Inphi Corp
Q3 2013 Earnings Call Transcript

Published: 30 Oct 2013

[Operator:]

Good day, ladies and gentlemen. Welcome to the Inphi Corporation Q3 2013 Earnings Call. My name is Phillip and I will be your operator for today. (Operator instructions.) As a reminder, this conference is being recorded for replay purposes. I would now like to turn the presentation over to Mr. John Edmunds, Chief Financial Officer. Please proceed, sir.

[John Edmunds:]

Good afternoon, everyone. We’d like to thank you for joining us today for our quarterly earnings call to discuss our Q3 financial results and business outlook. I’m John Edmunds, Chief Financial Officer for Inphi Corporation, and with me today is Ford Tamer, our Chief Executive Officer. First I will read the Safe Harbor then Ford will give you an overview of our business. After that I will give you a short financial summary and the outlook for Q4 2013, and then we’d be happy to take your questions. Before we begin today please note that during the course of this conference call we may make projections or other forward-looking statements. These forward-looking statements and all other statements made on this call which are not historical facts are subject to a number of risks and uncertainties that may cause actual results to differ materially. These forward-looking statements speak only as of today’s call. We do not undertake any obligation to provide updates after this conference call. For further information regarding risk factors for our business please refer to our registration statements on Form S(1) as well as our more recent annual and quarterly reports on Forms 10(k) and 10(q) all filed with the Securities and Exchange Commission, accessible at www.sec.gov. Please refer in particular to the sections entitled “Risk Factors.” We encourage you to read these documents. Also during the course of this conference call we may make reference to non-GAAP financial information. A reconciliation of this information is included in the press release and on our website which is available at www.inphi.com. This information is not a substitute for GAAP and should only be used to evaluate the company’s results in conjunction with corresponding GAAP measures. Now, to begin our review of the quarter I will hand you over to our Chief Executive Officer Ford Tamer. Ford?

[Ford Tamer:]

Good afternoon. Thank you for joining us for our Q3 2013 earnings call. We’re delighted to report that we’ve just completed the highest revenue quarter in Inphi’s history at $26.6 million and are now operating at a great than $100 million revenue run rate. We exceeded our revenue forecast by delivering a 9.5% sequential increase, and even on this higher number we’re guiding to $28.8 million for the Q4 revenue – an 8.0% sequential increase and a 26.0% year-on-year increase. Importantly, we delivered this all-time high result while keeping operating expenses controlled. We booked EPS of $0.04 per share per our forecast. Before John goes into details on our P&L let me take this time to discuss the major trends impacting our business and our Q3 accomplishments. Recently I’ve been describing Inphi as the FedEx for high-speed data movement. Our Layer 1 Interconnect semiconductor solutions move data at high speeds from a few millimeters up to 3000 kilometers with low latency, robust margins and low power. If data absolutely, positively has to be there we guarantee delivery with logging, notification, retries and confirmed receipt. We do this at high speeds, up to 400 Gbps today with a roadmap to 1 terabit per second in the near term. Our Tier 1 OEM customers deploy mission-critical applications where data delivery is expected, required, and taken for granted. We move data within and between datacenters, between datacenters and users, and in long haul and metropolitan networks. We have three primary product families

[John Edmunds:]

Thanks, Ford. Now I will briefly recap the key financial results for Q3. As Ford stated, for Q3 2013 we reported record revenues of $26.6 million which represented 9.5% sequential growth and was above the midpoint of our guidance of $26.3 million. Overall both our communication and server revenues are growing by double digits in the second half of 2013 versus the first half. We experienced more of a surge in server-related growth in Q3 and we expect more of a surge in communications-related growth in Q4. Overall the mix of product sales for the second half of 2013 is continuing to approximate something on the order of 60% server and 40% communications. While the backlog for Q4 LRDIMM sales have been growing of late our stronger Q3 mix of server-related revenue comes primarily from RDIMM as the server memory market has been growing along with the introduction of the Intel Ivy Bridge platforms. In communications for the second half of 2013 we are continuing to see more 100G product revenues, both from amplifiers, drivers, and SerDes as well as other product sales. On a GAAP basis, net income in Q3 was a loss of $2.8 million or a loss of $0.09 per diluted share. This included stock compensation expense of $4.7 million and the associated offsetting tax benefit of approximately $1.8 million. This compares to a net loss of $1.1 million or $0.04 per diluted share in the same quarter one year ago. To give you more detail on comparing numbers I will now provide some additional non-GAAP measures. Within the press release you will find a specific written reconciliation as provided, but in general the differences between GAAP and non-GAAP are due to excluding stock compensation expense – a net $2.9 million as noted earlier, as well as some other GAAP to non-GAAP differences in income tax expense of approximately $1.0 million. On a non-GAAP basis then, net income for Q3 was approximately $1.1 million or approximately $0.04 per diluted share which was at the high end of our Q3 earnings guidance. This also compares with non-GAAP net income of $1.5 million or $0.05 per diluted share for the same quarter one year ago. Non-GAAP gross margin for Q3 2013 was 64.2% which was down 40 basis points from the 64.6% in Q2 2013. In Q3 the gross margins came down based solely on the higher mix of server-related business. Our gross margins are expected to improve sequentially by approximately 30 to 50 basis points higher in Q4 based on a higher expected mix of communication product sales in Q4. Non-GAAP operating expense for the quarter totaled $15.5 million. This was up approximately $0.1 million over Q2 and less than our guidance for Q3. We plan to continue to hire critical engineering resources in Q4 as we continue to make measured investments to support projects with large OEM customers. We expect our Q4 operating spending to rise approximately $0.9 million +/- $0.2 million, principally due to additional heads as well as design and qualification activities in R&D. With regard to the non-GAAP pro forma tax provision, in Q3 the effective tax rate remained the same at 38%. As volumes continue to ramp in 2014 we expect to be able to bring the effective tax rate down; however, this is currently our best estimate for the balance of 2013. Other income in Q3 was approximately $0.2 million coming mainly from interest income consistent with Q2. Now turning to the balance sheet, cash and investments in marketable securities increased by $0.4 million for the quarter, from $118.9 million to $119.3 million – so basically about $3.80 per diluted share at September 30th. This net increase of $0.4 million in cash and investments was a result of approximately $4.0 million in cash from operations plus $1.8 million in cash from stock option ESPP exercises all offset by $2.9 million being used for property, plant, and equipment for mask sets, lab equipment and tenant improvements; and $2.6 million being invested in a small equity position in a technology-based startup. The $4 million in non-GAAP cash flow from operations comes from $1.1 million in non-GAAP net income, adding back $1.9 million in depreciation and amortization and $1.4 million in other non-cash based income tax charges as well as netting the use of $0.5 million in cash for net working capital expansion. Our cash flow and working capital in general has continued to stay efficient in Q3. Accounts receivable increased $1.5 million from $11.3 million to $12.8 million due to the linearity of shipments in the quarter. This represented 44 days sales outstanding, up slightly from the 42 days reported at the end of June. With regard to inventory we were down from 59 days on hand at the end of June to 52 days on hand at the end of September, which replied an inventory turns measure improving from 6.1 to approximately 6.9. Total inventory decreased by $0.1 million from $5.6 million to a lean $5.5 million at the end of September. Payables increased from $5.4 million to $6.0 million. Days payables outstanding remained flat at 57 days at the end of both June and September. Now for the business outlook for Q4. Again, I would like to remind everyone that the following statements are based on current expectations as of today and include forward-looking statements. Actual results may differ materially. We do not plan to update nor do we take on any obligation to update this outlook in the future. As a range of guidance for Q4 revenues are forecasted to be up sequentially 5% to 11% or $28.8 million at the midpoint +/- $0.8 million. Non-GAAP gross margins are expected to improve based on mix of business and be in the range of 64% to 65%. We expect non-GAAP operating expense to increase $0.9 million in Q4 +/- $0.2 million primarily in R&D due to ongoing hiring and new product design and qualification expenses. We are currently estimating the non-GAAP effective tax rate to be 38% for the year. We are confident these components should then align resulting in non-GAAP net income between $1.3 million and $1.8 million which on approximately 32 million estimated diluted shares would result in an estimated non-GAAP earnings per diluted share of between $0.04 and $0.06. We also estimate non-cash stock compensation expense to be $4.5 million to $4.7 million. This would imply a GAAP net loss of $900,000 to $1.5 million for the quarter or a GAAP earnings per share loss of $0.03 to $0.05 per common share for Q4. The outlook will not be updated during the quarter and up until the time of the next quarterly earnings release unless Inphi publishes a notice stating otherwise, so please ask your questions today during the general Q&A period. And now we’ll be happy to answer any questions you might have.

[Operator:]

(Operator instructions.) And your first question comes from the line of Quinn Bolton from Needham & Company. Please proceed.

[Quinn Bolton:]

Hi Ford, hi John. Thanks for letting me ask these questions. I wanted to start just on the server side of the business since you saw a surge there. It sounds like it came mostly from RDIMMs. I might have missed it but did you say whether or not the LRDIMM business also increased?

[John Edmunds:]

Most of the surge as we suggested in the prepared remarks, Quinn, was coming from RDIMMs. And we have had an increase in the LRDIMM and particularly in the bookings for LRDIMM here in Q4. So we do see somewhat of an increase coming in that area.

[Quinn Bolton:]

Okay, but it’s mostly RDIMM around Ivy Bridge launch that drove that business in Q3.

[John Edmunds:]

That’s right.

[Quinn Bolton:]

Great. My second question, Ford, you mentioned sort of the CPAK silicon optics module for datacenter applications in your prepared script. I’m wondering if you have any content actually in CPAK, maybe CDRs, or is your opportunity as CPAK is deployed really getting your SerDes on line cards and really driving up port density?

[Ford Tamer:]

We cannot confirm or deny that we are or aren’t in the CPAK. The CPAK is shipping so at some point you should be able to open it up and see what’s in there. We do feel that the introduction of technologies like the CPAK will be very positive for Inphi because you will need next to the optics a set of CMOS driver and TIAs which we’re very well positioned to deliver on. And so if you think of this transition to this new silicon photonics world we do believe our opportunity to serve that market is we’re very well positioned to deliver in a very unique way.

[Quinn Bolton:]

So it sounds like the CMOS drivers and TIAs actually could go into CPAK-like modules in addition to opportunities for SerDes to drive those modules?

[Ford Tamer:]

That’s correct. So all these modules would have driver TIAs, SerDes or CDR type of offerings in them, same as the older technologies based on indium phosphide. So we support both the older technologies as well as some of the newer silicon photonics technologies.

[Quinn Bolton:]

Great. And then just the last question from me, just wondering if you guys can comment whether you feel that the CDR and SerDes business is sort of on track to your revenue targets for 2013?

[Ford Tamer:]

Yes, it is on track and you should see the introduction of a SerDes on the line card next year as a new revenue driver.

[Quinn Bolton:]

Great. Thank you very much.

[Operator:]

Your next question comes from the line of Doug Freedman from RBC Capital Markets.

[Doug Freedman:]

Hi guys, thanks for taking my questions. If I could start off with, could you give us a lay of the land on what you’re seeing in shifting competitive landscape in the LRDIMM and RDIMM marketplace, and what you’re seeing as far as ASP trends? I know that one of your competitors recently came public – I’m wondering if that’s having an effect on the way that they are handling the market and any thoughts there.

[Ford Tamer:]

There’s always a slight shift when you get new competitors coming in but it’s been relatively stable compared to what we’ve seen so far. We are the only vendor on DDR3 who is qualified across both RDIMM and LRDIMM. So today if you go to the Intel website where they post the memory validation and the different modules that are validated you’ll see Inphi across the board as opposed to some of the competitors we have who play only RDIMM or only LRDIMM. We also do feel that the knowledge and understanding of the buffer technology we’ve gained through DDR3 has enabled us to deliver a better product for DDR4. So we are pretty confident that our DDR4 register software has been the lowest power and best performance. The buffer is actually a very complex part and we’ve actually delivered on a whole range of functionalities specified by JEDEC and we do feel like we will have one of the best buffer technologies out there as well. So we feel pretty good about DDR4 as well.

[Doug Freedman:]

Great. There’s no doubt you guys are executing beautifully. So on that, you’ve laid out a plan where you felt like you were going to be able to get to a $200 million run rate. Are you still comfortable with that goalpost and any change to that that you’d like to share with us?

[Ford Tamer:]

We haven’t changed that goalpost and obviously it’s a target at the end of 2015 and directionally we’re still targeting that goalpost. Could it shift by a quarter? Possibly but we haven’t changed the goalpost yet.

[Doug Freedman:]

And are there any inflection points that you think are necessary to achieve those goals?

[Ford Tamer:]

So right now John likes to refer to this as a staircase and so we’re putting a foot in front of the other every quarter. And we will be, as we noted on the last conference call, steadily increasing and delivering to what we tell you we’re going to do every quarter. We do believe some of the new products we’re working on will result in potentially nice increases in 2016.

[Doug Freedman:]

And my last one and then I’ll jump back in the queue. Do you happen to have your appending plan figured out for next year that we can sort of get an idea of what type of investment you’re making in support of these goals?

[John Edmunds:]

We actually haven’t quite finished our planning process at a granular level yet for next year but I think the street models out there are fairly indicative of what we’re thinking for the most part, Doug.

[Doug Freedman:]

Terrific. Thanks guys, and congratulations on great results. I’ve got to say they’re stand out in this environment that we’re seeing with your peer group.

[John Edmunds:]

Thanks, Doug.

[Operator:]

And your next question comes from the line of Jorge Rivas from Craig-Hallum Capital Group. Please proceed.

[Jorge Rivas:]

Hello guys, thanks for taking my questions. I have a question on your expectations for metro applications on 100G. I heard from some peers that are saying that this market could be four times the size of long haul so I’m just wondering what your expectations are for next year, how to introduce products [to seize] opportunity.

[Ford Tamer:]

We think the metro 400G will start seeing shipment in 2014 but the big year will be 2015. So you’ll see systems starting to ship next year but the big revenue will probably be a year after that.

[Jorge Rivas:]

Okay. And then one last question from me, can you give us the split between 40G and 100G in the communications segment?

[John Edmunds:]

You know, Jorge, we don’t typically break that out on a consistent basis but I think I’ve said in the past that 100G is 75% or more of really what we’re shipping at this stage. So we don’t have much shipping in the 10G or 40G markets. We’re most focused on the 100G.

[Jorge Rivas:]

Okay, alright. That’s all from me.

[Ford Tamer:]

Jorge, I would say that this is helping us in this market. So the fact that we have predominantly 100G revenue is supporting a revenue increase given the 40G and 10G are seeing pressure and not growing as fast.

[Jorge Rivas:]

Right, okay. That’s all from me, guys, thank you.

[Operator:]

Your next question comes from the line of Tore Svanberg from Stifel. Please proceed.

[Evan Wang:]

Yes, hi, this is Evan Wang standing in for Tore. Congratulations on the quarter and a great guide. I think in your prepared remarks you hinted at the linearity becoming stronger through the quarter. I was wondering if you could talk about how linearity has been so far in Q4 and how that has factored into your guidance?

[John Edmunds:]

It’s a good question, Evan. In general I think we see the business as strong going into Q4. Our backlog is in the typical position that we see it at this time in the quarter, maybe a little stronger but that’s not unusual coming into Q4. So we just continue to forge ahead and ship product based on the flow of the orders.

[Evan Wang:]

That’s great. And what kind of expectations do you have for the month of December, that being the last month? And is there any seasonality that you factor in?

[John Edmunds:]

Typically you’ll see in Q4 stronger shipping in the first couple of months and it really just depends on particularly the market environment for the product in that quarter. So you can see it continue to extend and ship into the third month in strong levels; you can also see people take advantage of the opportunity to do maintenance and that sort of thing. So it always sort of depends on the strength of the quarter as we move through but it’s typically stronger in the front half than it would be in the back half. And then as you go into Q1 next year there can be some seasonality depending on capital markets and capital spend and whether people want to be conservative coming out of the gate or not. So we just have to wait and see.

[Evan Wang:]

Okay, great. My second question has to do with the memory price. It seems like the price has been increasing lately just from all (inaudible) and the capacity constraints and so forth. How has that helped your business or how has it impacted your business in any visible way?

[Ford Tamer:]

So we go on regular price negotiation every six months, so there’s one coming up. And we’re happy to see that our customers are doing well so overall that typically bodes well for all the participants in the market. So it hasn’t impacted our pricing yet but we’re always happy. We measure our success by the success of our customers, so when our customers do well we’re happy to see that and that should hopefully mean good things for all the participants in that market.

[Evan Wang:]

Great, thank you very much. Thank you for taking my questions.

[Operator:]

Your next question comes from the line of Sundeep Bajikar from Jefferies. Please proceed.

[Sundeep Bajikar:]

Hi guys, nice job on the quarter. Has your visibility into the server market improved? And as part of that, in terms of new products beyond LRDIMM do you have any exposure to NVDIMM or other similar solutions being rolled out as the use of Flash increases in datacenters?

[Ford Tamer:]

So on the NVDIMM we have not announced anything but we have solutions being tested in that market. So it’s an early market and it’s one that will probably go through a standardization (inaudible) before it becomes a bigger market, so we’re definitely very involved in that market and very up to speed on what’s going on. So stay tuned, Sundeep, we haven’t announced anything formally in that market. On the visibility?

[John Edmunds:]

Yeah, I think in regards to your visibility question, Sundeep, visibility has probably improved marginally recently just because of the advent of Ivy Bridge and the interest in people upgrading and doing things with servers at this stage. So there has been some additional activity and visibility in the near term but it’s not a margin – it’s not a tectonic change in the way that world works.

[Sundeep Bajikar:]

Okay great, very helpful. And then on DDR4, just wanted to know if there is a change in timing. When do you think you would start to see sort of a volume ramp of DDR4 revenues?

[Ford Tamer:]

So Sundeep, our job is to get ready as soon as possible and be ready ahead of our partners, so we’re not going to predict timing. You should probably ask our customers and our partners about that. Our job is to make sure that we’re not going to delay anyone and be ready with very robust high-quality, low-power, high-performance, great margin offerings that support stuff like strong ESD and strong EOS, extended voltage tests, tremendous characterization, phenomenal system validation. So we literally have tens of people in the lab that are banging on this product to make sure they’re going to be the most robust in the industry and make sure that we’re ready before our partners are.

[Sundeep Bajikar:]

Okay, great. And then shifting gears, on the comm side can you give us more color in terms of how you think the operators in China are planning to deploy 100G upgrades. What portion of the network do you think they will be upgrading and in roughly what timeframe? This is one of the big trends I think you called out.

[Ford Tamer:]

So I think the contracts are being handled. The first phases have already been awarded and this is the next generation phase of these upgrades, right? So on the China Mobile side it was stated that they would go more aggressively on the upgrades as soon as the smartphones hit $150 and that has happened. And so we’re seeing both China Mobile and China Telecom roll out aggressive upgrades in 2014 for TD LTE handsets. And so the awards have been handed now and the rollouts should happen in 2014.

[Sundeep Bajikar:]

Okay. But in terms of the 100G links that are being added, what portion of the network do you think they’re getting added in?

[Ford Tamer:]

The 100G would be definitely in aggregation and core as opposed to access. So that would be closer to the core of the network as opposed to the access next to the customer.

[Sundeep Bajikar:]

Okay, great. Thank you very much.

[Operator:]

(Operator instructions.) And your next question comes from the line of Quinn Bolton from Needham & Company. Please proceed.

[Quinn Bolton:]

Ford, I just wanted to ask you a quick follow-up question, just coming back to the optical component market. Most of the TIAs and drivers have been [gas] or III-V for a while. It sounds like that’s switching over to CMOS-based solutions going forward at least for datacenter. Can you tell us how you see the competitive landscape for CMOS drivers and TIAs?

[Ford Tamer:]

Sure. Quinn, the first part I would say that that shift will take some time. So it’s not that you’re going to see us abandon our III-V efforts. We actually do feel that having both III-V and CMOS is a competitive differentiation. So we say if you want to learn what happens in a datacenter just look at long haul and metro, right? So we do feed our offerings, both TIA and drivers first and gas or silicon germanium or indium phosphide; and then as the technology matures then we port them to CMOS. And in our mind it’s very unique. There’s on other company that has both. You’ve got companies that have either III-V or CMOS but nobody can do both. And moving forward we think that’s going to be the same. We’re committed to support both and we do see this as a huge differentiation over our competitors. The interesting piece, if you want to learn what happens in a datacenter take a look at the long haul – so the long haul market went from multi-mode fiber to single-mode fiber to WDM to tunable laser and colorless networks, right? And if you look at those four trends our prediction is the same would happen in a datacenter. So we’re learning a tremendous amount from having those circuits tested, working at high-speed and then porting them to CMOS.

[Quinn Bolton:]

Okay. Thank you.

[Operator:]

And your next question comes from the line of Ross Seymore from Deutsche Bank. Please proceed.

[[Bob Gilardi]:]

Hi, this is Bob [Gilardi] for Ross. I was just curious, Ford, I mean are the deployments for the silicon photonics products, are these mainstream deployments or are we still very early days? And kind of how do you see, is this a mainstream technology in 2014 or is it really a 2015 where it really becomes meaningful for Inphi?

[Ford Tamer:]

So I would, we would comment but again you probably should ask directly Cisco, Intel and some of the other folks who have announced products in that space. So at ECOC in London in September Cisco showed their CPAK module and there was an announcement of a router by Cisco that would support the CPAK as the first platform. So this is not research; this is going to be a production product that is going to be shipping in high volume. And I would suggest you ask Cisco for the timing of that. Intel at their IDF also showcased a whole desegregated datacenter vision where silicon photonics was center stage and again, you should ask them about the timing of that. But I think that’s not research; that’s again going towards production. And there’s obviously a few other folks also working on silicon photonics. So I think you’re going to keep seeing a world where both will coexist; so you’ll see a world where some of the more traditional [vixel] technologies will coexist in certain environments along with technologies based on silicon photonics and CMOS for the foreseeable future. And on our side we’re not going to pick camps. We would keep supporting our customers in both camps and be agnostic to the chosen technology.

[[Bob Gilardi]:]

Great, that’s very helpful. The other one, I know you didn’t want to talk about the timing of DDR4 but my understanding is that the number of components will go up with the DDR4 generation. Does that increase your potential content opportunity with DDR4 or is it more of a case where probably not, you just sell more components and the content stays the same? I’m just curious of how you see it.

[Ford Tamer:]

I think the biggest change, Bob, on the DDR4 would be there is an expectation, we’ll see whether it happens, that the price of the register would go up compared to the current price of the register as compared to the buffer. So if you look at DDR3 the reason we’ve had such a focus on attach rate in DDR3 for the buffer is that the buffer is selling at let’s say 10 times the price of the register. In the DDR4 timeframe you will have nine buffers per register, and the expected price of those nine buffers compared to the price of the register is let’s say 1.5x to 2.0x. So you’re not anymore having a 10x delta in ASP or average selling price which even though it’s not as significant should help in de-focusing the debate from attach rate of the buffer to actually the market expanding because of the base price of the register expanding. Early indication that we’ve had talking to customers, the register still will dominate DDR4 and still will dominate sort of the TAM. However we do expect the buffer to become a larger part of the revenue going to DDR4. But we also expect the whole debate on attach rate for buffer would probably take a backseat compared to the overall TAM increase that DDR4 will create which could be significant.

[[Bob Gilardi]:]

Great. Very helpful, thank you.

[Operator:]

Ladies and gentlemen, this will conclude the question-and-answer portion of today’s call. I will now turn it over to John Edmunds for closing remarks.

[John Edmunds:]

Thank you, Phillip. Inphi plans on attending the RBC Conference in New York on November 12th and the Needham Conference in New York on January 14th to 16th timeframe. We’ll also be visiting with investors in Columbus, Ohio, and in Toronto with Piper Jaffray and in the Midwest with Craig-Hallum this quarter. Ford and I would like to thank you for joining us today and we look forward to speaking with you again in the future.

[Operator:]

Thank you for your participation in today’s conference. This concludes the presentation. You may now disconnect. Have a wonderful day.