I’ve had my issues with Inphi’s (IPHI) valuation, as you have to stretch and look at what the market has historically been willing to pay for exceptional growth to drive an estimate of fair value, but I’ve had no issue with the performance of the company. In a quarter where even many tech companies saw significant revenue declines, Inphi more than doubled revenue on the back of strong demand in data centers and telecom, and demand for leading-edge data center solutions doesn’t seem to be flagging.
It’s possible to argue that Inphi shares are worth more than $130 on the basis of what the market has historically paid for similar growth stories, and some sell-side analysts are arguing the shares are worth more than $160 based upon what the market is currently willing to pay for stocks like AMD (AMD) and NVIDIA (NVDA). I don’t really like valuing stocks this way, but it’s hard to argue that Inphi isn’t in a rare company where growth is concerned and that the opportunities in both telecom and data center remain very attractive.
Dusting Expectations Once Again
Although Inphi had some challenges with margins (more on this later), it’s hard to call the second quarter results anything but “good”. Management continues to build its “under-promise and over-deliver” reputation, with a revenue number that beat expectations by 17%. The company did give some of this back at both the gross margin line ($0.20/share) and opex lines ($0.04/share), but a $0.28/share beat is a fine result.
Revenue more than doubled from the year-ago quarter and rose 26% sequentially, making for an exceptional result even relative to healthy end-markets. Revenue from the long-haul and metro markets rose more than 100% yoy and close to 50% qoq, with strong sales from 5G and coherent DSPs, as customers continue to switch to merchant silicon. The data center business wasn’t quite as strong but still delivered better than 90% yoy growth.
Gross margin declined about six points from the year-ago quarter and 10bp from the prior quarter. The year-over-year change is largely due to the eSilicon deal, and while operating income did increase more than 200% (and nearly 50% qoq) on a non-GAAP basis, the company is going to be spending more on R&D – news that growth stock investors seldom like hearing.
Management set a bar for third quarter revenue that was 16% above the average sell-side estimate going into the quarter, suggesting around 3% qoq growth, and again I’d note management’s history of surpassing its own guidance. Management did also guide to modestly lower margins, though, with expected increases in R&D spending.
Still Waiting For Competition To Emerge As A Bigger Factor
Inphi benefited from multiple strong drivers in the second quarter. As mentioned above, the company is benefitting from a switch toward more merchant silicon among its telecom customer base, and that is driving better sales of Canopus DSPs.
Historically, long-haul/metro customers designed their own chips (moreso on the long-haul side than metro), but that’s starting to shift, and if the performance advantages and full lifecycle cost advantages continue to justify switching to Canopus, Inphi is likely looking at not only share growth in the DSP market but an expanding addressable market. Acacia (ACIA) has been the leader in this market in the past, but it remains to be seen if Cisco (CSCO) will continue to allocate the same level of resources going forward, and that could be another driver in Inphi’s favor.
Another major driver this quarter, this time on the data center side, was strong ongoing demand for the company’s PAM4 solutions. MaxLinear (MXL) wasn’t expected to be much of a serious rival to Inphi, but Broadcom (AVGO) was, and it looks as though that company just isn’t making much headway this year – as I mentioned in my last update on Inphi, I did believe that COVID-19 could delay qualification/certification efforts, and that seems to be happening. So, if you’re Google (NASDAQ:GOOG) (NASDAQ:GOOGL) or Amazon (AMZN) and you need a top-end PAM4 solution, you go to Inphi.
As you might expect, then, the strong 200G/400G ramps and the slower entry of Broadcom means that Inphi is maintaining a higher-than-expected market share this year – likely closer to 70% than the previously-expected 65% (which was a boost from the 60% expected around the start of the year). At the same time, though, hyperscale spending remains strong (stronger than expected), and management boosted its estimated addressable market in 2021 by more than 10% (to $500M).
ColorZ also remains a strong driver in the data center, with the company’s strong relationship with Microsoft (MSFT) largely driving this. There is still some risk around the transition to ZR in 2021, but it sounds as though ZR could ramp more quickly than expected, and I believe Inphi won’t really see much turbulence in that transition – it remains a risk (particularly with Microsoft), but not a high-probability risk in my view.
Just about everything seems to be going Inphi’s way right now, which I suppose could be a sort of bear argument in its own right – how much better can things go for the company than they are now? I do still expect strong ongoing investment in data center infrastructure (especially hyperscale), but I do also believe that Broadcom will eventually make more of a showing in the market. I’d also note that while Inphi is no longer counting Huawei in its guidance, it has still been shipping to Huawei, and I do get the sense that that has driven at least some of the upside seen so far this year.
With Inphi executing at an even higher level than I expected, data center spending remaining robust, and perhaps some emerging opportunities in DSPs, my long-term revenue growth rate goes up slightly – from the “mid-to-high teens” to the low end of the high teens. While greater investment in R&D will depress near-term FCF margins, I wasn’t expecting particularly impressive near-term FCF margins anyway, and Inphi isn’t being valued on the basis of its cash flow generation. For what it’s worth, though, I still do believe Inphi can generate adjusted FCF margins in the mid-20%s over time.
As I said in that prior piece, you can’t really value Inphi on its own merits and get anywhere useful – the shares trade well above discounted free cash flow, as well as typical norms for the company’s expected near-term margins. If you look at what the market has historically been willing to pay for similar growth, though (in terms of year-ahead PE), I believe there’s still an argument that the shares are undervalued below $130, and as I said before, there are some analysts arguing that the shares should trade above $160.
The Bottom Line
At some point, there’s going to be a hiccup in Inphi’s growth – whether that’s from a slowdown in data center spending, shares gains from Broadcom, or some other source. The impact on the share price will likely be ugly, at least in the near term, and so this does feel a little like a game of musical chairs now. That’s not meant as a sharp criticism of Inphi, but just the reality of investing in high-multiple growth stocks. That will probably be when investors like me take another look at valuation, but in the interim, I do see how and why growth/momentum investors can still find opportunities here.
Disclosure: I am/we are long AVGO. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.