From time to time you’ll see sell-side analysts put themselves through contortions that would make an acrobat reach for the Advil, and that often happens when they feel the need to justify ever-higher price targets for strong operating stories that have already breached prior targets. That seems to be the case today with ASML (ASML), as the sell-side has defended a lower 2021 outlook from this highly-valued semiconductor equipment supplier as “providing certainty and confidence”.
Let me be clear – I love ASML’s business and would be happy to own it at the right price. I just think it’s interesting to see how other analysts will justify their price targets when valuation gets extreme. I found these shares too expensive last quarter and they have since underperformed the NASDAQ and the SOX, and have actually declined a bit. Although the shares certainly aren’t a bargain by conventional means, it’s a name to watch if this recent tech correction accelerates.
A Strong Quarter, Driven By EUV
ASML did well for its shareholders in the third quarter, with revenue beating expectations by 7% on strong EUV equipment sales and good operating leverage that drove a 13% beat at the operating income line.
Revenue rose 33% yoy and 19% qoq, with system revenue up 33% and 27%, respectively, beating expectations by 8%. EUV shipments doubled from the year-ago and quarter-ago levels, generating two-thirds of equipment revenue (the first time EUV has exceeded 50%) and beating expectations by 20%. Overall units were down slightly from the prior quarter (60 vs 61), but the greater mix of EUV and the increasing ASPs for those systems (customers ordering higher-priced configurations) drove ASPs more than 31% qoq.
Service revenue rose 31%, but declined 3% qoq. That was good enough for a small beat versus the sell-side, but I was honestly a little surprised that ASML didn’t do a little better, as access to customer sites should have improved on easing COVID-19 restrictions. This is not a red flag, just one of those things that makes me go “hmm”.
Gross margin improved almost four points from the year-ago level and slipped about 70bp qoq, good for an in-line performance relative to the sell-side. Operating income rose 77% yoy and 34% qoq, with margin improving close to eight points from last year and three and a half points from the prior quarter.
A Mixed Outlook, But More “Noise” Than Noisome
ASML’s guidance offered a little bit of upside for the fourth quarter, which was of course welcome. Guidance for the next year, though, is leading to a roughly 3% decline in sell-side expectations so far. While guide-downs from richly-valued tech companies can hammer the stocks, the sell-side has been trying so far to spin this as “providing certainty” about the 2021 outlook given concerns about challenges and push-outs at Intel (INTC) and TSMC (TSM).
For my part, a one-year “burp” in the trajectory of the EUV ramp doesn’t really concern me all that much. Likewise, orders are always going to be volatile on a quarter to quarter basis. This quarter saw order value decline 44% yoy, but jump 160% qoq, with 73 tools going into the order book. Of that total, EUV orders amounted to 4 units, versus 23 last year and 3 in the prior quarter.
Although there have been some hiccups in the node transitions expected in 2021/2022, logic demand remains healthy, and I expect it to remain so, with improving demand from autos, ongoing growth in data center and 5G deployments, and some uplift from smartphones. On the memory side, management is looking for a recovery in equipment spending around the turn of the year, but a lot rests on inventory levels and smartphone demand (healthy data center demand doesn’t hurt). Consider the example of Samsung; this large memory company will be using EUV for its 1znm process, but tool demand is ultimately going to be tied to demand for LP DDR5 DRAM.
ASML management has been seeing high utilization levels for its tools, and that’s positive for tool demand. EUV penetration is also continuing to grow, with EUV being used on 10-plus layers at 5nm and management is looking for 20-plus layers at 3nm.
Stepping back a bit, I still like the long-term EUV story for ASML. EUV is already positioned as an essential tool at 5nm and 3nm, and ASML continues to drive tangible benefits from its R&D efforts. The new “D” system offers a roughly 20% improvement in productivity and a higher ($150M ASP), and I expect further improvements in productivity and price for the upcoming “E” system.
New trade restrictions (requiring technology export licenses) against Chinese semiconductor company Semiconductor Manufacturing International Corp (OTCQX:SMICY) (“SMIC”) have also created some noise. My understanding is that ASML can ship EUV tools from the Netherlands without running afoul of the U.S. government, but if SMIC cannot access the other tools its needs (etch, deposition, et al), it’s a moot point. Sales to China are about 20% of ASML’s business, and it remains to be seen if there will be further turbulence here – call this a “known unknown”.
Although Canon’s (CAJ) nanoimprint lithography technology could pick up some share, I don’t know of any companies using it outside of a single Japanese memory company (Toshiba), and I have my doubts as to whether it can be a disruptive force in advanced logic nodes. As far as logic goes, the EUV opportunity belongs to ASML and I see more risk to ASML’s ability to profitably scale up capacity than to demand over the next five years.
As far as the model goes, I’ve made some adjustments to 2020 and 2021 numbers, but I’m still looking for long-term revenue growth of around 10% (and ’19-’24 growth of 12%). Relative to the Street, my 2024 numbers are actually at or above the high end, so I don’t think I’m exactly selling the growth opportunity short. On the margin side, I expect FCF margins to improve into the 30%’s over the next five years, with a long-term FCF growth rate in the mid-teens.
Valuation is where everything goes screwy. ASML already trades well above industry norms for EV/revenue and EV/EBITDA, but given its effective monopoly position in a key enabling technology for advanced chip production, it should trade at a premium. You can argue that ASML should get an earnings multiple equivalent to other top-tier tech companies (typically in the low-to-mid-30’s), but that really only just supports today’s valuation. Likewise with discounted cash flow; you have to drop the discount rate to around 6% to drive a worthwhile fair value today.
The Bottom Line
I fully realize that you have to set aside valuation from time to time, especially if you want to participate in the great secular growth stories. I also realize, though, that even with the best growth stories there are still parameters of valuation you have to keep in mind. Give me a 10% pullback in ASML shares and I may be tempted to buy in, but this is still not a compelling enough opportunity for me at this price.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. 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.