Good day, everyone. Welcome to this Microchip Technology Second Quarter and Fiscal Year 2019 Financial Results Conference. As a reminder, today's conference is being recorded. At this time, for opening remarks and introductions, I'd like to turn things over to Mr. Eric Bjornholt. 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. In attendance with me today are Steve Sanghi, Microchip's Chairman and CEO; and Ganesh Moorthy, Microchip's President and COO. I will comment on our second quarter fiscal year 2019 financial performance, and Steve and Ganesh will then give their comments on the results, discuss the current business environment, as well as our guidance, and provide an update on our integration activities associated with the Microsemi acquisition.
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 in 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 want to remind investors that during the quarter ended June 30th, 2018, we adopted a new GAAP revenue recognition standard, which requires revenue to be recognized at the time products are sold to distributors versus our historical revenue recognition policy, where revenue on such transactions were deferred until the product was sold by our distributors to an end customer.
We are not able to provide guidance on a GAAP basis, as we are not able to predict whether inventory at our distributors will increase or decrease in relation to end market demand, and this is not how we manage our business. As evidence of this uncertainty, in recent years, we have seen net inventory at our distributors increase or decrease by a significant amount in a single quarter. Our non-GAAP revenue is based on true end market demand, in which we measure the revenue based on when the product is sold by our distributors to an end customer. We will continue to manage our business and distributor relationships based on creating and fulfilling end market demand. All of Microchip's bonus programs will continue to work based on the amount of revenue earned from fulfilling end market demand.
Therefore, along with the GAAP results based on sell-in, we will also report our non-GAAP results based on sell-through revenue recognition. I will now go through some of the operating results, including net sales, gross margin, and operating expenses. I will be referring to these results on a non-GAAP basis using revenue based on end market demand and expenses prior to the effects of our acquisition activities and share-based compensation. Non-GAAP net sales in the September quarter were $1.513 billion, just above the midpoint of our guidance and up 24.4% sequentially from net sales of $1.217 billion in the immediately preceding quarter. We have posted a summary of our revenue by product line and geography on our website for your reference. On a non-GAAP basis, gross margins were 61.7%, operating expenses were 23.4% of sales, and operating income was a record $579.3 million and 38.3% of sales.
Non-GAAP net income was a record $454.6 million. Non-GAAP earnings per diluted share was a record $1.81, which was $0.07 above the midpoint of our guidance of $1.74. On a GAAP basis, net sales in the September quarter were $1.433 billion. GAAP revenue was impacted by a significant reduction in the Microsemi distribution channel during the quarter, resulting in Microsemi distributor months of inventory being down to 2.6 months. We believe at the current levels that distributors are holding the amount of inventory needed to support end market demand, and that the inventory levels are in line with the levels maintained by our distributors for our historical business. GAAP gross margins were 48.1% and include the impact of $3.9 million of share-based compensation, $184.4 million of acquisition-related and acquired inventory valuation and other costs, and $56.1 million impact from changes in distributor inventory levels.
Total operating expenses were $586.6 million and include acquisition and tangible amortization of $169.9 million, special charges of $18.2 million, $6.6 million of acquisition-related and other cost, and share-based compensation of $37.5 million. The GAAP net income was $96.3 million or $0.38 per diluted share and includes one-time tax benefits of $115.6 million related to a variety of matters, including tax reserve releases due to audit settlements and statute of limitations expiring, tax reform and transition tax refinement, and intercompany restructuring of intellectual property. The non-GAAP cash tax rate was 3.5% in the September quarter, and we expect a similar rate for all of fiscal 2019.
We expect our non-GAAP cash tax rate for fiscal 2020 and fiscal 2021 to be 5% or less, exclusive of the transition tax, any potential tax associated with restructuring the Microsemi operations into the Microchip global structure, and any tax audit settlements related to taxes accrued in prior fiscal years. We have many tax attributes and net operating losses, tax credits, and interest deductions that will keep our cash tax payments low. The transition tax for the combined Microchip Microsemi group is expected to be about $364 million, was recorded last fiscal year, and will be paid over eight years. We have posted a schedule of our projected transition tax payments on the investor relations page of our website. For GAAP purposes, we had a significant tax benefit in the quarter for the variety of reasons I discussed earlier. Moving on to the balance sheet.
Our inventory balance at September 30th, 2018, was $836.7 million, including $120.1 million of inventory markup from Microsemi required for GAAP purchase accounting. Excluding the inventory markup, we had 117 days of inventory at the end of the September quarter, which was down two days from the prior quarter levels. Inventory at our distributors in the September quarter were at 37 days, compared to 40 days at the end of June. As indicated earlier, we believe that our distributors are holding an appropriate level of inventory to support end market demand. The cash flow from operating activities was a record $487.6 million in the September quarter. As of September 30th, the consolidated cash and total investment position was $464.2 million. We paid down $501 million of total debt in the September quarter, and the net debt on the balance sheet reduced by $315.5 million.
At September 30th, our debt outstanding includes $3.1 billion of borrowings under our line of credit, $2.733 billion of Term Loan B, $2 billion in high-grade bonds, and $4.481 billion of convertible debt. Our net debt to EBITDA, excluding our very long-dated convertible debt that matures in 2037 and is more equity-like in nature, was $4.9 billion at September 30th, 2018. Our leverage is higher than we originally projected, primarily due to lower EBITDA from the Microsemi business, driven by needing to correct distribution inventory levels through lower shipment activity in the month of June, as well as in the September 2018 quarter. As indicated earlier, we believe the distribution inventory correction for Microsemi is essentially complete.
We are committed to using substantially all of our excess cash generation beyond our dividend payment to reduce our debt levels. We expect our debt levels to reduce significantly over the next several years. Our net leverage metrics are based on 12-month trailing EBITDA, which will continue to provide some headwinds due to the significant distribution inventory reductions, which caused our shipment activity to be significantly less than end market demand for the past two quarters, as well as our guidance for the December 2018 quarter. Capital expenditures were $72 million in the September 2018 quarter. We expect about $50 million in capital spending in the December quarter, and overall capital expenditures for fiscal year 2019 to be about $230 million-$250 million.
We continue to add 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 are currently outsourced. These capital investments will bring some gross margin improvement to our business, particularly for the Atmel and Microsemi manufacturing activities that we are bringing into our own factories. Depreciation expense in the September quarter was $47.2 million. I will now turn it over to Ganesh to give his comments on the performance of the business in the September quarter and provide an update on some of the Microsemi integration activities. Ganesh? Ganesh?
Thank you, Eric, and good afternoon, everyone. As I review our product line performance, please bear in mind that the September quarter has a full quarter of contribution from Microsemi, while the June quarter only had a month of contribution from Microsemi. This, along with the fact that Microsemi revenue only maps into four of the six product line reporting categories we have, will, in some cases, distort the quarter-over-quarter comparisons of revenue by product line. Taking a closer look at microcontrollers, our microcontroller business was sequentially up 12.6% as compared to the June quarter. This benefited from a full quarter of Microsemi contribution in September, as compared to a partial quarter for June. Our September quarter microcontroller revenue annualizes at almost $3.3 billion, putting us within striking distance in the next few years of the two industry players who were ahead of us in the 2017 microcontroller rankings.
Our microcontroller portfolio continues to expand with several new product introductions. Our roadmaps remain strong, and our design and funnel is robust. We see all these as positive indicators for future growth. We believe we have the new product momentum and customer engagement to continue to gain even more share in the future. Moving to analog. Our analog business was sequentially up 33.6% as compared to the June quarter. This too benefited from a full quarter of Microsemi contribution in September, as compared to a partial quarter for June. Our September quarter analog revenue annualizes at over $1.75 billion, firmly in the top 10 of analog semiconductor players. With Microchip 2.0 and our total system solutions approach, we expect to continue to grow and gain further analog market share.
Next up is our FPGA business, which came to Microchip through the Microsemi acquisition. Our FPGA revenue hit an all-time record, even after going back through the Microsemi and Actel history. Our low power mid-range PolarFire family continues to garner strong market acceptance, with revenue more than doubling sequentially as compared to the end market consumption in the full June quarter. Despite being a new product category for Microchip, we're optimistic about the prospects for the FPGA product line based on the highly defensible markets and applications they're designed into, as well as the intense design win focus and broad-based market application and customer focus required. These requirements are very similar to Microchip's microcontroller business requirements, and therefore, we expect the FPGA product line will very nicely leverage our capabilities. Moving next to our licensing business. This business was sequentially up 40.3% as compared to the June quarter.
We achieved an all-time record for our royalty revenue. Additionally, our results also benefited from the sale of a patent license for a specific set of patents that can be used in a non-competitive fields of play. We did anticipate this patent license to close in the September quarter and included it in our guidance. We continue to retain indefinite rights for these patents for the fields of play that are of interest to us. Even without the patent license sale, the September quarter licensing business revenue was sequentially up from June. Let me share some background as to our thinking in regards to patent licensing. We have inherited a substantial patent portfolio from the companies we acquired over the last 10 years.
Several quarters ago, based on licensing requests we were receiving, our licensing business unit embarked on a strategy to monetize certain patents which have value to players in non-competitive fields of play. In all cases, we retain rights to use these patents in our products as well. The first results from this monetizing effort is what we saw in the September quarter results. There is a second such patent license sale for a specific set of patents, which is in the advanced stages of negotiation and is included in our guidance for the December quarter. Investors should expect that the revenue contribution in the future from this effort will be a little lumpy from quarter to quarter. Our memory business was sequentially down 3.8% in the September quarter as compared to the June quarter.
Finally, our MMO, which stands for multi-market and other business, was up 90.7% sequentially as we had a full quarter of Microsemi contribution in the September quarter, as compared to just a month in the June quarter. On a combined company basis, which has a full quarter of Microchip and Microsemi, our second quarter fiscal 2019 non-GAAP end-market demand revenue of $1.51 billion was split across the six product lines we report as follows. Microcontrollers were 54.2%, analog was 29%, FPGA was 6%, memory was 3.1%, licensing was 2.5%, and MMO is 5.2%. We expect this will be the approximate forward-going revenue percentages by product line, and that quarter-to-quarter changes from here onwards will be relatively small. Now an update on the Microsemi integration progress. The business unit integration continues to progress well.
We are restructuring underperforming businesses while implementing the strengths and best practices from both companies to drive improvements across the board. The key Microsemi business unit leaders we retained continue to run their businesses and are adapting to the Microchip culture and business expectations. New product development and design win momentum are continuing to happen as planned. The sales integration is also progressing well. We have converted the Microsemi sales team to a Microchip style non-commission sales approach. Teams from classic Microchip and Microsemi are working collaboratively as one team to drive new opportunities and pursue product cross-selling opportunities. Our internal reference designs are increasingly taking advantage of our combined total system solutions approach. A comprehensive analysis of our new channels and channel partners from the Microsemi acquisition was completed, and we're making the adjustments required for an effective combined company approach.
The operations integration work is an enormously complex undertaking, as little to no integration of prior Microsemi acquisitions had taken place. Because of the size of the business and the absence of ERP integration by Microsemi, the business system and operation integration will be done in phases. The first phase for one of the business units went live on November 1st successfully. Further phases have been planned with a quarterly cadence of go live events for many quarters. We expect the overall operational integration will take about two more years to complete. From a factory standpoint, the initial plans for insourcing some of what Microsemi was outsourcing have been completed and prioritized actions to execute these plans have commenced. In the G&A functions, we eliminated some more of the support organizational redundancies, and more will happen after we get farther along with our business system and operational integration plans.
Let me now pass it to Steve for some comments about our business and our guidance going forward. Steve?
Thank you, Ganesh, and good afternoon, everyone. Today, I would like to first reflect on the results of the fiscal second quarter of 2019. I will then provide update on our progress at Microsemi. I will then provide guidance for the fiscal third quarter of 2019. Our September quarter financial results were good. Our consolidated non-GAAP net sales came in just above the midpoint of our guidance that we issued on August nine, 2018. Our consolidated non-GAAP gross margin was strong at 61.7% of sales. Microchip classic non-GAAP operating margin was an all-time record. Consolidated non-GAAP operating profit, including Microsemi, was 38.3% of sales and near the high end of our guidance. Our consolidated non-GAAP EPS was an all-time record at $1.81, and near the high end of our guidance. There was a strong EPS contribution from both classic Microchip and Microsemi.
The accretion from Microsemi exceeded our original guidance of $0.15 for the September quarter. This was based on non-GAAP revenue, which is based on distribution sell-through and is representative of real market demand. On non-GAAP basis, this was also our 112th consecutive profitable quarter. Let me provide you some update on the progress we have made with the Microsemi integration. First, distribution inventory. We had told you last quarter that we will complete a distribution inventory reduction in two quarters, September and December. I am pleased to report that with a lot of work by the combined teams of Microchip and Microsemi, and in cooperation with the distributors, we almost completed the inventory reduction in one quarter. The distribution inventory at the end of September was 2.62 months, down from about four months prior to the closing of the acquisition.
This level of inventory, we believe, is the level distributors need to support end market demand, and is in line with the level maintained by distributors for Microchip classic business. This was accomplished without any negative impact on distribution sales out. The non-GAAP net sales from Microsemi, based on direct shipments and distribution sales out, was an all-time record. Second is termination of deals. Since closing the Microsemi acquisition on May 29, 2018, we terminated deals, discounts, and subsidies that were used prior to the acquisition in connection with sales into distribution and contract manufacturers. For example, we canceled the Arrow Supply Assurance program concerning sales into the channel of end-of-life product. At acquisition close, there was approximately $47 million of inventory in distribution under this Arrow Supply Assurance program. We won't take any inventory back, we will not ship any more into distribution under that program.
We also canceled the interest subsidy that was offered to some distributors to compensate them based on the inventory they held. We also canceled the production incentive program, under which there was a discount offered to contract manufacturers to take additional inventory. Cancellation of this program has had no negative impact on contract manufacturers willingness to purchase the product they need. Finally, there were a number of OEM customers that were transferred to distribution, and then the inventory shipped into distribution. This resulted in lower margin for Microsemi. As of today, we have transferred many of those customers that account for about $110 million of annualized revenue back to direct customers. Overall, we are pleased that we were able to reduce the Microsemi distribution inventory very rapidly.
There was one business unit called the High Reliability Discrete Products business unit, where the inventory at the end of June was very high at 8.6 months. At the end of September, we had brought it down to 6.7 months. We are continuing to reduce the inventory for this business unit further. The shipments are now quite linear. In the four quarters prior to being acquired by Microchip, about 57% of Microsemi's quarterly sell-in revenue was shipped in the last month of the quarter. In the September 2018 quarter, only 31% of sell-in revenue for Microsemi products were shipped in the third month. Next is Microsemi internal inventory. It is obvious that as we reduce the distribution inventory, the inventory that Microsemi would grow substantially unless we took corrective actions in the factories and subcontractors by cutting production targets.
That is what we did. Fortunately, a very large amount of Microsemi business was done from wafer foundries and assembly and test subcontractors. We aggressively cut wafer starts at the foundries and loadings at assembly test subcontractors. Microsemi had several small fabs and assembly locations owned by Microsemi We cut loadings in these factories quite substantially, some of them as much as 50%. We let go temporary workers in many of these facilities and implemented temporary plant shutdowns to control inventory. Microsemi did not produce a high percentage of their products inside, so the gross margin impact was fairly small. We maintained over 61% overall gross margin for Microsemi. We believe that the underloadings of the factories got somewhat counterbalanced by the discontinuance of all the discounts. As I have said before, the results are going to be a bit noisy for a couple of quarters.
At the end of September, the overall internal inventory at Microchip, including Microsemi, was 117 days, down two days from 119 days at the end of June quarter. We did a great job in controlling internal inventories. This may seem all too easy after the fact, but there was a tremendous focus and amount of work that went into reducing the distribution inventory and at the same time ensuring that our internal inventory does not balloon up. We have multiple joint teams from Microchip and Microsemi, some focused on distribution inventory reduction, others focused on controlling internal factory production, and others managing reduction of foundry wafer starts and assembly and test loadings. While addressing the inventory has provided some headwind on the distribution sell-in driven GAAP revenue, including sales to contract manufacturers, the non-GAAP revenue for distribution, which is based on sell-through, has not been affected.
I am pleased that Microsemi business achieved a record non-GAAP net sales based on end market demand. The second issue we highlighted last quarter was extravagance. In this area, we have been moving aggressively to implement Microchip's frugal expense culture. We have shut down the executive floor at the Microsemi headquarter building, canceled the private jet account, terminated or sold many of the sports sky boxes, canceled golf and auto racing sponsorships, and substantially pruned many memberships. As we reduced inventory and spending, Microsemi business has continued to do well. This was a narrative from the beginning that Microsemi has very good engineering teams and has very good products. The customer sockets are sticky, and there are good end market opportunities for the combined company. Our strategy for the better part of this decade has been to buy businesses and turn them into world-class performers in the likes of Microchip.
Here, we are starting with excellent products, excellent gross margins, and excellent engineering teams. With the distributor and contract manufacturing inventory reduced and with Microchip's operating expense approach, we are optimistic about achieving our long-term targets for accretion from Microsemi. Far, we are ahead of our original targets. I also want to provide you one more figure. In the September quarter, we paid $501 million of our debt. This was accomplished by squeezing all the cash out of nearly 100 subsidiaries around the world. We could not have done it without the new tax law. Despite the large debt paydown in the September quarter, the debt leverage has not moved, though primarily because the EBITDA has dropped, driven by lower GAAP revenue of Microsemi and also some softness in the Microchip classic business. I also want to provide you update on our classic Microchip business.
All of the issues that I highlighted last quarter that were built into our guidance essentially came true. I had described four concerns. First, long lead times for some of the passive components. This issue did affect some sales as we expected, since the customers could not complete the whole kit. Number two, tariffs. While we were one of the first to highlight the risk of tariffs and trade war, this issue caught investor and analyst attention during the quarter for the whole industry. We saw significant impact on our business during the quarter due to customer concerns about tariffs, and we saw significant weakening of our bookings. We believe that tariff and related customer concerns will continue to be an issue. Number three, ZTE. We saw some impact for not being fully able to ship to ZTE for both Microchip and Microsemi.
Despite the release from the Justice Department, there has been some demand destruction at ZTE. Number four, bitcoins. The bitcoin business has essentially evaporated. It was about 1% of our business supplying microcontrollers and power management products for the power supplies as well as Ethernet controllers. Now I will provide you guidance for the December quarter. The guidance we provided for the September quarter, which reflected our caution on business conditions, turned out in retrospect to be spot on and was a harbinger for broader industry weakness. We continue to be cautious about the outlook for the December quarter. We see weak market conditions for automotive and industrial, our two largest markets. Automotive business is negatively impacted by emission testing bottlenecks in Europe as well as weakness in China. The industrial business is negatively impacted by tariffs.
On the other hand, communication, data center, and aerospace and defense businesses are seeing strength. With all this commentary, we expect our total non-GAAP net sales to be down 5%-10% sequentially. We expect our non-GAAP gross margin to be between 61% and 61.5% of sales. We expect non-GAAP operating expenses to be between 24.9% and 25.4% of sales, and we expect non-GAAP operating profit percentage to be between 35.6% and 36.6% of sales. We expect non-GAAP earnings per share to be between $1.49 and $1.64 per share. Given all the complications of accounting for the acquisitions, including amortization of intangibles, restructuring charges, inventory write-up on acquisitions, and changes in distribution inventory, 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. We request that the analysts continue to report their non-GAAP estimates to First Call. With this, operator, will you please poll for questions?
Certainly. At this time, if you do have a question, that will be star one. We do ask that you please initially limit yourself to one question due to time constraints. Again, that will be star one for questions. We'll hear first today from Christopher Rolland with Susquehanna.
Hey, guys. Great quarter in light of some of the headwinds that we're having here. Great update on Microsemi as well. Elaborate a little bit more on some of the end markets there. You had a little bit more detail than others did, perhaps on auto. Maybe talk about industrial as well. Some bright points, maybe within data center and comms, maybe that's your optical business that you might be pointing to. Any more details there would be great.
Well, if you look at the auto business, we highlighted two areas. One is, there was a new emissions standard implemented in Europe effective September 1. There was a one-year notice on it almost, from September 1 of the prior year. However, testing agencies couldn't build up the capacity to test all the car production, so there was a significant bottleneck of cars waiting to be tested, which brought the European car production down. The whole China economy has been weak due to various reasons. Some incentives were taken away, the concerns about tariff in Chinese production, and all sorts of other reasons China automotive market has been weak. In the industrial area, the concerns are mainly tariffs, both driven by Chinese end products coming to U.S., which have a 10% duty today, going to 25% in January.
There are a lot of people concerned about whether they will be competitive or not in selling products in U.S., and because of weak Chinese economy, industrial is weak in China, too. The three markets which are doing stronger are the data center, communication, and defense and aerospace. If I take them in the reverse order, defense and aerospace, I think after years of drawing down defense spending, the defense and aerospace spending has been on the way up. We are very broadly designed in into essentially every single weaponry system, every single aircraft, helicopters, and others. That part of the business is doing well. The data center business, I think you probably have seen some other companies with higher exposure to data center, cloud, hyperscalers and all that. That part of the business has been stronger.
That business did well last quarter and is stronger relative to the automotive and industrial and other businesses. Third is the communication. I think we also see the communication business doing well.
You guys have talked about this potential weakness that we're seeing from most guys out there in the market a little bit earlier. Some, however, haven't guided for much weakness at all. Do you think that perhaps they're going to see it a little bit later and you might be working through this process a little bit earlier? Then are there any signs of stabilization in late October or early November?
Well, we're not going to talk about what other companies have said or what may happen to other industry players. I think you should ask them. Regarding Microchip, we first saw some impact in June quarter. What we were seeing in June, based on that, we had a fairly conservative guidance for the September quarter. Most people didn't talk about it till the results for the September quarter. We were four months earlier before we saw it. That's not first time. Numerous times in the past, we have seen the impact of industry events about three, four months ahead of the others see it. We see it early and we always end early. If we talk about the industry four months earlier, nobody tends to believe it because nobody else is talking about it, and you're not seeing it.
Therefore, we start talking about the industry, we just talked about some factors that were affecting our business. It turned out we were correct again, and everybody else saw it later. Now on the other side, we always come out earlier, but whenever we tell you that won't get believed either because others are not seeing it yet. We're not really ready to talk about the other end of the cycle yet.
Great. Thanks so much, Steve. Appreciate that.
Welcome.
We'll hear next from Harsh Kumar with Piper Jaffray.
Yeah. Hey, guys. Thanks for the MSCC update. Congrats on preserving EPS in a tough market. Maybe one for Eric. Eric, I think you had cited a goal of 0.7 turns of delevering, I believe, for this fiscal year. Can you maybe talk about if that gets impacted at all with this environment, and currently based on what you're seeing? Maybe one for Steve, and I want to ask a follow-up after this. Microchip was usually up in March. I think mostly MSCC was just all over the place. Maybe could you help us think about how we can think about modeling March, just from your initial read on what you've seen in the business?
Sure. I'll take the deleveraging one first and then pass it back to Steve. We made really good progress in terms of paying down debt, as both Steve and I mentioned in our prepared remarks. We took down $500 million of total debt on a net basis, $315.5 million. Did a really good job there of using all the cash that we could to pay down debt. We will continue to use substantially all of our excess cash generations after dividends to reduce our debt levels, and we expect our net leverage to decline dramatically over the next several years.
Now, with the correction of the Microsemi distribution inventory having a significant impact on our EBITDA in the June and September 2018 quarters, and then combining that with our guidance for the December 2018 quarter, our trailing 12-month EBITDA will have significant headwinds over the next year, and our net debt to EBITDA will not trend down as fast as we had indicated back in August. This is a very good cash flow business. Like I said, we'll continue to pay down debt. In the current quarter, we probably expect that debt paydown to be somewhere in the $160 million range. A broad range would be $150 million-$200 million, we'll have to see how things trend here over the next few quarters to give you a longer-term forecast.
Got it.
I'll pick up the second question you had, which was regarding the March quarter. I remind investors that when we bought Atmel, it was not a small acquisition either. Atmel was about 40% to half of our business. After we integrated Atmel, we didn't understand Atmel's seasonality for a while because companies on their own, their prior seasonality was based on selling, however they worked with the distributors, the whole end market mix was different. After we had Atmel for an entire year or more than a year, we fully understood the impact of Atmel's seasonality on our business to be able to guide properly. Same is the situation now with Microsemi. We don't fully understand Microsemi's seasonality. We haven't had a March quarter with the Microsemi in it.
We do not consider the March quarter on Microsemi's clock to be a meaningful indicator for us regarding seasonality, because that business was managed with a selling-driven mentality with a significantly high distribution inventory. Our guidance will be based on real market demand. The end markets for Microsemi are quite different. They're stronger in defense, stronger in data centers, stronger in communication, and almost no automotive, very little automotive and all that. I'm not ready to talk about the March quarter in terms of guidance, until we have some more experience on our belt in understanding their seasonality as well as being able to gauge the impact of the current correction when it is ending and how its impact would be in the March quarter.
That's fair. Thanks, Steve.
We'll hear now from Craig Ellis with B. Riley.
Yeah. Thanks for taking the question. I'll start with a clarification. With respect to the strong performance on the Microsemi inventory cleanup in the quarter, did that have a particular impact on any of the product segment revenues that you reported, or was the impact spread about equally among MCUs, analog, and other businesses?
It really did not have an impact on the sales out activity that Ganesh talked about, breaking out by product line. It was really just more of correcting the inventory sitting in distribution. I don't think there was really any end market impact from the actions that we took.
Basically, we haven't broken out the sell-in data by product line. We only have broken out the sellout data by product line, and that did not change based on lowering the sell-in and cleaning up the inventory.
Yeah. We have the sell-in data in our public filings, in our 10-Q, but that doesn't really give you an indication of where the distribution inventory drain was by product line.
Got it. Then the follow-up, there were a couple of references to accretion, and I just wanted to clarify, what was the exact amount of Microsemi earnings accretion in the quarter? Steve, are you retaining all the prior synergies and accretion targets, the $300 million in synergies as well as the $0.75 plus in year one and the $8 in, I believe, fiscal 2020 earnings?
Yeah. We're not breaking out the exact numbers of Microsemi accretion because we're kind of managing it as one company. We gave you an anecdote that the accretion in the September quarter was higher than the $0.15 for the September quarter we had originally guided. We feel very confident that going out of the first year, the accretion will be higher than the $0.75 that we guided, and we remain committed to a longer-term accretion of, or earnings per share of $8 that we guided before. There's a short-term impact based on this business cycle environment, whatever, one, two, three quarters it lasts, it doesn't really have a long-term impact over three, four years, because usually, the bounce back gets you back to where you were.
Thank you.
From SunTrust, we'll hear from William Stein.
Great. Thanks for taking my questions. First, really more of a housekeeping one. The share count was a little lower than expected in the quarter and in the outlook. Is that just owing to share price and the treasury method accounting for average shares in the quarter, or is there something else going on there?
No, that's exactly what it is. We post a schedule on our investor relations page under supplemental financial information that walks through what the share count approximately would be based on various prices. As you know, we have several convertible debt instruments outstanding, the share dilution from those has gone down with the stock price falling.
That helps. Thank you. The other is to address the GAAP to non-GAAP revenue adjustment. I think it was $81 million in the quarter. You noted that the channel inventory for Microsemi now approximates the sort of heritage Microchip business channel inventory. It sounds like maybe there's a little bit more work to do, but it's mostly done. While I understand that disti inventory is going to fluctuate somewhat quarter to quarter, after this $81 million delta that we saw in the quarter you just posted, going forward, should we expect this number to sort of fluctuate around zero or average around zero over time? Is that the right way to think about it?
I'll take a stab at answering that, Steve or Ganesh can chime in. We're really pleased with the progress that we made in getting the Microsemi disti inventory down to 2.62 months, which is right in line with where Microchip has been historically. We really are not able to project what distributors are going to do. Some of that is going to be based on environment, it's going to be based on working capital needs, it's going to be based on lead times. There's so many things that factor into it. We have 120 or so distributors that we work with worldwide. We don't manage our business in a way where we get that type of visibility from them and really are driving for end market demand forecast versus sell-in.
It wouldn't surprise me in an environment like this that distributors are going to be very tight with their working capital. Again, for us to be able to provide any sort of forecast of what it's going to do is quite difficult, and we don't feel we have capability to do that. Steve, do you want to add anything?
Well, what I would add is, we do not put any effort into managing our business to put the inventory into distribution. We put our entire effort into creating market demand, which pulls the inventory out of the distribution and then allow the asset managers at the distribution to buy the product they need to serve the end market demand. As the quarter end is approaching, we don't look at sales out is this and the sell-in is this, and we got to go make another $10 million a deal, so sell-in is lower than the sellout or to hold back product because it's going to go upward or lower. We don't manage both those numbers. We only manage one number, which is sales out. Like I told you last quarter, the standard has changed. The GAAP standard is sell-in.
I told you that before, we disagree with that. We have to report a GAAP sell-in number, which we do, and we'll continue to do it. All our measurements, all our bonus programs, what board manages us is to a market demand driven number that we report as a non-GAAP revenue. Sell-in would be whatever distributors want based on their own asset management, their cash flow needs, lead times, hundreds of things go into what they do with that inventory. We don't manage that.
Understood and appreciated. Thank you.
Thank you.
We'll hear now from Craig Hettenbach with Morgan Stanley.
Yes, thank you. Steve, understanding you don't want to talk about the March quarter and seasonality yet, just for the core business. Your commentary about kind of cautious into December, what type of signals are you looking for in terms of whether it's the distribution or customers, in terms of how far along we are for the correction here?
Well, we interact with a large number of customers and distributors over the course of this quarter. Based on that collective engagement with hundreds, if not thousands, of customers in all three geographies, U.S., Europe, Asia, we will have an assessment by the time we announce our December quarter results, which will be in early February. If there is meaningful information that develops before that and gives us enough data points that we can kind of make a call, then we'll use the opportunity at one of the conferences to make comments on it. Fresh coming from the elections yesterday, with 1% of the precincts reporting, we cannot make the call on the election.
Understood. Appreciate the color on the end markets, where you're seeing strength and weakness. Any thoughts on consumer appliances? That's a market that we've seen some potential slowing in China as well. Do you have any thoughts there?
That part of the business, appliance, consumer part of the business, is weak also. I kind of put it in industrial. People who make air conditioners and appliances and that's kind of industrial kind of business. You may call it consumer, yes, that is weak also.
Okay. Thank you.
You're welcome.
From Stifel, we'll hear from Kevin Cassidy.
Thanks for taking my question. With reference to China, have you seen any change in behavior on new designs where maybe because of the tariff and the trade war, if there's a preference for non-U.S. microcontrollers as an example, rather than Microchip?
We haven't really seen a lot of impact in reality. There is a lot of talk about it, but it's not changing the funnel size in any way, partially because very large portion of Microchip business is proprietary. Such functionality with such low power or such features or performance for that price really is not available in any local kind of Chinese parts, which are really fairly low end. I don't think we have seen any meaningful difference in the funnel size or design activity. There's a lot of talk about it.
Maybe I didn't understand the question directly, but the tariff impact on our products directly is very, very small, right? It's the products that our customers are building with our products where the tariff applies. I think it's probably too early for us to see any real impact on the design side, Kevin.
If they were to use a Chinese microcontroller or analog part to build an end product, an appliance, a washer or dryer or whatever, that will have duty also coming in here. Not designing a Microchip product does not evade the duty coming into U.S. That is a bigger impact.
Right. It's just more whether there's an anti-U.S. view from a design point of view. Even if it's, say, an ST Micro, something from Europe rather than a U.S.-based. That's what I was trying to get at.
I think your question is the right one, and we hear that, but like I said, fortunately, we're not easy to design out.
Okay, great. Thank you.
We'll hear next from John Pitzer with Credit Suisse.
Yeah, guys, thanks for letting me ask the questions. Congratulations on the good results. Steve, relative to the industrial weakness, you're calling out tariffs. I know this is probably an impossible question to answer, but I'm going to ask your opinion anyhow. To what extent do you think customers are reacting to the 10% tariff that's already been levied? To what extent do you think that they're actually trying to be anticipatory to the raise to 25% come the beginning of the year? I'm just trying to get a sense as to whether or not you think there's another shoe to drop or not. My second question, you did a great job kind of talking about the distribution inventory at Microsemi. I'm kind of curious about direct customer inventory. What did you see as the quarter unfolded? Was there any meaningful adjustments you needed to make there?
Are those behind you? To the extent they might be behind you, could there be a potential tailwind, which I think you referred to on the last conference call once this inventory depletion was done?
Let me take the first one first, which is a tariff question. What's happening is, as the percentage tariff and the products which have tariffs on keeps moving, it is kind of creating waves in customer sentiment. Originally, back in June, July timeframe, when there were some tariffs ready to go on in August, some people will try to get ahead of it, build the product, so they bring it to U.S. and avoid the tariff. Others will hold back production. They want to draw down their inventory because they don't really know whether they'll be able to pass the tariff to the customers or they will have to eat that entire tariff. It kind of creates a wave both ways, and that was for the 10% tariff.
Through the quarter of September, customers were essentially cutting down production, drawing down their inventory because they felt that they'll be uncompetitive passing on the tariff to U.S. customers who had a choice to buy the product made in Taiwan, made in Singapore, made in Vietnam, made in Mexico or wherever. Now very recently, we're hearing the talk that the expectation many of the customers had was, heading towards the elections, the whole thing was going to get settled. You made a call at the UN. Trump was talking so bad about Canada, that Canada wasn't cooperating and all that, and there may be an agreement with Mexico only. One week later, they signed an agreement, announced the agreement with Canada. The thought was similarly, a few days before the election, there may be a settlement with China, and people were holding back.
Well, that didn't happen, now the feeling is, I don't think the settlement is close. Now some of the customers are thinking tariffs are going to go to 25%, and we should pull in and build some of the stuff and bring it to U.S. at 10% tariff rather than bringing to 25%. You could see that the wave's moving all over the place, and 50,000 customers we might have in China are not all aligned. It's very noisy. It's very hard to make decisions, and lead times are short, and we're responding to every customer, and they're whipsawing as we see it. I don't know if that answers your first question.
No, that's helpful. Then just on the direct customer inventory situation at Microsemi.
On the direct customer, the inventory we saw on the direct customer side was with contract manufacturers, because contract manufacturers, it's considered OEM business. It's a direct business. It's not through distribution. There was really not much inventory in the direct customers that we saw. We wouldn't know it either. You don't know every customer's inventory. Where the deals were made, where the discounts were given, were to the contract manufacturers under a program called, what was it called, a program? Production incentive program. We canceled that program, as we canceled that program, contract manufacturers only started taking what they need for their needs, and they drew down their inventory.
Perfect. Thank you.
Again, for questions, that is star one at this time. We'll hear next from Chris Danely with Citi.
Hey, thanks, guys. Steve, I know you're not going to comment on March seasonality. Hypothetically speaking, if March revenue was down, would you be able to keep OpEx flat to down?
Well, we definitely will have the March OpEx down, not even flat to down. It'll be down, because we're still continuing to reduce Microsemi expenses as we do integration. If March quarter is down, as you're supposing, bonuses would be lower, so bonus accrual would be lower. Yeah. I would say in that situation, March expenses will not be flat. They will be down.
Okay, great. For my next question, any sort of comparisons on this correction versus previous corrections? Are you seeing anything in common, or is this completely new?
Well, I think that's more of your department than mine.
I'm just as clueless as you guys are
in 2015, we saw this impact, I believe in August-September timeframe. We made a call, I believe, in late September timeframe that we see China slowing down. It was driven by China, nobody else was seeing it. When the numbers came out, numbers were cut, they were cut again, everybody was down as much, most stocks went down lower than they had gone down that day when we had made the call. Everybody was about four months behind. Same thing is happening this time. I think we saw the impact three to four months behind everybody else saw. That is common. What's driving it today versus what was driving it back then, I think the reasons are a little bit different. Tariffs, maybe lead time becoming shorter. Some are same, some are different. Tariff is a new phenomenon.
Got it. Okay. Thanks, guys.
We'll hear now from Vivek Arya of Bank of America Merrill Lynch.
Thanks for taking my question. Steve, for the first one, can you give us some more real-time sense on production plans? Are you still looking to reduce them further versus seasonality, or do you think they have kind of stabilized at these levels?
Well, if you look at Chris Danely's question, you have to make an assumption on March. We cut back production quite a bit. With that, the internal inventory is in good control. Actually, internal inventory went down by a couple of days. If the business is stable from here, if this quarter marks the bottom, then factories start building back slowly. If there is a further downside, which was an assumption possibly Chris was making, I don't know if he was making that assumption, then you have to adjust to the reality.
I see. On the OpEx side, if my math is right, you had $354 million non-GAAP OpEx in September. I think December, you are guiding to $352 million, kind of flattish. Why are we not seeing more cost synergies, or does it just mean we'll see them later on?
Well, we took a lot of costs out early. I think if you were to combine the Microchip and Microsemi models from back in March or even a year ago, you're going to see that significant expenses have been taken out, and that doesn't mean that there's still not more to come. I think we're actively managing that. Our OpEx came in significantly lower than our guidance for the quarter that we just completed. We're managing based on the environment that's presented to us, and I think we're doing a good job of that.
Got it. Just one quick housekeeping. I think, Eric, you said tax rate under 5%. I believe in the past it was between 3% - 4%. Just for modeling purpose, should we model like 3.5% or 4.5% for the next one or two years? Thank you.
Well, when we gave 3% to 4% before, that was kind of the short term, and we've got now questions coming in terms of what's it going to be over the next few years. It's not a perfect science to forecast this, under 5% or 5% or less is where we're at today. It's not saying it can't be 3.5%, that's a pretty tight range.
I think, you're modeling 100+ subsidiaries around the world and varying tax rates in various countries and interest payments and deductions and all these taxes, and it's an average of all that. It's not a perfect science. It can change based on mix.
Got it. Thank you.
Moving on to Mark Delaney with Goldman Sachs.
Yes. Good afternoon, and thanks for taking the questions. The first question is hoping you could help us better understand some of the trade-offs that you've made with Microsemi's business now that you've had it under your wing for a longer period of time. It sounds like there's been several efforts taken to improve gross margins, and I think that business had run in the low 60%s on a standalone basis. With all the changes, some of the product pruning and things along those lines, do you think it gets to above that 63% range that you're guiding for on a combined basis? Or just toward the higher end once utilization rates normalize?
Related to that, maybe you can help us think about, has there been any share shifts or the size of product pruning that you may have done given some of these changes that you're making to the business model?
I think, Mark, one mistake most people make in looking at Microsemi's gross margin is that they did an acquisition of a company called Vectron, which closed in late November, early December timeframe. It had gross margins in the mid-30s and was $100 million revenue. Its impact was really never seen by investors in a non-GAAP basis. December quarter was fairly short, just one month. March quarter, we announced the deal on March 1. March quarter was really never fully announced based on a non-GAAP.
Right. They filed a 10-Q.
Yeah, they filed a 10-Q.
Didn't have a conference call or
Didn't have a conference call, so you wouldn't really know the non-GAAP margin, taking away all these charges and everything else that acquisitions do. Microsemi gross margin wasn't 63%, driven by all those acquisitions. Margin was below 60%. We have already improved some and will continue to improve the gross margin, but looking for that gross margin to go over 63% would be quite a stretch right now.
Yeah. We, as you said, have margin improvement programs that are in place, and we still have confidence in the combined long-term model of 63%. This last quarter, we were at 61.7%, so it's not like we're light years away from that.
We got to get the factories to full production again. Because of the inventory correction, as you ship less to distribution, all that inventory stays in-house unless you cut production. We did, and we take the under absorption in the current quarter. We don't capitalize it into the inventory. As the production starts going back, you'll have some accretive effect on gross margin.
That's helpful. Then for the second part of that question, obviously, there's a different focus on the types of businesses Microchip will be participating with there. Can you just help us think about the scope of revenue that you may not participate in anymore either because you're focusing on a narrower set of product lines, any share shifts that we should be contemplating, things along those lines so we can think about the right revenue run rate for that business once we adjust for some of these inventory dynamics that have gone on the last couple of quarters?
We said that before. We are going to keep, and we like all the businesses. The combined Microsemi, all businesses together, P&L is very good. There are product lines which are lower gross margin. There are product lines which are higher gross margin than average. There are product lines that grow less than average. There are product lines that are growing more than average. Microchip Classic is no different. Our gross margins are lower on certain Microchip businesses, mix and match overall gross margin is excellent, and that is the case with Microsemi. Every business is creative, we're not going to offload a business just because it's lower gross margin than the average. If it's creative and growing and adding and providing additional attach, is a part of the total system solution. If all the other things are working, it's a good Microchip product.
Yeah. Every business unit leader that we have is tasked at in not just growing the revenue, improving their gross margin and operating margins over time. Within the Microchip system, we work together to make that happen.
Yep. Thank you.
We'll move next to Kristen Siaka with Nomura.
Good evening. Thanks for taking my question, congrats on the great execution of Microsemi. I just wanted to follow up on the OpEx question from before. It looks like the OpEx from this quarter came in a little bit lower than expected, I was just wondering if that was just due to better than expected execution on synergies, or was there something else going on in the quarter there that we should be thinking about?
There was nothing else going on. Although, we would say that since we saw the effect of this downturn coming much earlier than anybody else, we hunkered down earlier. We canceled the majority of the open racks. We cut back on capital. We canceled some orders that we had on capital to the extent we were able to cancel them, and the rest, we slipped them out. We hunkered down, and we kind of batten down the hatches.
Ask our employees to treat the spending as if it's their own money.
Exactly.
Do that on a global basis. It has very good results. I think Steve's right. We just got out ahead of this quicker.
Okay, great. That's helpful. Just moving to the inventory correction from Microsemi and how you completed that about a quarter earlier than you'd stated on the prior call for the most part, what exactly caused that correction to happen faster than expected? Was there something in particular that Microchip did, or was it just better end demand, or what was kind of driving that force?
Let me take that. I think, especially when you buy new business, you have a lot to learn for that business. You're getting to know the people. There are concerns with employees regarding what Microchip could do with various businesses. Let me get this in. Various analysts write about various businesses that we will sell or we should sell. I always ask you, please don't do that because you create problem for us. The employees of that business read those reports. They get concerned that their business is maybe sold or not get focused on. We have said that we're going to keep all the businesses, every time you look at some analysts will make their speculation why data center business doesn't fit with Microchip, or Opto business doesn't fit with Microchip, or something else doesn't fit with Microchip.
All that does is creates concern for our customers and creates concern for our employees. What I'm saying is, we're working very hard on all these items, learning the businesses, keeping the customers at peace, talking to the employees, learning the inventory. When such a large challenge comes up, like the inventory, it just takes an enormous amount of energy forming the teams and focusing the teams to go clean up the inventory. It also depends on how in mix that inventory was. If that inventory was highly out of mix, then you have to continue to ship additional product for which there is demand. If there is lot of slow-moving inventory that we call sludge, then that will take much longer time.
In certain business units, the inventory was in good mix. Certain other business units, inventory was not as good of a mix. Basically, huge amount of work, combined teams working together, creating a goal for everybody to clean up this inventory faster than the two quarters we guided to. I think that's where we are. Also working and getting cooperation from the distributors.
Excellent. Thank you.
It's just a lot of grunt work. There is not much magic there. It's just a lot of hard work.
Yeah. As Steve mentioned in his prepared remarks, there's still a business unit or two that's inventory is higher than we'd like it to be. It'll come down over time. There may be a business unit or two that's inventory is too low. These things kind of balance each other out. We think at 2.6 months, we're in a very good position today.
In this business unit, number one, the inventory was very high, over 8.6 months. Clearly, there was a lot of excess inventory. When you build that much inventory also tends to age and gets out of mix. Where it was shipped such a high amount of inventory and customer demand shifts, a lot of the inventory becomes very slow-moving and takes time to correct.
I guess I'll give kudos to all the business units that were involved, the sales and distribution organization, our distributors, the planning folks that's with us. It was a very large team, everybody did a good job of executing.
There was also $110 million of product that we shifted back from distribution to direct. Customer by customer and distributor by distributor, whichever the pair was, the distributor shipping to customer, you had to go work that relationship and convince a customer and convince a distributor. Distributor is going to lose that business to direct customer, we'll ship it directly, make sure distributor cleans out their inventory so the new business and Microchip ships, and creating accounts for those customers, the receivables, learning where the product has to be shipped. Some places, sometimes it goes into hubs of distributors, sometimes there are multiple locations. All that challenging task by the shipping teams and marketing teams and sales teams, it's just purely a guess when we first talk to you regarding how long would it take.
I'd rather tell you two quarters and get it done in one quarter, rather than tell you one quarter and it takes two quarters.
Understood. Thanks, guys.
From Raymond James, we'll go to Chris Caso.
Yes, thank you. Thanks for taking the question. Just a question with regard to customer inventory levels, and I'd say this probably more pertains to the Microchip classic business. How much of the slowdown that you're seeing now do you attribute to some of the demand function issues that you talked about, as opposed to just customers holding too much inventory and typically, when we've seen these slowdowns in the past, it's some combination of the two. Do you think about it any differently this time around?
Well, when customer buys less product, you're unable to differentiate whether that is just purely an inventory correction or his demand is slower. It always tends to be a combination. If you're building 100,000 widgets a week, then you have certain amount of inventory at your loading dock, in your factory floor, some in raw material, some in work in process, you have a certain amount of inventory. If your demand drops because of your tariff concerns, let's say you only want to build 90,000 widgets, then the entire inventory in your structure divided by 90 is going to become more weeks of inventory than at 100. Then when he draws it down, you could call it inventory correction, you can call it because of lower demand. I don't know how you separate it.
Right. Understood. As a follow-up, if you could talk about the linearity of your order rate, as you proceeded through the quarter and into October, I assume that the order rates have slowed down since we last spoke in August. Have they sort of stepped down to a certain level and then stabilized that level, given the uncertainty about tariffs and such you spoke about? Did they continue to worsen on a month-on-month basis?
The order rate slowed down and stayed pretty constant at the low level and have recently bounced back. Partially because lead times are fairly short, they got to place the orders for December and January. They don't need to place orders for next June, they do need to place the orders for the next quarter. The order rate, as you said, slowed down, but has recently bounced back up.
In the month of October?
In the month of October, yes.
Okay. Helpful. Thank you.
We'll hear next from Vijay Rakesh with Mizuho.
Yeah. Hi, guys. Just wondering, on your de-leveraging, have you changed? Are you accelerating that? How are you looking at that?
I tried to answer that in an earlier question. Essentially, our de-leveraging is going to be at a slower pace than what we had originally guided to the street. We've got headwinds of very low EBITDA in both June and September due to distribution inventory reduction. We're having a down quarter this year. At the midpoint of guidance, we're going to be down 7.5%. That just has an impact and is going to stay with us now for 12 months because this debt calculations or EBITDA calculations are all 12-month trailing. The cash flow from the business is going to be high, but if you look at our accounts receivable balance, it's down $120 million quarter-on-quarter, and that translates to cash.
It's going to take us some time, but all of our cash, really all the significant cash in excess of what we need to run the business and pay dividends is going to be used to pay down debt. With that, we will reduce the debt balance, but the EBITDA is going to be something that sticks with us for the next year.
Got it. Just a follow-up here. Getting past some of the near-term tariff issues, the lack of visibility into the March quarter and first half, let's say. Any thoughts, one of the perennial concerns on microcontrollers has been, especially as you look at automotive, where does content go? Just wondering to get your thoughts on how you look at that. Thanks.
Ganesh, you want to take that?
Ganesh had to run to catch a flight.
Oh, he had to catch a flight.
Yeah.
I think we are not concerned about the long-term prospects of microcontroller, analog or any of our businesses. Our fundamental premise of being able to sell total system solution into a customer socket, whether it's automotive, whether it's industrial, with microcontroller, analog, connectivity products, Wi-Fi, Ethernet, Bluetooth, LoRa, and now with lots of Microsemi products, including discrete products, FETs and diodes, and RF and other products, I think there is fundamentally no concern about long-term Microchip and the design win funnel and all that are very healthy. I think we're just talking about this impact, some from the tariffs and some from the slowdown in the business. You could go back, there was an earlier question, how it is different or same as the 2015 cycle. In that 2015 cycle, some of the assumptions were like, this thing is over.
The industry just goes away and all these things don't get designed in. You can see how the business bounced back and we had record quarters and record years and outstanding market share gains and everything else. There's no long-term concern here, just short-term.
All right, thanks.
We'll hear next from Harlan Sur with J.P. Morgan.
Good afternoon, Thanks for taking my question. Maybe just touching on some of the pockets of strength in the business. In aerospace and defense, it's obviously very sticky, very high profitability segment. Sounds like near term it's kind of doing well. In order for you guys to sort of take advantage of this very profitable area, how much effort is required to get your MCU and analog products, mil-aero rad-hard qualified? Is there a lot of work to be done here, or is the core Microchip team already engaged in cross-selling?
There's a lot of work underway. There was a small aerospace and defense business that we have acquired as part of the Atmel acquisition, which was headquartered in France. They did business largely with French aerospace companies and didn't have any market share in U.S. After acquiring them in the last year and a half or so, we've been putting efforts. We expanded their charter to take Microchip standard microcontroller, analog, and memory products and make them rad-hard, recharacterize them to the needs of military and aerospace with broad temperature ranges and all that. All that work has been underway for the last year or so in militarizing our products. Microsemi brings even just a much higher level of expertise in that area and about 10x to 15x larger business than Atmel had in aerospace and defense.
In our integration, we have combined the Atmel aerospace and defense business together with the Microsemi aerospace and defense business as one business unit. That business unit is accelerating, taking Microchip products to market in that segment. That effort is about a year old and is really getting significant boost right now. We're not shipping core Microchip products in military for revenue today. Any Microchip products going into aerospace and defense sector are still standard commercial products that are bought from distributors and just kind of use standard commercial product. In some non-sensitive applications, aerospace and defense is allowed to use standard commercial products. The military products, rad-hard products, is an effort underway, but it's not making revenue today.
Got it. Thanks for the insights there. In the FPGA segment, you guys drove record revenues. Obviously, the Microsemi team built a strong mil-aero leadership franchise there, I think Ganesh talked about some of the new opportunities with the mid-range PolarFire family, which is more for kind of like multi-market applications, but sounds like overall the business is doing well. Can you guys just give us a sense on how strongly the FPGA business is growing on a year-over-year basis?
I don't have a year-over-year comparison. Somewhat it won't be valid because it will be a sell-in by putting a high channel inventory versus a sell-through that we are measuring. That's why we said, based on a sell-through basis, the numbers were a record. It was all-time record. I don't have a year-over-year comparison since we haven't had the business for a year. The business is doing very well. It's doing very well in defense and aerospace. It's doing very well in the multi-market, has very good design in funnel. The next generation architecture, which is Gen5, is just going into market and just introducing at the sample level. There's a significant demand pull on that, so I think we're very optimistic about that business.
Yep. Thanks for the insights, Steve.
Sure.
We'll go next to Jefferies, Mark Lipacis.
Hi, thanks for taking my question, just one. Steve, a number of the contract manufacturers with a global footprint have talked about some of their customers already shifting manufacturing outside of China and accelerated "activity" with their customers to do as much. I was wondering if you were noticing that, any evidence or anecdotes on that front, and to the extent that you think that might be causing some near-term disruption. That's all I had. Thank you.
You're very correct. We're seeing that too. Two things are happening. One, if there are two sources of production, one in China and one outside of China, then some attempt is being made to take the Chinese production and sell it to Europe and Asia, and take the non-Chinese production and ship it to U.S. This way you can optimize and not pay any duties. Two challenges in that. Number one, it's a lot of manual work, and most people aren't set up to really differentiate the product where it should go. Manually, and then writing some automation, it's really being done. It's not perfect, because in many cases, the U.S. demand is not exactly equal to the non-Chinese production, and Europe and Asia demand isn't equal to Chinese production, so there are still problems.
To solve that problem, you're correct that many customers are moving Chinese production to outside of China, but they are rapidly running into capacity issues. Where the outside factories are full, there was a little bit space, and they are now full, and there is a lot more to be transferred from China, but there's no place to put it till they really expand capacity, build more square footage and all that, and that's a slow process. Yes, we're seeing that.
Very helpful. Thank you.
At this time, I'd like to turn things back to Steve Sanghi for any closing remarks.
Well, we want to thank all of our customers and analysts for your patience and for your understanding. I think this has been a tough four months here, but I think we have done a lot of good work in cleaning up the Microsemi inventory. At the same time, it's really the first acquisition whose timing was such that the wind is not on our back. We did a number of other acquisitions and the timing was great. Here, along with the Microsemi inventory challenges, we're also facing this downturn in the business, we're dealing with dual challenges. I think we posted a pretty good quarter, and despite sequentially down guidance, I think we're managing the inventory very well. The internal inventory is in very good shape.
Overall gross margin, operating margins are still very, very good, and we look forward to really getting through these issues and getting back to our long-term $8 earnings target. Thank you very much.
That does conclude today's conference. Again, thank you all for joining us.