Tuttle Capital Concentrated Memory Stack ETF (HBMX): A Nuanced Play on AI’s Memory Bottleneck
This blog post takes a look at the recent launch of the Tuttle Capital Concentrated Memory Stack ETF (HBMX). It’s a new investment option aiming to ride the wave of booming demand for memory semiconductors, all thanks to the AI revolution.
We’ll dig into HBMX’s investment strategy, what sets it apart from other memory-focused ETFs, and the thinking behind its focus on this crucial piece of the AI hardware puzzle.
Unpacking the HBMX Strategy: Targeting AI’s Memory Appetite
The Tuttle Capital Concentrated Memory Stack ETF (HBMX) just launched on Cboe, and that’s a pretty notable development for anyone watching AI-driven investments. The fund comes with a 0.95% expense ratio and puts its money into pure-play companies across the entire memory semiconductor ecosystem.
HBMX covers everything—from the early development and manufacturing stages, to packaging, testing, and finally, getting those memory solutions out to market. That’s a lot of ground.
At its heart, HBMX wants to tap into what some folks call “AI Alpha.” The logic? AI workloads are exploding, and memory is shaping up to be a real pain point for growth. By zeroing in on companies working to solve this memory crunch, HBMX is hoping to catch some serious upside as demand keeps climbing.
Diverging Paths: HBMX vs. DRAM ETF
There are already other ETFs trying to capture the memory market, with the Roundhill Memory ETF (DRAM) being the most obvious comparison. Both focus on memory, but their strategies and rules for picking companies are pretty different.
DRAM got off to a fast start, pulling in over $14 billion in assets. It’s known for a tight, concentrated portfolio—usually less than a dozen companies. That’s a pretty focused approach, sticking with the biggest names.
HBMX, though, aims for broader exposure. It plans to hold around 20 to 35 companies, casting a wider net across the memory landscape.
Eligibility Criteria: A Tale of Two Funds
The biggest differences come down to how each fund decides who gets in.
- DRAM’s Strict Approach: DRAM is picky. Companies need to get at least 50% of their revenue from memory-related business. They also need a $10 billion market cap and at least $5 million in average daily trading volume. That means only larger, liquid companies make the cut.
- HBMX’s Flexible Framework: HBMX is looser with its requirements. It doesn’t care about market cap or trading volume. The main thing is that a company must have at least 25% of its revenue tied to memory.
That flexibility lets HBMX include smaller or less liquid companies in the mix. Some of these lesser-known names could end up being big winners if AI-driven spending keeps ramping up in the sector.
The Philosophy Behind HBMX: High Conviction, Optimized Risk
Matthew Tuttle leads Tuttle Capital and has a clear vision for HBMX. He says the fund is intentionally both highly concentrated and narrow in focus.
Why? He wants to keep high conviction in the holdings but still optimize the fund’s risk-reward profile. The main strategy sticks to equity holdings or closely related financial vehicles.
Tuttle points out the fund didn’t just appear out of thin air. It’s a response to two pretty big shifts in the market:
- The uneven impact of AI’s advancements on established tech companies in different sectors.
- And then there’s the growing awareness that memory is a key bottleneck—companies that solve these memory challenges could be in for real rewards.
HBMX carves out a spot as a strategic entryway into the AI hardware supply chain. The fund wants to capture the momentum from ongoing government and corporate investments flowing into memory tech.
DRAM gives a concentrated, large-cap, and global look at memory, but HBMX goes for something a little different. It zooms in on sector-specific strategies and digs into the details behind AI’s memory demands.
Here is the source article for this story: Tuttle’s New HBMX Targets Memory Semiconductor Ecosystem