Broadcom Inc (AVGO) — Q4 2018 Earnings Call Transcript

Thank you, operator, and good afternoon everyone. Joining me today are Hock Tan, President and CEO, and Tom Krause, Chief Financial Officer of Broadcom. After the market closed today, Broadcom distributed a press release and financial payables describing our financial performance for the fourth quarter and fiscal year 2018. If you did not receive a copy, you may obtain the information from the Investors section of Broadcom’s website at broadcom.com. This conference call is being webcast live, and a recording will be available via telephone playback for one week. It will also be archived in the Investors section of our website at broadcom.com. During the prepared comments section of this call, Hock and Tom will be providing details of fiscal year 2018 results, guidance for fiscal year 2019, and some commentary regarding the business environment. We will take questions after the end of the prepared comments. Please refer to our press release today and our recent filings with the SEC for information on the specific risk factors that could cause our actual results to differ materially from the forward-looking statements made on this call. In addition to U.S. GAAP reporting, Broadcom reports certain financial measures on a non-GAAP basis. A reconciliation between GAAP and non-GAAP measures is included in the tables attached to today’s press release. Comments made during today’s call will primarily refer to our non-GAAP financial results. So with that, I will turn the call over to Hock.

Well, thank you, Bea, and thank you all for joining us today. As you saw, we closed the fiscal year on a very high note. Consolidated net revenue for the fourth quarter fiscal ’18 was $5.45 billion, a 12% increase from a year ago, and EPS came in at $5.85, a 27% increase from a year ago. Importantly, free cash flow was $2.53 billion or 46% of our net revenue. I would like to provide you more color on the top line today, and I have a lot to cover. Please note fourth quarter results do not include any contributions from CA. Starting with wired, on the quarter results, revenue was $2.2 billion, growing 3% year-on-year and the wired segment represented 41% of our total revenues for this quarter. Looking deeper though, fourth quarter wired results reflect very strong year-on-year growth for our networking and computing offload businesses driven by robust demand from the cloud data center markets as well as traditional enterprises. Networking and computing offload represented over two-thirds of our wired segment in the quarter and grew 22% year-on-year in the quarter. This is on the back of growing 10% year-on-year in the third quarter. So, this part of the wired segment continues to be very robust. On the other side, as anticipated, cyclical headwinds in certain parts of our broadband business reflecting weak carrier spending in those areas continued to impact this part of our wired business in the fourth quarter. As a result, broadband was down year-over-year again in the fourth quarter and offset partially the strong growth from data center spending. Turning to enterprise storage, revenue was $1.3 billion, representing 23% of revenue. Consistent with what we experienced in wired networking businesses, robust enterprise IT spending drove over 96% year-on-year revenue increase. Now, of course, this includes contributions from Brocade, which we acquired about a year ago. But even if we strip out Brocade, enterprise storage grew double-digits year-on-year in the quarter. Moving on to wireless, revenue was $1.7 billion, which was down 5% year-on-year. The wireless segment represented 31% of our total revenue. Wireless revenue, however, was somewhat better than our expectations for the fourth quarter as we benefited from upside volumes of legacy phone generations and our North American OEM customer. And finally, our last segment, industrial. In the fourth quarter, the industrial segment represented 5% of our total revenues. Distribution re-sales, which is how industrial outsold for us, continued to be strong contributing to high single-digit year-on-year growth in the industrial business. With that now, let’s talk about the segment performance for the full fiscal year 2018, which interestingly enough could be in stark contrast to the Q4 results I just articulated. Wired for us in fiscal 2018 was up 1% as networking expanded while broadband was down. Meanwhile, enterprise storage was significantly supported by Brocade as well as strong organic growth in our server storage connectivity business and industrial performed extremely well, up 12% helped by a healthy macro backdrop. Finally, despite all the quarterly fluctuations, wireless was actually up 20%. Our revenues hit a new record high growing 18% year-on-year to nearly $21 billion for fiscal ’18. This clearly demonstrates how our diverse set of businesses drives stability and sustainability in our consolidated revenue despite quarterly and even biannual volatility in specific segments. In addition, viewing our business broadly, you can see we have created over the years, organically and through acquisitions, a substantial core revenue stream in semiconductors based on technology enabling connectivity solutions across a broad set of end-markets. We continue to remain focused on the sustainability and growth of this core business. But in addition, with our acquisition of Brocade, we created a complementary revenue stream to our semiconductor solutions that we are now calling infrastructure software. With the acquisition of CA now, we will grow this revenue stream and build upon it through acquisitions consistent with our business model. As a result going forward, our two primary segments will be semiconductor solutions and infrastructure software. For fiscal 2019, the outlook for business is as follows. In the semiconductor solutions segment, we expect continued robust demand from cloud customers with the ramp of next generation Tomahawk 3 stretchers and from the launch of our next generation routers, Jericho 2. We also expect to see a recovery of spending by carriers and operators in cable as well as in communications as we expect the broadband market recovery to start to progress through the year. We have already seen that happen this quarter. Storage, we believe will be stable relative to fiscal 2018. And as we previewed last quarter, we believe the reset in our wireless business in the first half of 2019 from share loss in the current phone generation will be followed by a substantial recovery in the second half as we take share back for the next generation. While there will be lots of puts and takes here, our outlook for the semiconductor business is for modest revenue growth in 2019. This may be somewhat dampened relative to our long-term mid single-digit growth expectation by wireless. Now, turning to the infrastructure software segment, before providing our outlook, I should take a few minutes to outline the substantial changes we are making to the CA business model. We expect these changes to result in a dramatically more profitable revenue base, which is more aligned to the rest of Broadcom and that we expect will grow. First and foremost, gone are the days of trying to land new products with new customers. We are focusing all our attention on renewing existing products with existing mainframe-centric customers, customers that represent virtually all of the world’s largest enterprises and largest spenders on IT. We are also targeting expansion opportunities within this core mainframe customer base. The cost of running this renewed and expanded model will be substantially less than the legacy model of acquiring all costs and importantly renewing and expanding plays to CA’s strengths. Today, over 70% of CA’s revenues are derived from its top 500 accounts. In almost all cases, these top customers have been licensing CA mainframe products for more than a decade and oftentimes several decades. At this same customer, enterprise products are sold, but sold as perpetual licenses with maintenance and data in the license agreements. At each of our top customers, we have two primary objectives: 1) we want to expand our efforts on mainframe and make sure that we are realizing the full value that our mainframe tools are delivering to our customers. And as we discussed on the prior call, usage has been growing at double-digit rates at all these top accounts. We are focused, because of that, on pricing mainframe based on consumption. We also feel there is a huge opportunity for customers to save money by leveraging our broad mainframe portfolio to drive more conversions to CA tools. The second objective is to expand our enterprise software products within the same top accounts. It is true that lower costs and lighter weight SaaS alternatives have been creating challenges for CA for some time in the enterprise software market. What’s interesting is that CA actually has very highly rated enterprise software, in fact, Gartner upper right-hand Magic Quadrant categories for enterprise software on behalf of enterprises. While we are very well suited to the private cloud IT environment of the largest enterprises, these enterprise software are just too expensive relative to SaaS. We are growing away to move away from the inflexible, perpetual license model for enterprise software to an enterprise-wide all-you-can-eat license for all of our core accounts. By doing this, we expect to remove the friction caused by selling expensive upfront perpetual licenses so that the incremental costs for our customers to expand the use of enterprise products will be highly competitive relative to SaaS-based alternatives. Bottom line, we are adopting a fully ratable subscription model for the Broadcom software business. This new business model we believe plays to our strengths, focusing on the largest 500 customers tied to mainframes with the ability to up-sell enterprise software competitively using an all-you-can-eat subscription-based model. We expect this transition, however, to take a couple of years given the timing of contract renewals. But once completed, we expect revenues to stabilize at over $3.5 billion annually and grow from there. To support that revenue base, we do not expect to spend more than $900 million per year. As a result, we expect to achieve more than $2.5 billion per year in operating profitability from the CA business once we go through this transition. We are well underway today with the CA restructuring and integration process, including an announcement of the Veracode divestiture to Thoma Bravo and the outsourcing of the CA services business to HCL. Now, let me talk about our outlook for the infrastructure software segment in 2019. SAN switching and private channel SAN switching performed beyond our expectations in 2018 on the back of very strong enterprise demand as well as meaningful share gains. While we expect to continue to see healthy demand, we do not expect in this forecast to have this sustained through 2019. Furthermore, since we are moving mainframe and enterprise software products to a fully ratable revenue recognition model and just focusing on the top 500 accounts, we expect a reset in the CA revenue starting Q1 2019. Our revenue forecast for the infrastructure software segment for 2019 will be approximately $5 billion. Combining with semiconductor solutions, in summary, we are forecasting consolidated revenue to be approximately $24.5 billion fiscal 2019.

Thanks, Hock, and good afternoon everyone. My comments today will focus primarily on our non-GAAP results from continuing operations unless otherwise specifically noted. Let me walk through our results for the fourth quarter of fiscal 2018. Fourth quarter net revenue was $5.45 billion, ahead of the midpoint of guidance. Our gross margin from continuing operations was above the high end of our guidance at 68.4% as we benefited from a more favorable product mix in the quarter. Operating expenses were slightly lower than expected at $863 million. As a result, we achieved record profitability in the quarter. Operating income from continuing operations was $2.86 billion and represented 52.5% of net revenue. Adjusted EBITDA was $3.02 billion and represented 55.4% of net revenue. This figure excludes $132 million of depreciation, and the company delivered $5.85 of EPS in the quarter off of a $435 million weighted average fully diluted share count. This represents 27% EPS growth compared to the same quarter last year. Working capital, excluding cash and cash equivalents, increased approximately $105 million compared to the prior quarter due primarily to an increase in receivables driven by seasonally higher shipments in the last month of the quarter. In addition, we spent $106 million on capital expenditures. As a result, we had record free cash flow from operations at $2.53 billion or 46% of revenue. This represents 47% growth in free cash flow from Q4 of 2017. In the quarter, we returned $2.26 billion to stockholders, including $723 million in the form of cash dividends and $1.53 billion for the repurchase of 6.4 million AVGO shares. We ended the quarter with $4.3 billion of cash, $17.5 billion of total debt, $408 million of outstanding shares, and $432 million of fully diluted shares outstanding. Now, let me turn to our fiscal year 2019 non-GAAP guidance. We do intend to update our annual guidance on our quarterly earnings calls throughout the year. This guidance is for results from continuing operations only. As Hock discussed, net revenue for fiscal 2019 is expected to be approximately $24.5 billion, including approximately $19.5 billion from semiconductor solutions and approximately $5 billion from infrastructure software. IP licensing is not expected to generate a material amount of revenue. Operating margins are expected to be approximately 51%. I would like to note that post-CA integration restructuring, we do expect to move closer to 55% operating margins in 2020. Net interest expense is expected to be approximately $1.25 billion and reflects maintaining a target cash balance of approximately $4 billion and servicing total debt outstanding of approximately $37 billion following the close of the CA deal. This forecast does not contemplate any debt pay-down in fiscal year 2019. The tax rate is forecasted to be approximately 11% and includes a slight negative impact from the CA acquisition. Depreciation is expected to be approximately $600 million. CapEx is expected to be approximately $550 million. As a result, free cash flow from continuing operations is expected to be approximately $10 billion. Stock-based compensation expense is expected to be approximately $2.1 billion. Now, this is a substantial increase in our stock-based compensation expense. We are implementing a special broad-based multiyear equity award program for our employees, including our new CA employees. Each multiyear equity award will vest on the same basis as four annual equity grants made on March 15 of each year beginning in 2019. A maximum of approximately 31 million shares of common stock in aggregate will be issued which will vest over the next seven years. This is the same number of shares as we would have expected to grant over the next four years annually. The spike in the 2019 stock-based compensation will start to come down in 2020 and decline from there back to our normal level by 2022. This accounting dynamic impacts the stock-based compensation in 2019 but providing four years of equity grants upfront provides clarity regarding future compensation that creates a powerful retention incentive in an otherwise tight labor market and sharpens the focus on long-term stockholder value creation. In addition, it allows us to maximize the use of the remaining authorized share reserves under our 2019 Avago equity award plan, which unfortunately is expiring in 2019. Broad-based employee stock ownership is a fundamental tenet of our company, and it is important that we continue this legacy while our current equity plans enable us to do so. Hock is not participating in this program and as previously disclosed, will not receive another equity grant until at least 2021. For executives, 50% of the awards are PSUs, the vesting of which is tied to total shareholder return, similar to our prior annual awards to executives. Finally, no further annual grants are planned for employees who receive this award until at least 2022. Now, let me turn to capital allocation plans before we open the call for questions. Consistent with our capital allocation strategy, we are focused on returning approximately 50% of our prior year free cash flow to stockholders in the form of cash dividends, with the balance being allocated to a combination of stock buybacks and acquisitions. In addition, we plan to continue using our balance sheet to fund acquisitions while focusing on maintaining our investment-grade credit rating. Based on approximately $8.2 billion of free cash flow that we generated in fiscal 2018, we are increasing our target quarterly cash dividends starting this quarter to $2.65. This constitutes an increase of 51%. We plan to maintain this dividend payout throughout the year subject to quarterly board approval, which means we plan to pay out over $4 billion in cash dividends in fiscal 2019. We will reassess the dividend at this time next year based on our fiscal 2019 free cash flow from operations. Given the dilution stockholders are bearing from the multiyear grant and given the free cash flow yield that Broadcom is currently generating, we are also budgeting to return an additional $8 billion to stockholders through stock buybacks in fiscal 2019. Coupled with the dividend, this means we are planning to return approximately $12 billion to stockholders in fiscal 2019, which constitutes all of our projected free cash flow, plus the excess cash that we have on our balance sheet today.

Thanks for taking my question, and congratulations on the good execution. Hock, I understand and appreciate keeping the focus on longer-term trends, but just because removing the guidance on a quarterly basis is a big change. Just for this quarter, could you give us some color on how Q1 trends are shaping up especially given all the concerns around trade and tariff and your largest customers? Even if you can't quantify everything, if you could just give us some color commentary on what’s going on in different segments in Q1, that would be very helpful.

I’ll give you the answer, it’s okay. Remember, we have backlog out 18 weeks for most of our products, that’s longer than a quarter, which runs 13 weeks. And based on what we have in place, it’s trending pretty well compared to Q4. It’s there are puts and takes even in all of this. Broadband starts to recover, as I mentioned before, finally, long last, and networking offload computing is still nicely holding up, but handset wireless, you’ve seen it out there, we expect to see a seasonal downtake. So, storage flattish back to moderation. All combined together, things are kind of what it is.

Yes, thanks for taking the question. I want to understand maybe the puts and takes a little bit better in the software – infrastructure software guide. If we look at CA’s results on a standalone basis, it looks like they’re about $3 billion to $5 billion. You’re stripping out the services business. You’ve sold Veracode. So, can you help us bridge a little bit more the uplift you’re seeing from that level of revenue to that $5 billion guide for the full year? Thank you.

Hey, Aaron, it’s Tom. Keep in mind there are now two substantial businesses. As you properly pointed out, there’s the CA business for a couple of data points. Veracode run rate business is about $150 million a year and it was growing, and we’ve outsourced the services business, so that business will start to tail off through the course of 2019 and largely be gone in 2020. But then keep in mind also, there’s Brocade, the SAN fiber channel switching business, which is performing very well for us. We’re not breaking out the specific revenues for that particular business, but it’s also providing a substantial portion of the overall $5 billion. So in total, we see a reset in the CA business starting in Q1. We do expect based on the renewal expectations around our core 500 customer base to grow throughout the year with CA, and we also expect to continue to maintain reasonably high levels of revenue with the Brocade fiber channel business.

Thanks for taking my question, guys. When I think about that $2.5 billion operating profit target from CA, can you just talk about that timeline to achieve that? And when I look at the accretion or the incremental contribution you get from post-CA, how much of that is going to come in cost versus OpEx for you guys?

Well, very interesting question. Let me outline again what I went through in my remarks fairly quickly. We start 2019 partly because of a resetting from recognizing perpetual licenses on a salaried manner to ratable subscription-based revenue recognition. ’19 will take a step down from what you typically expect the rate to be, and it will rapidly build up over the next 2-3 years to the level I spoke about closer to over $3.5 billion. On the spending side, if you recall, before we acquired CA the last quarter, stripping out services, taking out services which was a wash, total spending was about $2.4 billion to $2.5 billion per year. We’re bringing it down to $900 million. We’re able to bring it down because a large part of that $2.4 billion of spending was attributed to the various sales motion, development motion of trying to land new customers. The largest 500 customers in the world are already our customers through mainframes. We’re saying that going forward, we’ll get to the same revenue number with much less spending. That $3.5 billion, say conservatively we get to in year 2 or year 3 from today, less the $900 million end-state spending is where we believe we get to around the $2.5 billion operating profit target.

Great. Thank you so much for taking the question. Hock, you talked about your intention to regain share in the RF business next year. I appreciate the time you spend with your customers in designing these products, but I was under the impression that the SKUs for next year hadn’t been set. What gives you the confidence that you can indeed regain share in that business?

We are just confident. We have not been idle; we have been working. These are very difficult products, complex and technologically advanced products to develop, and we have been working on this for over a year with customers.

Good afternoon. Thanks for taking my question. Hock, you talked about continued strong trends fiscal ’19 in networking demand, cloud, and enterprise. I was hoping you could quantify a bit more. Next year, it’s still looking like the cloud guys are growing their spending again albeit at a lower rate versus this year, but then you are there on the 200 gig, 400 gig upgrade cycle with Tomahawk 3 and you have got the ramp of some of your AI and deep learning and SmartNIC ASIC programs. Given all of this, do you still feel the team can sustain double-digit year-over-year growth rates for this segment fiscal ’19?

Very good question. Yes, cloud - public cloud has been part of our networking and compute offload business. The public cloud side, which is at least half of our revenues in that sector, continues to be extremely strong. In 2018, we didn't launch any major new milestone products and we still grew double digits. In 2019, along with that natural momentum, we have two significant product launches. The top of the rack switch, the Tomahawk 3, which is a 12.8 terabit per second throughput switch that will be used by hyper cloud providers. This will be a big driver of growth, in addition to the Jericho 2 router coming mid-year. This puts us in a favorable position as the demand continues to increase for cloud services. Overall, our growth looks promising.

Thank you. Tom, I just wanted to confirm the facts on the stock option grant, so $2.1 billion for fiscal ‘19 and you have that coming down over – is it over a four-year period and does it return to the fiscal ’18 levels or some level above that?

Yes, that’s the right way to think about it, Romit. It’s a four-year grant accelerated and done in one shot this year as opposed to doing it over four years. So in aggregate, you wouldn’t have any difference, but from an accounting perspective, you'll have to take all the step up this quarter. It will start to come down next year and decelerate back to where we were over a four-year period. The 2018 level would be roughly $300 million a quarter type stock-based comp run rate as the company’s normal steady state basis.

Great, thanks, congratulations on the quarter. I’m particularly focused on the dividend; there was a significant increase this quarter. When we contemplate the company’s ability to grow the top line long-term, expand margins, and your capital allocation plans, what does management expect the long-term growth rate of that dividend to be?

Sure, Will. So I think we have spelled it out fairly clearly on both the policy in terms of returning 50% of free cash flow from the prior fiscal year. When you take into account the buyback expectations of approximately $8 billion and the outstanding shares should come down, alongside free cash flow increasing, we expect a good setup to continue to drive the dividend well into double digits over the next several years.

Hi guys, thanks for taking my question. Can you elaborate on the all-you-can-eat model that you are developing for the enterprise software business? Does that basically mean one license that a customer takes for anything that you buy going forward? If that’s true, how do you grow the business without taking those rates up over time if you are still selling to the same customers? What does that model actually look like?

Yes, that’s a very good point. You are right. We provide that enterprise-wide license to those core customers only, product by product. It’s not across all our enterprise products at the same time, only when the customer is adopting it. If a customer is a big core customer who adopts our agile operations or rally for projects, we will provide a multi-year contract under that enterprise-wide unlimited license. After the contract is three years, we expect inflationary improvements and if we innovate and improve the product, we can sell them additional products, and we expect that’s how we grow.

Thanks for taking the question. I think it was in your comments where you mentioned the aspiration for 55% operating margins in fiscal ‘20, but since that would represent a 400 basis point increase from what you are targeting in fiscal ‘19, can you walk us through some of the assumptions that could lift the operating margin to that level? Thanks.

Yes, good question. There’s a number of things with that. First and foremost, we see growth in the business, especially in the core semiconductor business, where we expect to see a return to mid-single-digit growth rates in 2020. The software side will also continue to grow into the ratable model. We are also focused on gross margin expansion, especially on the semiconductor side. Finally, we will be reducing expenses dramatically with CA as we change the business model and focus on top 500 accounts. This leads us towards achieving 55% operating margins.

To be specific, this fiscal ‘19, when we buy a company as complex and large as CA, it takes us a year or two to transition through the end state. I estimate we are carrying about $1 billion of transition expenses in fiscal ‘19. It won’t all evaporate by fiscal ‘20, but a big part of it will evaporate, which along with revenue increase gets us to that 55% operating margin.

Yes, thanks. Hock, any particular feedback from large customers now that you have Brocade and CA together? Anything you would like to discuss in terms of synergies and overlap of customer base?

Great question. Yes, I have met with several CIOs of some of the largest customers of CA who happen to also be the largest end-use customers of Brocade. A lot of these CIOs are thinking through the high levels of IT spending. They’re figuring out the best structure for their data centers and many are considering building their own private cloud. We have all the technologies, hardware, and software to enable them to build private clouds. These CIOs are able to spend several billion dollars a year in IT, so there are potential synergies here as we engage in dialogue with them.

Hi guys. Hock, you mentioned the all-you-can-eat model for software. I was wondering if you continue to do more M&A on the software side that you can stack on that same model?

That’s a great idea, and we definitely want to do that, because we developed with CA the platform for support and ensuring customer success, as well as directly engaging with those largest 500 customers. As we add more products, especially enterprise software products, we expect to build on that revenue stream in infrastructure software.

Hi, guys. Congrats especially on the cash return side. Hock, I wanted to ask a bigger question; with all the uncertainty in China trade, etcetera, have you noticed any change in any of the end markets given these uncertainties on customer behavior in anyway?

Yes, I am sure they are. But I think it’s hard to tell if some of it is related to macroeconomic variations versus the bigger concern with respect to tariffs. We see some segments down, but we also see segments up. Overall, it’s not completely clear yet how this will affect the business we are in, which largely focuses on enterprises and operators. Our exposure to consumer is limited to high-end phones, and the phone market has not been strong in recent months.

Hock, you made some comments about cloud/hyperscale, that has been an area where growth has been particularly strong this year. Can you discuss if that growth is sustainable into ‘19, and differentiate between your core Ethernet business and maybe some of your new emerging ASIC business around AI acceleration?

The cloud guys are still spending. Their spending pattern starts to track that of operators and gets lumpy, but if you look across a year, they sustain high spending levels. It includes both U.S. and China providers. We are selling them more products than just switching, including interconnects like fiber optics, which increased in content as they upgrade. We are also benefiting from sales of deep learning chips and related products. We believe that trend will continue as data centers expand their services.

Hi, thanks. Tom, I’m trying to get a bridge on the $24.5 billion relative to the $23.9 billion shown as pro forma in the CA presentation. I know you’re losing Veracode and some around HCL, but you’re also getting a bump from the change in the model. Can you provide an apples-to-apples bridge?

It’s a challenging bridge only because you’re looking at two accounting standards with 605 versus 606 on the CA side. The $24.5 billion figure is indicative of modest growth in semiconductors, plus the two businesses of CA and Brocade contributing to the $5 billion. I feel confident in this forecast considering the expected growth driven by these solid businesses.