate that we do not expect to recur in the fourth quarter. Our estimate for 2011 R&D, capital expenditures, and depreciation are unchanged. It might be useful for me to provide some insight into our expectations for acquisition-related charges as we go forward. As a first estimate, after $250 million in the fifth quarter, we will be complete with the impact from the inventory write-up adjustment. As a result, we expect first quarter 2012 acquisition-related costs to decline to about $150 million, dropping to about $110 million in the second quarter. After that, these charges should decline about $10 million per quarter until we reach $80 million, which is the amortization of intangibles amount. This will continue for 8 to 10 years.

We will update you on our expectations as we move forward, and we will continue to make these charges available and visible for you as we report. In summary, the market environment continues to be difficult, yet we believe we have already seen early signs of stabilization. With our acquisition of National Semiconductor, we are taking a strategic step that will strengthen both our market position and our ability to serve our customers. The integration mass that is ahead of us will be a multi-year effort and will be challenging. We recognize this and are up for the job. In the end, we are confident that we will deliver attractive and growing returns on this acquisition investment for many years to come. With that, let me turn it back to Ron.

Speaker 0

Thanks, Kevin. I realize that we threw a lot of data at you in our prepared remarks. We do have a copy of these prepared marks already posted on our website for your convenience. Operator, you can now open the lines up for questions. In order to provide as many of you as possible an opportunity to ask your questions, please limit yourself to a single question. After our response, we will provide you an opportunity for an additional follow-up. Operator?

Speaker 4

Very good. We will go first to John Pitzer of Credit Suisse.

Speaker 0

Hello, John.

Speaker 3

Yeah, guys, can you hear me?

Speaker 0

Yes, we can. Go ahead.

Speaker 3

Yeah, quickly, guys. On the core selling revenue being down about 7% since last December, I was wondering if you could just give us a little bit of color of how that might play out between Analog, wireless, and Embedded.

Speaker 0

John, I'm not sure that we have the core number, I don't believe we broke out for the December quarter. I think what you heard us say was what core did for the third quarter. I don't have a separate forecast yet for you going forward into the December quarter, other than the consolidated forecast that we typically provide. Do you have a follow-up?

Speaker 3

Ron, I guess I'm just trying to get a sense of sort of the operating revenue in semis being analog versus maybe the data and business, OMAP and connectivity. Is there any color you can give me around the sequential expected growth relative to guidance for the December quarter?

Speaker 0

Again, John, that's typically what we tend to provide in guidance, the total revenue number. I think this quarter, what you were thinking about when you said core revenue was when we told you the difference that semis would you're in versus the total revenue because of the impact of calculators. I don't have a more detailed breakout across wireless, analog, or embedded processing. Do you have a follow-on?

Speaker 3

Quickly, just on the utilization front, do you think the December quarter is kind of the bottom in utilization? At what level are you bottoming here?

Speaker 5

Yeah, John, we do expect our utilization to decline further in the fourth quarter from third, just like third declined from second. We don't have a forecast as to whether or not it declines further in the quarter. That's going to be more a reflection of our outlook as to what first and especially second quarter revenues might look like, as it takes quite a bit of lead time to put material. We do expect utilization to drop a few more points in the fourth quarter from third, but we don't have a forecast beyond that.

Speaker 0

John, I think we even tried to explain back at mid-quarter that this is mostly a macro-driven downturn. Any time customers reset their production levels, typically they'll auto-reset or rationalize their inventory levels to that new low production. What we have seen and what we believe we've seen is that the customer's own inventory reset is largely behind us. At this point, our customers' production and therefore our own shipments will track with basically end demand, which is more macroeconomically driven. I think we have a reasonable view of the fourth quarter. Once you go beyond that, it really ties to what the macro economy does and which way that tends to take end demand for our customers' product and therefore how that would drive our own. Thank you for your questions. We'll move to the next caller.

Speaker 4

We will go next to One Yungas City.

Speaker 2

Thank you very much. Do you guys have a sense as to whether or not you're actually undershipping distributors' sales through, I suppose you would call it, in Q4? In Q3, can you just give us a sense as to the relative stock between OEMs and TI?

Speaker 0

Okay. Glenn, in terms of Q3, we had sales out of our distribution channel down a couple of points, a couple of percentage points compared with the third quarter. Yet against that, they were able to reduce their inventory levels. I think you would translate that to say our shipments in were below, or yeah, below their shipments out. Therefore, they were able to reduce the levels of their TI inventory. I'm sorry. You had another part of that question?

Speaker 2

How does the OEM business compare relative to distribution business?

Speaker 0

Okay. We have less visibility into OEM other than where we have consignment programs, which is significant. OEM consignment programs probably represent 25% of our total revenue, so we have good visibility there. Absolutely, if we own the inventory, we keep it lean, we keep it high turning. For about 35% or so of our revenue that tends to be operated on traditional backlog and order entry systems, we have less visibility. That is really where I was saying previously, our view of what we've seen, and we actually saw it in our revenue trends in the quarter, is that customers, because of the economic uncertainty, basically reset their own desired inventory levels. They accomplished that reset earlier in the first half of this quarter, and then we've seen revenue pick up somewhat as that inventory completion, or inventory reset, has been completed.

That is not saying they're off and running by any means in terms of their own production levels, but their inventory reset of TI components we think is largely complete at this point. Do you have a follow-up, Glenn?

Speaker 2

Yeah, I do. I couldn't help but notice in the list of phones that you gave with OMAP 4 in it that a lot of those sounded like they were phones someone else had in the Honeycomb generation of Android. It sort of raises the question of stickiness of past processors as we change generations of operating systems. I wonder if you can give us a sense of that. How sticky do you think OMAP 4 will be in Ice Cream Sandwich? More importantly, how sticky do you think it will be in the next generation of Google operating system?

Speaker 0

Glenn, I don't want to imply any, how would I say it? That we take any of these programs for granted. We have to win our business and rewin our business every day. I would say at the same time, for a lot of these customers, TI has been a long-term supplier of many products into their handsets. My own view is I think Honeycomb was a bit of an anomaly, especially on the tablet side, because there was a race by many OEMs to have some type of demonstration product available at last year's Consumer Electronics Show. That drove, at least what I would say was very visible as wins for maybe another supplier.

As I would have described even back in last January, a lot of those same OEMs that were demonstrating platforms based on another supplier at Consumer Electronics Show, we had active development programs underway at that same point in time with OMAP. I think the difference is we're a supplier that they've had a history with. They're comfortable. They know we can support both their business terms as well as the kind of volume production requirements that, operationally, these high-volume handsets require. I think we're pretty confident in our position with these various players. At that same time, we have proven over the years our ability to do a good job in rewinning that business every day that we're engaged with these customers. Thank you, Glenn. We'll move to the next caller.

Speaker 4

We will go next to Crystal's Carousing Company.

Speaker 2

Thanks for taking the question. Braun, you had mentioned you're seeing some signs of a cyclical bottoming. You mentioned a slowing in order deceleration activity in August and September. Are there other things that you're seeing out there that lend conviction that we are getting close to a bottom?

Speaker 0

I think just, you know, the fact, I tried to identify three things. One was the order trends. The second one was that, you know, just in general, lead times are short, which, you know, in and of itself is not indicative of a bottom, but usually exists at the bottom of a cycle. Finally, just the fact that, I'm going blank on my third one. Kevin, do you remember what it was?

Speaker 5

We saw revenue profile shift through the quarter.

Speaker 0

Yeah, there was another one that I had mentioned. Anyway, I'll find it here shortly. Oh, just that, as I said before, inventories are lean both at OEMs and distributors. You know, we don't have additional depletion of inventory that needs to take place for growth to start to recover. That doesn't mean we're at a bottom, because revenue will go down in the fourth quarter, especially if you compare on an apples-to-apples basis. One thing I'll also note is that if you look at the delta to just seasonal, third quarter, we were about 7% below our seasonal norm. In the fourth quarter, if you take the middle of our range, we'd also be about 7% below the seasonal average.

One way of looking at that is that at least we're not, to the extent we accomplish that mid-range point in revenue for the fourth quarter, we're kind of, maybe you call it bumping along the bottom as opposed to operating in an environment that is getting progressively worse. We'll see how it lands, but that's our perspective. Do you have a follow-up, Craig?

Speaker 2

Yeah, thanks. I think TI is historically a strong notebook power management position. Given all the flooding issues in Thailand, is there anything that TI is picking up that would leave you uncomfortable with build activity in the fourth quarter for that business? You know.

Speaker 0

Craig, I'm a little uncomfortable getting too far out on the edge there because there certainly is a lot of uncertainty. That's a rapidly changing situation. You know, I can walk you through some considerations. First is that TI does not have any facilities there. That being said, we do have some assembly subcontractors as well as a few component suppliers, for example, some very specific lead time or lead frame suppliers. We do have some products that, as a result, are impacted. To the best we've been able to gauge, our current assessment is that it would impact potentially about 1% of our revenue in the fourth quarter. We're working with the customers that are impacted, working with them on contingency plans. We're going to work through that.

This does not include any impact to the customers themselves, either customers that are operating in Thailand or customers that might have their own production plans changed because they're unable to get supply from other suppliers that have a more direct Thailand impact. We're trying to work through all that. Just a note, we do have programs where customers, and these are programs where maybe there are some second sourcing type of arrangements on high-volume programs such as hard disk drives, where customers have contacted us requesting us to pick up additional support for products that their other sourced semiconductor supplier might have been more impacted by Thailand from an assembly test perspective and they're asking us for upside support. Again, we're working through all that.

The best we can say, at least as it directly affects our production, is very minimal impact, but we're still working with the customers to understand their end of it as well. Okay, Craig, thank you for your questions. We'll move to the next caller.

Speaker 4

We will go next to Vivek Arya of Bank of America Merrill Lynch.

Speaker 1

Thanks for taking my question. Also, thanks for the transparency around the National. It's very helpful. I'm trying to compare your Q4 guidance with the Q3 results. If we exclude the one-time charges, we get $0.60 in pro forma EPS for Q3. If we do the same for Q4, we get $0.47. Apples to apples, excluding charges, EPS is down over 20%. You're guiding sales down only, I think, 2% or so. What is driving the much larger decline in pro forma EPS?

Speaker 5

Among the things that we talked about sequentially, you've got, as you pointed out, lower revenue that is probably about $0.02 worth of the delta there. We're also expecting lower utilization. That's about a penny of the delta. I mentioned to you during my prepared remarks that the reduction in OpEx that we saw in the third quarter was due to some catch-up adjustments for accruals on variable compensation. Those will not recur in the fourth quarter, and that will cause our TI OpEx, if you will, to go up by about a penny. We also had a benefit in the third quarter from a lower tax rate from a catch-up adjustment that was similar. The fourth quarter will be on a year-on-expenses recurrence tax, so that's about $0.02. Finally, we'll have a full quarter's worth of National Semiconductor's OpEx in our numbers, and that's about $0.07 in transition.

When you run the math, that's how you get to the $0.47.

Speaker 0

That's from the $0.60 to the $0.47 in both cases where acquisition costs are excluded. Is that right, Kevin?

Speaker 5

Correct.

Speaker 0

Vivek, to make sure I emphasize, when Kevin talked about the return of TI OpEx that had been reduced in the third quarter, the third quarter reduction in OpEx included two components, basically. One was ongoing reduction. The second component, basically, was the catch-up for the fact that we had been accruing compensation expense at a higher rate during the first half of the year. The ongoing reduction will continue into the fourth quarter. It's just the catch-up amount that doesn't recur that results in some sequential increase in the fourth quarter. A very similar story on the tax reduction in the fourth quarter and a slight increase in the third quarter. I'm sorry, the fourth quarter. Do you have a follow-on, Vivek?

Speaker 1

Yeah, just one question. Your buyback appears to be holding up quite well. How do you look at that as we go through this softer macro period? Are you allocating a certain percentage of free cash flow to it, or how are you strategically looking at the buyback activity?

Speaker 5

Vivek, there's a number of different elements that go into it. As we had indicated when we announced the acquisition of National Semiconductor, we would moderate the amount that we spent on buybacks versus what you'd seen us in prior periods, largely in preparation for beginning to set aside cash reserves to begin to repay debt as it comes due. For example, National Semiconductor debt that we acquired includes $375 million that's going to be due on June 15, 2023. We'll need to set some cash aside to pay that. Further, going into 2013, of course, we'll have some of our own debt that we issued for the acquisition coming due, and we'll begin to build cash for that.

All that being said, we did announce last month that our board of directors approved an increase in our dividend by a little over 30% from $0.13 per share to $0.17 per share. That's effective this quarter, so that is part of how we're returning cash to our investors. We'll continue to return cash for buybacks, albeit at a more moderate rate, over the foreseeable future as we shift to accumulating some cash for the maturity of the debt when it comes due.

Speaker 0

Okay, Vivek, thank you for your questions. Operator, we'll move to the next caller.

Speaker 4

We will go next to Jim Cavello of Goldman Sachs.

Speaker 2

Great. Good evening, guys. Thanks so much for taking the question. First question would be just on margins. What are going to be the big drivers of the margin delta from quarter to quarter over the next several quarters? I understand your comments about fab utilization driving about a penny of the margin, or utilization is driving a penny of the EPS decline in the fourth quarter. What's going to be the most important going forward? Mix versus fab utilization versus even for the next quarter, some of the expansion issues on the National inventory?

Speaker 5

Yeah, Jim, I think the biggest driver next quarter is going to be the market to fair value of the inventory that we acquired from National Semiconductor. Basically, there's going to be almost $100 million worth of additional cost in cost of revenue that would normally not be there. I think if you do the math on that, it probably falls through on an after-tax basis at somewhere around $0.055, $0.06. Clearly, that's weighing us down until we get through that fair value accounting adjustment. As you already pointed out, we've also got lower utilization that will be weighing us down. Once the utilization is, certainly once the fair value adjustment is behind us, we'll get a little bit of a lift just because that won't be recurring.

In addition, as we begin to see the bottoming out that Ron mentioned and ideally a resumption of end-to-end on more of a growth vector, utilization will start being a benefit for us. Most importantly, in the longer term, with the acquisition of National Semiconductor, Analog now amounts to almost half our total revenue from somewhere around the low 40%, 43%, 44% earlier in the year. On a go-forward basis, clearly that mix will help us. I'll remind folks on the phone that National's overall gross margins actually were higher than TI's prior to acquisition. That alone should help upmix a little bit. More importantly, the total mix of analog revenue will become a bigger portion of TI's revenue as we go forward. That alone will help us as we march on our long-term goal to a 65% gross margin for the company.

Speaker 0

Okay, great. Do you have a follow-up, Jim?

Speaker 2

I kind of have a follow-up, one A and B. I'll let you guys decide if you want to answer both or one. The first would be I'm assuming that the expensing of the inventory is your one-quarter issue, and you're going to get all that back then in the first quarter margins as we think about. The second thing is with the increase in margins from National as you referenced, is there any updates or when might we hear an update to the long-term margin model of the company? Thank you.

Speaker 5

All right, Jim. I don't know if I want to take one A or one B, so maybe I'll just give you both. On the expensing, you are correct. That is a one-quarter issue by our estimate. That should be essentially complete through the fourth quarter and should have that behind us as we enter the first quarter. As to increasing or adjusting the model for the company, we want to be able to consistently deliver up to 55% before we go off and start looking at a new model for the company. We'll hold to the model that we've had for several years now, which is 55% gross and 30% operating, is what we're trying to design the company to operate at.

Speaker 0

Okay, thank you for your questions, Jim. We'll move to the next call, operator.

Speaker 4

We will go next to the shareholder of UBS.

Speaker 0

Sure. Thank you very much. Ron, can I just ask you real quick on connectivity and the wireless business? We've seen OMAP grew, baseband grew, but connectivity was down. Can you talk about two things? One is what's causing that shift in connectivity with the rest? Also, how much more resilience shall we expect in the baseband business? Could you.

Speaker 5

How much more? What was your last?

Speaker 0

What is the uplift for the baseband business? Should we still continue to model that to go down?

Speaker 5

Okay. Connectivity is pretty straightforward. It'll be a little bit awkward of a response for what will become obvious reasons. Our two largest connectivity customers currently are not doing as well in the marketplace as some of their peers. As a result, TI's connectivity revenue has not held up as well as some of our other, as for example, the OMAP revenue. Keep in mind, there's also a mix even within the customers. For example, our connectivity product line tends to be very, it tends to be centric around Wi-Fi and combos that use Wi-Fi or centered upon Wi-Fi. That tends to be the very highest end of the customer's product range. While smartphones are doing well overall, the mix within some of our customers has been stronger at the lower end where we don't supply connectivity versus the higher end where we are engaged.

The outlook for baseband going forward is unchanged. What we've seen this quarter is the kind of noise that we've told you to expect in baseband where, during the first half of the year, we had some inventory reduction taking place that our customer this quarter represents a bit of a snapback from that. We've moved above, call it that historical or the linear trend line. The best guidance we can give you going forward is draw a straight line between third quarter of this year and first quarter of 2013 being zero. That's a rough approximation of how we expect that revenue to end. We expect that it will be completed essentially by the end of 2012. I don't know if that was one or two questions you'd say, but do you have a follow-on?

Speaker 0

Yeah. Separately, can I just ask you about some of the moving parts in the fourth quarter guidance? Obviously, excluding the contribution from National Semiconductor and their specific end markets, where within what you describe as bottoming process, are there specific end markets where you are seeing stronger or weaker as we look at Q4? Thanks.

Speaker 5

I think somebody else had taken a run at that. As you're aware, we don't break out our revenue forecast by various end markets. I'm sure we'll have certain areas better than others, but we're not forecasting that. Keep in mind the pressure that we're feeling on demand is economically driven. I would say it'd be more so on the consumer side than on the enterprise side, but it's economically driven, which tends to weigh pretty much across the board. Just as in third quarter, things generally across the board were well below seasonal, same kind of expectation going into fourth quarter. Okay, Ichai, thank you. We'll move to the next caller, please.

Speaker 4

We will go next to Sean Webster of Macquarie.

Speaker 2

Great. Thanks a lot. My first question is on the supply chain and inventory. Are there areas from your perspective that still seem lean? On the other side, are there areas which seem like they need to digest some inventory?

Speaker 5

As Ron mentioned earlier, just as we look at certain supply chains, we see where our inventories are at. Our distribution inventories, as Ron mentioned, were down about a day, still holding there at about six weeks. Our consigned inventories and our consignment customers continue to be very lean by definition because they pull only as they consume. That represents about two-thirds of channel that we can see. The other third, which is more classic sales to OEMs, we don't really have great visibility there. We also don't have a lot of concern there, given that broadly speaking, we actually came out of third quarter a little bit higher than we expected versus the mid-quarter upbeat. That was from a variety of customers actually ordering material a little bit more than we expected in the quarter.

Those signals all tell us that the inventory is probably in pretty good shape across the entire customer base and where we supply it.

Speaker 0

Do you have a follow-on, Sean?

Speaker 2

Yes, I do. Thanks. In terms of other income, is there any guidance there? Also, can you update us on the capacity of your 300 millimeter and expansion plans? Thanks.

Speaker 5

Yeah. On the other income, Sean, I think that the main thing that you saw in the second quarter, third quarter, excuse me, was interest expense of about $50 million. That was an all-in quarter for the debt that TI now holds and will hold for a couple of years. Fourth quarter will be a little bit higher. It'll be about $22 million. That will be adding in the debt that we assumed with the acquisition of National Semiconductor. That's probably the main change I'd point you to on the other income and expense. As to the logistics income, we have, as you know, been having that factory for some time now, especially stepped at a ramp pace when the earthquake happened in Mejia, as we do hold a lot of products. We continue to ramp product into that factory.

The total capacity for what we've purchased equipment for would generate about, when it's all loaded, as we indicated, up to $2 billion worth of revenue out of that factory. We're considerably below that today as we continue to ramp.

Speaker 0

In fact, I think our current levels we're running up now 20% of that of the capacity level. The other point I need to make is $2 billion that Kevin references is the equipment that we have purchased in-house today, which, as you're aware, Sean, only represents about half of the floor space inside of that fab. We have lots of room for expansion, but we also have plenty of equipment already on the floor that we're looking into. Again, 20%, lots of room to grow just even within the equipment we've purchased, but lots of floor space beyond that. Okay, Sean, thanks for your questions. Let's move to the next caller.

Speaker 4

We will go next to Braun Slaymaker of Barclays.

Speaker 3

Good afternoon. Thank you for taking my question. I guess first question, in terms of your comments on seeing stabilization, I was curious if you're seeing that across the board or specific to any certain end markets.

Speaker 5

CJ, I guess we'd say that the size stabilization is more a question of we're seeing a bottoming process beginning to form, not so much stabilizing, as evidenced by the fact that we still expect our revenues to be down next quarter. It's more important to note that what we're seeing through our order patterns, through our lead times, and through overall inventories is that we'll begin to see the bottom of this demand decline begin to shape or form a bottom, although it's not there yet because we expect the fourth quarter to decline.

Speaker 0

Okay. CJ, a follow-on.

Speaker 3

Yeah, as a quick follow-up, just to revisit gross margins. If I look at the OpEx guide and that $22 million other income or other expense, rather, it looks like you're guiding gross margins roughly flat. I guess, A, is that correct? Then, B, once we get through selling of the inventory without margins, should we be thinking we get immediately back up to another 100, 150 bps higher on the gross margin side even without a pickup in utilization? How should we think about that?

Speaker 5

Yeah, CJ, I'm not really sure with how you're building your gross margin assumption in your model. I would point out that there is about $100 million in cost of revenue in the fourth quarter that will depress our gross margins on a GAAP basis quarter over quarter. In addition, we expect lower utilization, so there'll be more pressure on gross margins from that standpoint. If you're extracting that $100 million, then maybe that's where you're getting your gross margins to be roughly flat quarter over quarter. I'm not really sure what your assumption is there, but I would want to point out those two items I just mentioned.

Speaker 0

Right. $100 million, CJ, as you probably alluded already, will be about 300 basis points, approximately, pressure in the fourth quarter. That will be completely gone in the first quarter. There may be other things that come about in the first quarter, of course. That inventory, the expensing of the inventory write-up, is a fourth quarter effect only. It'll be cleaned out by the time we get into the first quarter. Okay, CJ, thanks for your questions. Let's move to the next call.

Speaker 4

We'll go next to Chris Danely of JP Morgan.

Speaker 2

Thanks, guys. On the National acquisition, I think you threw out a target of $100 million in cost savings. Do you still feel confident in that target? Anything that's surprised you on the upside or the downside with the acquisition so far and integration?

Speaker 5

Chris, on the $100 million cost savings, we still hold to that forecast. Just to remind the other listeners on the phone, what we were talking about there is that by the end of the first year after acquisition, so that would be probably the end of the fourth quarter of next year, we'd expect to have enough of the redundant costs removed that we would see about $100 million of annualized cost savings or about $25 million a quarter starting about a year from now. We still expect that to be a reasonable forecast for planning purposes. As to any surprises on the upside or downside, I'd say it's probably just a little bit too late for us to really find any of those yet.

We've been pretty focused here on Greg and the management team getting out and visiting with customers on the world tour and focusing on updating our online systems and our marketing material to be able to make this a seamless appearance to our customers and that there's one place to go that has all 42,000 plus analog parts in it. We've also been focusing on the beginning steps of the overall integration of our systems and methodology. Not enough time yet to really have pluses or minuses beyond that we can talk about.

Speaker 0

Hey, Chris, I will add that this acquisition is hugely important to TI. As a result, we understand it's also really important to our investors. You will see ongoing progress updates and lots of detail from TI in terms of our progress there. Stay tuned. You'll see them as we get them. Do you have a follow-on, Chris?

Speaker 2

Yeah, I guess one more broad-based question on National. Now that you've had a chance to at least, you know, take a look at the piece under the covers, kimono, whatever you want to call it, what do you guys think are the two or three keys to, you know, sort of turning around the share loss there? What sort of landmarks can we look at or any sort of timeframe there?

Speaker 0

Kevin, what did you find as you looked under the hood?

Speaker 5

I think probably the encouraging thing that we found early on, certainly from Greg Lowe's tour with his managers, the reception that we've gotten from those in our distribution channel and importantly from our customers has been extremely encouraging. The sound bite that I keep gravitating to or hearing, if you will, is that the customers really, really liked National's parts. They're glad that they can now comfortably and safely design them back into their future products. I actually find that very encouraging. Beyond that, we clearly have some challenges in front of us also, which is making sure that we are able to integrate the companies from an overall process standpoint, to be sure that our systems integrate properly and we don't wind up with blind spots and holes and so on. There are always challenges when you do an integration of this size.

Speaker 0

Chris, I think it's been pretty exciting for some of the National management that have been out on this world tour as well because of what they, I mean, they're finding a reception as part of TI that they haven't been used to for a while. I think there is a refreshing upside that they're seeing that makes them pretty excited to be part of TI as well. Okay, Chris, thank you for your questions. Operator, let's go ahead and go to the next caller, please.

Speaker 4

We'll go next to Ramesh Shah.

Speaker 1

Hey, Ron. Just want to confirm that my math is correct. National had about $300 million in revenue in Q4. It'll be about, excluding acquisition charges, about $0.10 diluted to EPS. Am I in the ballpark?

Speaker 0

Kevin, thinking about the EPS, I can confirm you're in the ballpark with $300 million of Q4 revenue.

Speaker 5

Yeah, Ramita, it will add some to EPS on a pro forma basis. You need to be careful on a GAAP basis because, in fact, we will have an increased acquisition-related charge that I mentioned, $160 million of acquisition costs plus $100 million of the fair value markup of their inventory that we go through cost of goods sold.

Speaker 0

Right. The amount that goes through, you know, that'll be reported as part of our analog segment, which is more the ongoing operations, will be profitable. You're probably not too far off in terms of your estimate there, Romit. Again, that's exclusive of all acquisition-related charges. Okay. Do you have a follow-on?

Speaker 1

Yeah. I saw that you reiterated your CapEx plans for the year. Can you just share with us your thoughts about that going forward into 2012 and how you're managing headcount in this environment?

Speaker 5

Yeah, Ramit, the CapEx, we haven't given a 2012 forecast just yet, but I would remind you that we did take advantage of the downturn and acquired substantial capacity, especially in the more expensive wafer-fed side of the CapEx bill during the last 18 months or so, which implicitly means that our need to spend money on that portion of capital should be lower than normal for the next couple of years. Beyond that, we'll continue to spend, of course, on our assembly and test operations. Our model is 5% to 8% of CapEx spend. In fact, if you take a look over the last 12 months, we've actually spent about 7% of revenue on CapEx, with depreciation running about 6.5% of revenue. Consistent with where we've been, I'd expect those kind of factors to apply going forward.

As to staffing, we do continue to hire people on a selective basis in the areas where we expect to grow. Those will be in Analog and Embedded Processing, as well as in field apps engineers and talented salespeople.

Speaker 0

Please remember, we just hired over 5,000 new TI employees. Thank you, Romit. Operator, I think we have time for one final caller.

Speaker 4

Very good. That will come from Ron Seymour of Deutsche Bank.

Speaker 2

Thanks for squeezing me in, guys. One clarification really quickly. The $0.47 midpoint of the pro forma guide, that excludes that inventory write-up on the National side. If we X that out of the fourth quarter, you don't have to worry about it in the first quarter reversing itself. Is that correct?

Speaker 5

Correct.

Speaker 2

I guess a bigger picture question. What's your assumption for distant inventory levels in the fourth quarter? Is your guidance assuming that the inventory gets burned further, stays the same, rises, etc.?

Speaker 0

Ross, we don't tend to provide a look break out there. Let me just say TI distribution inventory is very consistent with where it has been historically. I don't anticipate that you're going to see a big change there. I think we're just over six weeks at the current level. We went down a stay even in the third quarter. The one thing I would say is that with National, there's adjustments going on with distribution in general there, so there can be some slight realignment. I think they finished third quarter, or what we had in the channel there when we completed the third quarter, was about 10 weeks of inventory. Historically, they tend to run, I think, 9 to 10 weeks. You'll see over time further adjustments there as we implement consignment programs with distribution and those types of value-add programs.

That'll be over the course of time, and literally, that'll be years as we implement that. Did you have a follow-on, Ross?

Speaker 2

Yeah, just really quickly, the $100 million in synergies that you guys plan on with National, any shot at how much of that comes on the OpEx lines rather than the COGS lines?

Speaker 5

Virtually all of it will come on OpEx, Ross.

Speaker 2

Great. Thank you.

Speaker 0

Okay, Ross, thank you. To all of you, thank you for joining us. A replay of this call is available on our website. Good evening.

Speaker 4

As we conclude today's conference call, we thank you for your participation and have a wonderful day.