If Micron’s CEO is right, these 3 memory stock ETFs can rally through 2027

Micron's (NASDAQ:MU) Fiscal Q3 2026 results include something you should look into a little more closely. The company's CEO said, "We expect tight conditions to persist beyond calendar 2027..." The tight conditions here refer to the supply crunch for DRAM and NAND, which has led to MU stock soaring by 176% year-to-date.

Of course, the stock has cooled significantly from its peak, and many other related stocks have taken a dive, but if he's right, we could soon see a reversal. Memory businesses may have much more pricing power and growth left before any cyclical slowdown or downturn.

Past selloffs of this scale have led to an even bigger surge down the line. Micron isn't too big yet for this to happen, given it's still a sub-$1 trillion company (albeit by a hair) and trades at a 6x forward PE ratio.

Here are three memory stock ETFs to look into if the company's CEO is right about the memory cycle being longer:

Roundhill Memory ETF (DRAM)

The Roundhill Memory ETF (BATS:DRAM) is the first pure-play ETF focused on memory and storage chip makers. This ETF remains the most popular way to play the memory trade, and I have no doubt DRAM will surge if the memory rally continues through 2027 or plunge if the cyclical downturn hits earlier. If you believe Micron's CEO, the latter is less likely.

Before making a decision, the biggest thing to keep in mind is China. If the U.S. government allows major companies to import components freely from there, you're going to see a massive influx of Chinese components.

But that doesn't mean the supply crunch is entirely artificial due to U.S. policy.

China does not have the advanced EUV lithography machines for the highest-end memory chips, so they can only brute-force the mass production of standard consumer memory. Many memory makers have already exited those fields, so I expect the DRAM ETF's holdings to continue climbing as long as AI spending remains solid.

The DRAM ETF is up 91% year-to-date. It was up 191% at one point but has cratered since. Its largest holdings are a mix of U.S. and non-U.S. memory makers.

Franklin FTSE South Korea ETF (FLKR)

Speaking of non-U.S. memory, you can look into the Franklin FTSE South Korea ETF (NYSEARCA:FLKR). There is a difference between this ETF and its more popular counterpart, the iShares MSCI South Korea ETF (NYSEARCA:EWY). The difference is that FLKR comes with a 0.09% expense ratio, whereas EWY charges 0.59%. Total return has been essentially identical.

And if you are unaware of why we're looking at Korea specifically, it's because the country is home to two memory heavyweights: SK Hynix (NASDAQ:SKHY) and Samsung. SK Hynix made a blockbuster debut in the U.S. stock market just days ago.

South Korea's stock market has also been on a roll, as the government is propping it up through a "value-up" program to bridge the "Korea Discount." Korean stocks have historically traded cheaply relative to global stocks, and you could argue this remains the case, as memory stocks trade at just 6-7x earnings.

But again, it's hard to say whether or not we've reached a top yet. FLKR is down 25% from its June peak. A similar selloff happened from late February to late March, so I wouldn't be too fearful.

VanEck Semiconductor ETF (SMH)

The two ETFs above will let you dip into most major DRAM and NAND stocks. VanEck's Semiconductor ETF (NASDAQ:SMH) does not expressly target memory, but if you believe that the memory rally will continue through 2027, you must also believe that semiconductor stocks will ride along. Both components are necessary to train and run AI models.

SMH has arguably been the single best major ETF you could've owned as a buy-and-hold play in the past 20 years. No one knows what the next 20 years may bring, but it's not a stretch to believe that the rally could go on for at least one more year. The "cool-off" in the past month is a drop in the bucket compared to SMH's 363% 5-year return, as it is only down 7.1% in the past month.

There have been two 30%-plus corrections in the past, but the SMH still recovered every single time. The demand for chips is broader and was outperforming the broader market well before AI became a thing. Thus, you may as well load up on SMH if you are loading up on memory stocks. In fact, I believe SMH will outperform any memory-focused ETFs because it is less cyclical. The expense ratio is 0.35%, which is negligible against the performance.

Contact [email protected] for any questions or corrections.