원문 보기
I've already talked about VanEck Semiconductor Fund on this channel. I added it to my tier list and it's one of the best ETFs for semiconductors and that's still absolutely true as we shoot this video. But there's another semiconductor ETF that a lot of people have been talking about in the comments that has been quietly outperforming in 2026, but I haven't mentioned it on this channel, so I definitely want to make sure we do.
It's called SOXX. It has almost identical companies that we see in SMH, charges almost the same commission, and it has outperformed SMH by almost 20% year to date, while we see SMH up 68 % and SOXX up 89%. Same sector, same expense ratio, 20% difference within this ETF due to a structural rule you've probably never heard of unless you actually own this ETF.
So , today I want to take a look at SOXX for the first time and show you what exactly this rule is, and if it's worth it to you and really worth it, then it deserves a place in that portfolio. So, SOXX is the iShares Semiconductor ETF issued by BlackRock. It was launched on July 10, 2001. That's right.
This is one of the oldest sector ETFs still trading with a 25- year track record, multiple semiconductor cycles, including that 2022 bear market that we saw with a 35% decline, which is just crazy if people held it through all that noise. It tracks the New York Stock Exchange's semiconductor index. This includes 30 US semiconductor companies spanning the entire value chain.
We're talking about chip developers, hardware manufacturers, memory suppliers, who, as we know, are going crazy right now. The figures for 2026 are extraordinary, with an 89% increase. Over the last year–179%. In three years–54%. 10-year annualized rate, 35% return over the past 10 years. And if you think about SMH, it was 31%.
So, they outperformed the last decade and both absolutely outperformed the S&P 500, which returned 10.24% over the same period. The expense ratio is almost identical for both of these ETFs, and price is not really a determining factor. The real difference lies in the structure of this ETF. The top five assets are Micron, AMD, Nvidia , Intel, and Broadcom.
Here's the first thing that really catches your eye. At SMH, Nvidia makes up about 16 to 18% of this ETF overall. SOXX, Nvidia occupies 6.94%. Micron, not Nvidia, has the largest position with a score of 8.77. This is no coincidence, and this is where the rule of this ETF really comes into play . It is now limited by the 8% rule.
If you haven't heard of it , it's a structured rule for this ETF that was mentioned in the chorus of this video, and it explains the entire 20% performance gap. No company in the SOXX can exceed 8% of the portfolio at any given time before it is rebalanced. Now the top five is limited to 8%. The remaining 25 companies here, because remember it's made up of 30, are limited to 4%.
This entire rule, in this rule alone, changes how the fund is actually structured compared to SMHH. So why did this 8% cap we saw result in the SMH ETF outperforming the SMH ETF by 20% this year ? That's because 2026 was a year of broad semiconductor growth, far beyond the Nvidia we're seeing. We are talking about mega-capitalized leaders in artificial intelligence, Micron, AMD, Intel, and other hardware manufacturing companies.
If you look at Applied Materials, Lam Research, K L M, they've all shown outstanding earnings this year as the chip manufacturing cycle has really expanded beyond GPUs , which Nvidia focuses on the most, into compute technologies for memory hardware, even advanced chassis, and what that actually looks like from the beginning.
As for SMH, Nvidia, as we said, occupies approximately 16 to 18%. So, depending on Nvidia's performance itself, whether it's good or bad, it could have a significant impact on the returns of this fund. Since SMH has a large stake in Nvidia, that means every other semiconductor company that outperforms Nvidia will essentially leave it behind.
SOXX, with a market cap of 8%, is making it hard to hold other companies in this semiconductor space. Micron, with a share of 8.77%, is a prime example of this. Micron was one of the best large-cap stocks in 2026. This was back when there was a memory shortage for AI, and we still see it when we discuss the DRAM ETF that many people are holding onto .
SOXX has almost double the weighting of SMH, and that started the cycle, which is why this ETF will just completely crush it in 2026. Now I want to be very, very clear about one thing. It's not getting rid of SMH to buy a SOXX video. These two funds are actually two different expressions of the same semiconductor thesis that we're talking about, and it really depends on what you think the next cycle will be like.
So let me show you exactly how it will work out and how it will change. Because if you think Nvidia will continue to lead the semiconductor market, as we start looking at just one company, but getting into, you know, quantum computing, looking at semiconductors, having the majority of the market, the SMH concentration going forward will be on Nvidia, which is really leading.
If you think this cycle will continue with memory ETFs, mid-range hardware companies that will really make a big name, balancing SOXX and really the structure will capture more than what's outside Nvidia. In 2026, the expansion thesis truly won. Whether it will continue in 2027 is, frankly , everyone's business and everyone's question.
Now there is another nuance that is really worth paying attention to. Over a long history, from 2011 to 2022, we look at the decades when AI truly took off or continues to see it. SMH grew by 781% compared to SOXX, which grew by 811. The difference of this 11% was very, very modest, and the AI cycle began precisely in SOXX.
The 10-year annualized return of 34% was slightly above 31% for SMH. The structural difference in the broad cycle is when Nvidia dominates and when Nvidia doesn't. Now here's another thing that a lot of people don't think about when it comes to SOXX. He has a cheaper brother. This is SOXQ. Same index, same 30 assets, same 8% structure, same BlackRock issue.
The only difference is the expense ratio. This is very interesting because we have seen this before with QQQ and QQQM. They create a cheaper version of essentially the same ETF. They have almost identical exposure to what we see. If value is important to you and you are really holding this for the next 15 years, the SOXX index does use a methodology compared to the SMH, it is worth noting the SOXQ because one caveat that they do this, and the reason they do this, is liquidity.
It is significantly smaller when looking at assets under management, which means significantly larger when it comes to buying and selling. Long-term buy and hold, it will make a difference in the cost structure for the trader and really moving the blocks there. SOXX is what you need. So, honestly guys, my honest verdict is, does SOXX make the tier list, and does it really have a place in a portfolio?
It belongs to level A, next to SMH, because it's not higher , not lower, but if you look at these two exceptional tools that look at the semiconductor market, and the niche that we're looking at. If you look at SMH, it has global liquidity, global supply chain researcher, if you look at TSMC, ASML, SOXX is ahead of the 10-year annualized returns, and that's across the broader semiconductor cycles that we're seeing with much more complexity.
If we start looking at the semiconductor position from scratch today, there can only be one winner. I would ask myself : do you think Nvidia will be the leader in the next phase of the next 5-10 years? And seriously, do you think this will be a broader market cycle? Nvidia leading the cycle, SMH, in the broader cycle, it would be SOXX that actually outperforms the competition.
Both are very dynamic when it comes to hedging these risks. And if you want to separate the two, of course, because 71% of this entire portfolio is completely overlapping, except for those international companies that we see in SMH. And if that's a cost-determining factor, SOXQ has the same impact, has half the expense ratio.
The funds you don't know about are usually the ones that do all the work most quietly. So, sure, guys, leave a comment below. Have you heard of SOXX or SOXQ before this video? What do you think the next semiconductor cycle will look like? Will Nvidia dominate? Will it be more widespread? I read every comment and will respond to it fully.
Subscribe if we covered something new here, or if it was for educational purposes. We cover ETFs, and of course we do that every day, and I'll see you guys in the next one.
Comments 0