Microchip Technology Incorporated (MCHP) Q4 2015 Earnings Report, Transcript and Summary

Good day, everyone, and welcome to this Microchip Technology Fourth Quarter and Fiscal Year 2015 Financial Results Conference Call. As a reminder, today's call is being recorded. [Operator Instructions] At this time, I'd like to turn the conference over to Microchip's President and Chief Executive Officer, Mr. Steve Sanghi. Please go ahead, sir.

Thank you. Good afternoon, everyone. During the course of this conference call, we will be making projections and other forward-looking statements regarding future events or the future financial performance of the company. We wish to caution you that such statements are predictions and that actual events or results may differ materially. We refer you to our press releases of today as well as our recent filings with the SEC that identify important risk factors that may impact Microchip's business and results of operations. As you know, in addition to earnings, we also announced the signing of a definitive agreement to acquire Micrel. During this conference call, we will not cover anything in relations to the acquisition and we'll take your questions only related to the earning. After we conclude this conference call, you will have to dial in to the second conference call where we will cover matters related to the acquisition and take your questions. The second conference call has to be filed with the SEC, hence the separation of 2 conference calls. The second call will begin at 5:45 p.m. Eastern Time. With that, let me now pass this call to our Chief Financial Officer, Eric Bjornholt.

Thanks, Steve, and good afternoon, everyone. In addition to Steve Sanghi, Microchip's President and CEO, Ganesh Moorthy, Microchip's COO, is also participating in this call. I will comment on our fourth quarter and full fiscal year 2015 financial performance, and Steve and Ganesh will then give their comments on the results and discuss the current business environment as well as our guidance. We will then be available to respond to specific investor and analyst questions. I want to remind you that we are including information in our press release and this conference call on various GAAP and non-GAAP measures. We have posted a full GAAP to non-GAAP reconciliation on the Investor Relations page of our website at www.microchip.com, which we believe you will find useful when comparing GAAP and non-GAAP results. I will now go through some of our operating results including net sales, gross margin and operating expenses. I will be referring to these results on a non-GAAP basis prior to the effects of our acquisition activities and share-based compensation. Non-GAAP net sales in the March quarter were a record $547.2 million and were modestly above the midpoint of our guidance, which was $546.6 million, and were up 10.9% from the March 2014 quarter. Revenue by product line in the March quarter was $353.6 million for microcontrollers, $127.6 million for analog, $33.2 million for memory, $23.6 million for licensing and $9.2 million of other. Revenue by geography was $106.2 million in the Americas, a record $126 million in Europe and $314.9 million in Asia. I remind you that we recognize revenue based on where we ship our products to, which tends to skew some of the revenue towards Asia where a lot of contract manufacturing takes place. On a non-GAAP basis, gross margins were at the midpoint of our guidance at 58.3% in the March quarter. Non-GAAP operating expenses were 25.9% of sales, well below the bottom end of our guidance range. Non-GAAP operating income was 32.4% of sales and net income was $148.8 million. This resulted in record earnings of $0.68 per diluted share, which was at the high end of our guidance. For fiscal 2015, on a non-GAAP basis, net sales were a record $2.161 billion and up 11.9% year-over-year. Gross margins were 58.8%, operating expenses were 26.4% of sales and operating income was 32.4% of sales. Net income was a record $593.9 million or $2.66 per diluted share. GAAP net sales in the March quarter were $543.2 million, and as originally guided, were $4 million lower than non-GAAP net sales due to the contractual relationship change we made with the ISSC distribution network to move them to a sell-through revenue recognition model in the December quarter. The revenue impact from this change was fully recognized as of the end of March and will not create a GAAP to non-GAAP revenue difference in future quarters. On a GAAP basis, gross margins including share based compensation and acquisition-related expenses were 57.7% in the March quarter. GAAP gross margins included the impact of $2 million of share-based compensation, $1.8 million of gross profit on the ISSC revenue recognition change previously mentioned and $1.5 million in manufacturing shutdown costs associated with the Supertex wafer fab in San Jose. Total operating expenses were $203.3 million or 37.4% of sales and include acquisition intangible amortization of $47.1 million, share-based compensation of $13.2 million, $0.7 million of acquisition-related expenses and special charges of $0.8 million. GAAP other expense includes a $50.6 million loss on the retirements of 50% of our 2,037 convertible bonds and an $18.5 million gain on the sale of a portion of investments that Microchip acquired as part of the SST acquisition back in 2010. GAAP net income was $99.4 million or $0.45 per diluted share. GAAP net income includes nonrecurring favorable tax events of $26.7 million, which were primarily driven by the closure of a tax audit and the expiration of the statute of limitations for certain previously established tax reserves. For fiscal year 2015, GAAP net sales were a record $2.147 billion. Gross margins were 57.3%, operating expenses were 37.4% of sales and operating income was 19.8% of sales. Net income was $369 million or $1.65 per diluted share. In the March quarter, the non-GAAP tax rate was 11.4%. The GAAP tax benefit rate was 58.9%. The GAAP tax rate was favorably impacted by the $26.7 million of nonrecurring tax events that I mentioned before and the tax impact of certain of the one-time events in the quarter. Our tax rate is impacted by the mix of geographical profits, withholding taxes associated with our licensing business and the tax effect of various nonrecurring items. Excluding any nonrecurring events, we expect our longer-term forward-looking non-GAAP effective tax rate to be about 10.75% to 11.25%. To summarize the after-tax impact that the non-GAAP adjustments had on Microchip's earnings per share in the March quarter, acquisition-related items were about $0.217, share-based compensation was about $0.052, nonrecurring favorable tax events were about $0.121, a favorable difference on the sale of a portion of investments acquired in the SST acquisition was about $0.084, the loss on the retirement of our 50% of our 2,037 convertible bonds was about $0.144, the difference in the GAAP and non-GAAP noncontrolling interest in ISSC was a favorable of about $0.05 and noncash interest expense was about $0.019. The dividend declared today of $35.75 per share will be paid on June 4, 2015, to shareholders of record on May 21, 2015. The cash payment associated with this dividend will be approximately $72.4 million. This quarter's dividend will be our 51st consecutive quarter of making a dividend payment. We have never made reductions in our dividend and in fact, this quarter's increase marks the 45th occasion we have increased the dividend payment and our cumulative dividends paid amount to over $2.5 billion. This program continues to be an important component of how we return value to our shareholders. During the time period that Microchip has paid dividends, we have also purchased back $1.4 billion of our stock, including the stock that we bought back with the issuance of our convertible debenture back in 2008. Our cash return to shareholders since the inception of our dividend program is over $3.9 billion. Moving on to the balance sheet. Consolidated inventory at March 31, 2015, was $279.5 million or 111 days, flat to the levels at the end of the December quarter. Inventory at our distributors increased by 1 day in the March quarter and are at 37 days. I want to remind you that our distribution revenues throughout the world is recognized on a sell-through basis. We intend to grow the days of inventory on our balance sheet in the June quarter to support our business growth and customer service levels. The cash generation in the March quarter, excluding the purchase of additional shares of ISSC, our dividend payment, changes in cash from the issuance and retirement of our convertible debentures and changes in borrowing levels under revolving line of credit was $155.5 million. As of March 31, the consolidated cash and total investment position was $2.34 billion, and our borrowings under our revolving line of credit were $462 million. Excluding dividend payments and our acquisition activities, we expect our total cash and investment position to grow by approximately $160 million to $180 million in the June quarter. Capital spending was approximately $29.4 million for the March quarter and $149.5 million for fiscal year 2015. We expect about $40 million in capital spending in the June quarter and overall capital expenditures for fiscal year 2016 to be about $160 million as we are adding capital to support the growth of our production capabilities for our fast-growing new products and technologies and to bring in-house more of the assembly and test operations that we are currently outsourcing. Depreciation expense in the March quarter was $25 million. I will now ask Ganesh to give his comments on the performance of the business in the March quarter. Ganesh?

Thank you, Eric, and good afternoon, everyone. Let's take a closer look at the performance of each of our product lines, starting with microcontrollers. Our microcontroller revenue was up 8.4% in the March quarter from the year-ago quarter. For fiscal year 2015, our microcontroller business was up 11.4% over fiscal year 2014, crossing the $1.4 billion mark for the first time and setting an all-time record. Also for fiscal year 2015, each of our 3 microcontroller segments, 8-bit, 16-bit and 32-bit, set revenue records. Microcontrollers represented 64.6% of Microchip's overall revenue in the March quarter. As we mentioned in our January conference call, for competitive reasons, we will only be providing overall microcontroller revenue and growth rates and not break out by segment, consistent with the practice of most of our competitors. Gartner Dataquest just released their microcontroller market share report for 2014. We are pleased to report that Microchip has regained the #1 position for 8-bit microcontrollers. Four years ago, it took the merger of 3 Japanese semiconductor giants, NEC, Hitachi and Mitsubishi, in the form of Renesas to knock us off the #1 spot for 8-bit microcontrollers. We assured you at that time that we would work relentlessly to gain market share and to wrest back the #1 spot in the coming years. Post their merger, Renesas' 8-bit microcontroller business in 2010 used to be 41% larger than ours. In every year since 2010, we have closed the gap versus Renesas, and in 2014, we wrested back the leadership position and finished 10.5% larger than Renesas. In the 16-bit microcontroller market, we were again the fastest-growing 16-bit microcontroller supplier among the top 10 suppliers in 2014, growing at over 2x the rate of any of the other top 10 suppliers. While our substantially faster growth rate closed the gap versus higher-rank suppliers, we remained in the #5 spot in 2014 and expect to continue to gain share and move further up the rankings in the coming years. In the 32-bit microcontroller market, we moved up to the #9 spot in 2014 from the #11 spot in 2013, continuing our relentless climb up the rankings since we entered this market a few years ago. Now, Gartner Dataquest's report is a backward-looking indicator where we are performing very well. Now to look at a forward-looking indicator. In April also, UBM Tech, which is the parent company of EE Times, released the results of their annual Embedded Market Study. Once again, Microchip was rated by the embedded system design engineers as their #1 choice for new designs using 8-bit and 16-bit microcontrollers in the performance range that we compete in, and we were the #2 choice among 32-bit microcontrollers. We are humbled and gratified by the overwhelming preference by engineers for our solutions and see this as a positive sign of future growth. Our overall microcontroller results as well as each of our 8-bit, 16-bit and 32-bit results are clearly outperforming the market, with year-over-year growth rates well above the market and what we've seen reported by our competitors in their results. These outstanding results in a very competitive market are a tribute to the tireless effort by the worldwide Microchip team. We have gained significant market share and have the new product momentum and customer engagement to continue to gain even more share as we build the best-performing microcontroller franchise in the industry. Analog products. Our analog business was up 18.7% from the year-ago quarter. In fiscal year 2015, our analog business was up 17.6% over fiscal year 2014, crossing the $0.5 billion mark for the first time and also setting an all-time record. This business continues to have strong design win momentum in a broad range of applications and represented 23.3% of Microchip's overall revenue in the March quarter. We continue to develop and introduce a wide range of innovative and proprietary new products to fuel the future growth of our analog business. Now moving to the memory business, which is comprised of our Serial E-squared memory products as well as our SuperFlash Memory products. This business was up 0.2% versus the year-ago quarter. In fiscal year 2015, our memory business was down 1.8% versus fiscal year 2014. We continue to run our memory business in a disciplined fashion that maintains consistently high profitability, enables our licensing business and serves our microcontroller customers to complete their solutions. Our memory business represented 6.1% of Microchip's overall revenue in the March quarter. I have one more topic to cover today. Given the importance of the Internet of Things market, some of our investors and analysts have asked for more color about the size of our IoT business. First, there is no consistent way in which semiconductor companies report their IoT revenue. As a result, there are some who report their entire embedded market revenue or the revenue of business units focused on the embedded market as IoT revenue. By that measure, pretty close to 100% of Microchip's $2.2 billion of revenue could be classified as IoT revenue, but it would not be as meaningful to investors in terms of its potential as a future growth driver. We have a stricter definition for our IoT revenue in that we only count end applications or things that are both smart and connected reflecting what the promise of IoT really is, which is the possibilities that open up by adding connectivity to smart systems. Examples of smart and connected applications for Microchip are things like alarm systems, access control systems, thermostats, smoke detectors, exercise machines, asset tracking, wireless audio docks, set-top boxes and other consumer electronic systems, hospital personnel locators, patient monitors, automotive telematics, building automation, vending machines, Smart Grid monitoring, wired and wireless networking systems, GPS navigation tracking systems and many, many others that are too numerous to list. More applications will continue to be added as they get connected. I'm sure you will agree that applications like the examples we just provided, which are a combination of smart and connected applications, are more real IoT applications and the products going into them would better characterize our IoT revenue. Microchip solutions that enable the IoT market include low-power microcontrollers, analog, sensors, memories, wired and wireless communication products. So using our stricter definition for IoT revenue, Microchip estimates that our current annual IoT revenue is approximately $275 million and has been growing at a greater than 35% compounded annual growth rate over the last 3 years. This revenue is already included in the public reporting segments for Microchip. Let me now pass it to Steve for some general comments about our business as well as our guidance going forward. Steve?

Thank you, Ganesh. Today, I would first like to reflect on the results of the fiscal fourth quarter of 2015 and fiscal year 2015 then I will provide guidance for the fiscal first quarter of 2016. We were very pleased with our execution in the March quarter. We achieved record net sales in the March quarter and slightly ahead of the midpoint of our guidance. Our non-GAAP earnings per share of $0.68 was a record and at the high end of our revised estimates. Recall that we had revised our earnings per share higher after taking into account the accretive effect of our financing transactions. Looking at the fiscal year 2015. It was another record year for revenue and the first fiscal year to cross the $2 billion revenue mark. For fiscal year 2015, revenue came in at a record $2.16 billion, up 11.9% from fiscal year 2014. All of our strategic product lines of microcontroller and analog posted record revenue performance for fiscal year 2015. The March quarter was also our 98th consecutive profitable quarter. I want to thank all the employees of Microchip for their contribution in this remarkable achievement of 98 consecutive profitable quarters. We're also planning a celebration for our 100th consecutive profitable quarter, which is 2 quarters away. In the last 2 conference call, we told you that we will be shutting down production in Supertex, San Jose fab as well as test facility in Hong Kong. The San Jose fab is now closed with the last wafer outs occurring in early April. The fab building is undergoing restoration to be returned back to the landlord. The transfer of the Hong Kong test operations to our high-volume Thailand facility was also completed in February. Now I will provide guidance for the fiscal first quarter of 2016. As we look at the business environment, it is clear that many of our industry peers are seeing weakness across various end markets including PC and industrial, slowing growth in China market and currency impacts from the strong dollar. Taking all this into account, we believe that our net sales in the June quarter will be between $547 million and $564 million. On a non-GAAP basis, we expect our gross margin to be between 58.3% and 58.5% of sales. We expect operating expenses to be between 25.75% and 26% of sales. We expect operating profit to be between 32.3% and 32.75% of sales, and we expect non-GAAP EPS to be between $0.69 and $0.73 per share. Finally, I want to remind you that given all the complications of accounting for the acquisitions including amortization of intangibles, restructuring charges and inventory write-up on acquisitions, Microchip will continue to provide guidance and track its results on non-GAAP basis. We believe that non-GAAP results provide more meaningful comparison to prior quarters, and we request that the analysts continue to report the non-GAAP estimates to first call. Before we open it up for questions, I want to remind everybody, in case somebody joined the conference late, that in this conference call, we'll be only taking questions regarding our earnings and we'll have a second conference call starting at 5:45 p.m. New York time, at which we will talk about the Micrel acquisition and take questions related to that acquisition. With this, operator, will you please poll for questions?

[Operator Instructions] Our first question comes from William Stein with SunTrust.

Steve, the guidance, a couple percentage points below consensus and perhaps normal seasonal patterns. Understanding that that's not out of line with what other companies are guiding this season as well, I wonder if you can comment on whether you see that small weakness extending further in the next quarter. You often have some strong opinions on the cycle and we -- I think we'd like to hear them.

I have blocked my opinion on the cycle. I'm not going to share with you anymore. Basically, I think as far as the seasonality is concerned, a company with a constant stream of acquisitions that we have done, we believe seasonality often is miscalculated by the investors depending on how many years back you go. We have had Supertex acquisition, which closed in April last year; ISSC acquisition last year. If you go 5 years back, we closed SST acquisition also in April. So when you look at the June quarter numbers, sometimes you do with or without acquisitions and many times, the breakdown of those numbers is not available 2 or 3 years out. So I believe that seasonality changes a lot with these acquisitions. And I think -- but the largest reason is you describe yourself, which is industry conditions are soft, just like announced by other peers. That's what we are seeing and beyond that I'm not going to comment on cycle or anything else for the future.

Okay. I understand that. If I can ask 1 follow-up, please. Wonder to what degree do you see this -- the weakness as FX related, either sort of first order or second order effect and whether there are any end markets, you were good enough to call out IoT. I think historically you haven't done a lot of talking about specific end markets. But is there 1 particular end market like either PC or wireless infrastructure that may be weaker or stronger?

So we don't really have much end market commentary. We serve 90,000-plus customers and a vast majority of them through distribution and many time, at the customers, we work with multiple divisions. For some of them the business is PCs. For some others the business is industrial. For some other it could be called consumer. So we don't really have any end market flavor. But Ganesh, in his commentary, described that our IoT business, the way we defined it, much narrower definition of IoT for applications that are really smart and connected, is growing 35% a year. So that's doing very well. But if you take a broader definition of IoT like include all the embedded business and all that, you could almost call 100% of Microchip's business as IoT, which really would not be helpful. So that's really what I have to say there.

I think the question was on FX.

FX, I don't really think we can delineate the effects that much. But we do business worldwide in U.S. dollars, so 99% plus of our business will be in U.S. dollars. So we don't take a hit on our receivables that way, and when the currencies become weak overseas, it doesn't change our revenue right away because our prices are in dollars. But then it makes the products more expensive for our customers. So as you're renegotiating the price, the next bid comes in, next order comes in, you're negotiating and then you will see some impact. But doing it in dollars gives us more control, allows us to manage it, split it with the customer or do whatever, but really allows us to manage it rather than have an automatic [ph] full FX impact to drop down. So I think the way we do it is probably better and has a delayed effect. It also impacts the business of our customers. If a business of our customers includes lots of products from overseas, which in most cases it does, then their end product becomes more expensive and they're having more trouble in terms of ramping their own products. So those results are real. Some are first tier, some are second tier, some are even third tier and it's very hard to figure this out with 90,000 customers worldwide doing business in 100 countries and I don't think we can decipher that.

Our next question comes from Chris Caso with Susquehanna Financial Group.

The first question is along the lines of currency also, but I noted that your business in Europe appear to be very strong in the quarter. I wonder if you could talk about that a bit. I didn't know if perhaps there were some currency related effects in there, perhaps having an effect on customer demand in the quarter. I guess, was that European strength surprising to you? And just generally, what would you think may be behind that?

Go ahead, Eric.

So Chris, if you look back at our historical results, the March quarter is always a strong period in Europe and you're having growth of almost 19%. It was a really good quarter in Europe, but that's not out of the line with what we've seen historically. And so with the FX rates really changing kind of late 2014, early 2015, the impact on the quarter, from an FX standpoint, was not significant in the March quarter and Steve talked about what the forward effects could be. But it's always a very strong quarter in March in Europe and this quarter was no different.

Okay. Great. Just as a follow-up, I know you're trying to take inventory levels up. If you can remind us kind of where you're looking to take them, how long you think it's going to take to get there. And is there sort of a leverage benefit on margins as you do that?

I think we like our inventory. Product mix has gotten very complex. Today, a significant portion of our business is analog. Lots of our microcontrollers business is very analog-like. There's a high amount of analog on microcontrollers with lots and lots of SKUs. So we're basically finding that our inventory level needs to be between 115 and 120 days. A few years ago, we used to talk about 110 days. So I think it's gone up by about maybe 7 to 10 days to have a reasonable amount of inventory to be able to do business in such a complex mix of SKUs, and our inventory today is 111 days.

111 and we put in the release that we'd expect, based on the guidance for the quarter, for inventory to grow between 3 and 9 days. So take the middle of the road of that and we're kind of getting real close to what our model is for the longer term. So I think inventory is in pretty good shape.

[Operator Instructions] We'll go to Craig Hettenbach with Morgan Stanley.

Just to follow up on the current environment given that the quarter itself came in line. Can you just talk about maybe how it progressed? And it looks like within the last 6 months, we've had a couple times a bit of a pause here. Just what you make of that from a demand or inventory perspective?

Well, in terms of how the quarter progressed, it was a very standard March quarter that starts slow with the new year holidays around the world then picks up some steam and then slows down into the Chinese New Year and then you have a [indiscernible] in March. It's probably the most back-end loaded quarter we do compared to June quarter, September quarter, December quarter. March quarter is the most back-end loaded, predominantly driven by the Chinese New Year and slow start in January, so very difficult.

Yes, and I'd say that was within our expectations.

Okay. Any then -- any color in terms of just kind of canvassing the landscape from a demand and just some of the sluggish trends you're seeing out there?

Well, I hate to go back and visit last year, but late last year, the estimates for the industry were double-digit 10 percentile [ph]. And we were able to see that, that that's not doable. It just didn't look possible with what we were seeing in the PC, the inventory we were seeing in the channel and what we were hearing from our customers. We just didn't think the industry had a driver to really create another 10% growth year. And as you look at it now, the growth rate for the industry has been cut in half. So I really have to say that I'm not surprised. It's -- I think that's what we thought it will be.

Okay. And then maybe just a longer-term question. If you look across, you've been successful in really ramping the analog business as well as on the MCU side in terms of share gains over time. But just as you look at those 2 businesses, any differences for long-term growth rates in terms of which one you might think grow the fastest longer term?

Well, we're a very profitable company. As you can see, 32.3% operating profit. So our microcontroller and analog business is both very well funded and relatively independently. We're not driving to a model that 3 years out, exclusion of a business will be this division or that division. They're all well-funded and it depends on basically achieving product success and marketplace penetration and they meet in manufacturing and they meet at the customer base and they meet in application. Wherein the same application, we got lots of microcontroller as well as analog products going into a single board. So there is a lot of synergy that way. But in terms of making investments, there's not a competition because we can really take care of both. So it will largely depend on where we have larger success. Analog has been growing faster somewhat because that's where we have done some of the acquisitions. Microcontroller acquisitions are a bit more difficult to do because everybody has a different architecture and you got different issues in trying to get synergy in microcontroller. So we had been able to acquire capability in analog both organically and through acquisitions, and it has been a smaller business compared to the microcontrollers. Microcontroller is a $1.5 billion business and analog is about 1/3 of that, $0.5 billion. So it's been growing faster. But it's not something that we are -- if microcontroller can grow faster with IoT with other applications, 32-bit is going very fast, 16-bit is going very fast, 8-bit did a record, but it does have a $1.5 billion gorilla that has to move. And we are #3rd in the overall microcontroller market, third or fourth depending on which year you take, versus in analog, we are much lower. There are much, much larger analog companies that we can take share from for long time.

We'll continue on to Kevin Cassidy with Stifel. Kevin E. Cassidy - Stifel, Nicolaus & Company, Incorporated, Research Division: On your CapEx plan for fiscal year '16, you said $160 million. Can you break it down a little bit? Is it front end or back end and just where the spending will be?

So it's a combination of both, Kevin. We're continuing to invest in new processes and kind of technologies in both the front end and back end and investing to not only support the growth of our organic business, but then invest to bring some things in-house that were previously outsourced for acquisition. So kind of the same story that you've heard before, it is definitely a split between front-end manufacturing and back-end manufacturing. Kevin E. Cassidy - Stifel, Nicolaus & Company, Incorporated, Research Division: Maybe to add on to that, what do you expect your split to be at the end of fiscal year '16 for internal versus outsourcing?

So if you're meaning how much of our wafer fab will be done in-house, so in fiscal '15, about 39% of that was outsourced and 61% was done in-house. There'll be modest movements from that in fiscal '16, won't -- nothing will change dramatically.

Unless there is a large acquisition that moves the mix. From an organic basis, you can move 1 or 2 points a year or something. So it's not very huge. But significant movements had happened when we acquired SST, which was 100% outsourced; when acquired SMSC, that was 100% outsourced. You could ask that question in the second conference call on Micrel and maybe I can answer it there. Kevin E. Cassidy - Stifel, Nicolaus & Company, Incorporated, Research Division: Okay. Yes, maybe what I was trying to get a feel for too is like I think with SMSC, the idea was to bring the test and packaging in-house even though the wafer fab was out. Is that still the longer-term idea to do more of your own testing and packaging?

Yes, so when we answered your question on 39% inside, it was not on a COGS basis. It was basically on a fab basis, front end. In terms of back end, we do...

88% of the test we do in-house and kind of high 50s, low 60% of assembly we do in-house and that moves around with the acquisitions as Steve noted.

Yes. So we have continued to bring work from SMSC inside. We have closed Supertex test facility in Hong Kong. All that is running into Microchip. So yes, case-by-case basis. When it runs outside, we try to bring it inside for lower cost.

Our next question will come from Harsh Kumar with Stephens.

Steve, I was wondering if there were any bright spots in terms of end markets, some markets that actually held up better both as far as March as well as maybe for the guidance.

I think it's fashionable to say IoT. That's what I would say probably. That's -- it grew 35% year-over-year, and I'm sure it did very well quarterly, too.

That's fair. And then I think you, in response to a previous question, you had talked about linearity of the March quarter. We found that very helpful. I was wondering, not specific to this June quarter but historically, maybe you can give us an idea how typically the June quarter trends for you guys last couple of years.

In terms of monthly linearity?

Yes, just any kind of color you want to give us.

We'll say June quarter is probably the most linear quarter. The March quarter is the most nonlinear quarter because of the Chinese New Year. Then when you get to September quarter, you have August, which is weak usually because of holidays in Europe and parts of Asia; and December quarter, you have holidays at the end. So June quarter is actually the most normal linear quarter, and I don't think we expect anything different.

And Chris Danely with Citi.

So I guess, if you just take a step back, can you compare this soft patch to what happened late last year or maybe even 2012, 2011? Do you think it's similar, better, worse? Do you think there's any chance that things could get worse? Or did we just kind of fall down a month ago or a few weeks ago and now we're kind of bouncing along the bottom and waiting for things back up? Any perspective there?

I think, Chris, I resigned from my industry forecasting job last year. Maybe could you...

Good luck with that. I guess, as my follow-up, how about inventory out there with distributors and your customers now that demand is a little lower? Do you expect some sort of inventory burn this quarter? Or do you feel like everybody is pretty comfortable out there?

So I think our distribution inventory is not high. Number one, we don't take it as a revenue, so it doesn't really matter as much. But if the distribution inventory becomes too high and then they correct it, we can have a downstream effect than if we have to lower our own inventory. Our own inventory is in the lower side and we like it higher. The distribution inventory is in the middle. So when you combine those 2 inventories in that high, if anything, they're in the low side despite the weakness, despite the guidance. So based on current numbers, our inventory is still really on the south or the middle. So I don't think the inventory is a problem. We're working very hard in our factories to make more product for our customers and have expedites and challenges and others with the mix being very complex, having closed down Supertex fab, accommodating all those products in our fab. Manufacturing is challenging and inventory -- high inventory is not an issue right now.

And John Pitzer with Crédit Suisse. John William Pitzer - Crédit Suisse AG, Research Division: Do you mind I give my shot and try to bring you out of retirement and kind of follow up to Chris' first question? I'm just kind of curious, you're guiding flat up sequentially for the June quarter. A little bit below normal seasonal patterns, but you guys are comfortable enough to continue to build inventory. I'm just kind of curious, if you thought this was going to be prolonged or deeper, would you do anything around utilization? Would you not build inventories? The fact that you're willing to build some inventory, kind of a test inside, if you think this is more of a soft test than anything else?

Well, I mean, we really aren't making a call either way. I mean, our manufacturing is running normal. If you look at our December quarter, we have holiday related shutdown we have to do to really do the maintenance on the facilities. So you lose a few days, and it's usually tied on total output in December quarter, and in the March quarter you get the full days. And related to that, we're working hard to build a few days of inventory. We said 3 to 9 days. Let's see at the end it comes out to be 4 or 5 days. That would still be only 114, 115 days of inventory on the lower end of 115 to 119. So like I said, in answer to Chris' question, I don't really think we have conviction either way to drive the inventory much higher and much lower. We're largely basically trying to get to the middle of our range and only going to get to the bottom of our range probably. John William Pitzer - Crédit Suisse AG, Research Division: That's helpful, Steve. And then maybe as a follow-up, Ganesh, really appreciate the kind of the more detailed definition of IoT and you talked about a lot of the discrete IP and technology you guys provide. I'm kind of curious, to what extent you guys are providing that as a turnkey solution when you think about microcontroller, analog, wired/wireless connectivity? How do we think about kind of -- how much of your IoT revenue is complete turnkey solutions versus discrete? And what's the opportunity to go to turnkey? And if you could talk a little bit about security within that IoT bucket as well, because clearly, there's a school of thought that secure and connected is what defines IoT. I'd be kind of curious as to how you guys thinking about the security issue?

Okay. Good questions. So when we go to market, there are obviously many customers for whom we can provide the complete bag of solutions: analog, microcontroller, wireless or wired, whatever connectivity they want to have. In other cases, we may only win a portion of that, and it gives us a foothold of the customer to come back and look at their next generation of designs. So I don't have a good split of which way it is across so many different customers there are. But our obvious go-to market is doing as much as what's on the board that constitutes both the smart side of the equation as well as the connectivity side of the equation. And your question on security is a very good one as well. So we know it's a key part of many, many of these applications. We enable that security, in some cases, through hardware that's built into our products in combination with software that we make available as well. And I think there's a lot more coming in front of us from the industry at large on how to ensure that as IoT fills out the range of applications, the security part of it continues to be addressed sufficiently. So yes, any time you have a node that is connected, its security is a key part of what we need to think about in the solution we deliver.

And with no additional questions in the queue, I'd like to then turn the floor back over to our speakers for any additional or closing remarks.

Okay. We want to thank everyone for attending our conference call. After a little break, at 5:45 Eastern Time, we'll be starting our second conference call, which is regarding the acquisition of Micrel, and we'll hear some of you again on that call. So thank you.