Summary
- SOXQ has been a source of wealth destruction since its inception last year.
- Valuations look attractive.
- The underlying fundamentals in the industry look unappealing.
The more inventory a company has, the less likely they will have what they need. - Taiichi Ono
Introduction
It would be a good time to segue into an analysis of one of the newer semiconductor-based products around; I’m referring to the Invesco PHLX Semiconductor ETF ( SOXQ ) which only made its debut last year in June.
You could argue that SOXQ picked a rather inopportune time to come to the bourses, as general sentiment towards most risk assets has been quite lackluster for over a year now. This has no doubt played a part in the fund’s negative returns since inception, but then again, also consider that during this period, it has also ended up underperforming both the S&P 500 and the Nasdaq!
Could this underperformance continue? Well, here are my thoughts.
What to consider?
After this prolonged underperformance, I’d like to think it could stimulate the attention of a few bargain buyers, who would love the opportunity to own a cohort of stocks that have traditionally served as useful sources of alpha but have rarely ever traded at cheap valuations. For additional clarity, do note that SOXQ’s holdings can currently be picked up at a weighted average P/E of 15.4x (typically these stocks usually trade at multiples of over 20x) which is a lot cheaper than the corresponding multiples of the S&P 500 (17.4x) and the Nasdaq ( 23.3x ).
Looking at the P/E ratios in isolation, particularly for growth cohorts, can be an incomplete exercise. It's also worth reiterating that semiconductors are typically pursued for the earnings growth potential that they provide, and when one ties this in with the P/E on offer, do the valuations make sense?
Well, if you look at most of SOXQ’s top 5 major stocks, who have sizeable aggregate weights (almost 40% of the total portfolio), you have to say there’s pretty good value on offer here, as their current forward PEG ratios are all trading at hefty discounts ranging from 23-78% versus their historical averages, more than justifying the P/Es. Only Intel (INTC), the fifth largest stock within SOXQ, comes across as unappealing from a PEG angle, as its earnings look poised to decline YoY.
Forward PEG | PEG-5 Yr Average | Differential % | |
Texas Instruments (TXN) | 1.7 | 2.3 | -26.4% |
Broadcom (AVGO) | 0.8 | 1.1 | -24.5% |
AMD (AMD) | 0.6 | 2.7 | -77.8% |
NVIDIA (NVDA) | 1.8 | 2.35 | -23.4% |
Source: Seeking Alpha
Then, as noted on yesterday’s Lead-Lag Live podcast episode with Eric Balchunas, tracking ETF flows can be quite useful in giving you an edge in an ultra-competitive marketplace. If one looks at the most recently available fund flow data for SOXQ, we can see that Q3 has pretty much been dominated by fund inflows giving you a sense of sentiment is slowly beginning to shift.
Thus, whilst the valuations and fund flow angles look encouraging, I’m not sure the underlying fundamentals in the sector look in great shape. The words- “chip shortage” was something of an omnipresent phenomenon across most financial portals last year; whilst that still lingers in some pockets, it’s fair to say that things have probably peaked and look to be on the way down. Data from Bloomberg shows that chip delivery times have been on a declining trajectory for the past two months, coming in at around 26.8 weeks in August.
Notable semi-entities such as TSMC have been warning of significant inventory build pressures which would need to be unwound until H1-2023, at the very least. What's also contributing to the adverse inventory situation is the drop-off in demand, in places such as Europe. According to Jefferies, inventory levels in that region are at historic highs, and with the energy crisis eroding purchasing dynamics there, expect significant pressure to linger. As can be seen from the image below, indices measuring new orders for tech equipment, auto parts, and electronics all continue to slide.
All in all, according to World Semiconductor Trade Statistics, after seeing double-digit growth over the past two years, global chip sales will only grow at a miserly rate of less than 5% in 2023.
Conclusion
SOXQ has had a rough ride this year and it is not alone; in fact, as noted in this week’s ‘Leaders-Laggers’ section of The Lead-Lag Report, the steady underperformance of the broad tech sector relative to other growth and high-beta cohorts has been a cause for concern; to clarify further, a ratio measuring the strength of tech stocks over the S&P 500 has now dipped below its 20DMA.
This week’s latest CPI reading could only make things worse for the SOXQ and the broad tech sector. As noted in a tweet on the timeline of The Lead-Lag Report, the probability of a 100 bps point hike this month has picked up steam; with even tighter monetary conditions ahead, semis are a hard sell.
For further details see:
Invesco PHLX Semiconductor ETF: What To Consider?