© Golden Dayz / Shutterstock.com
$10,000 parked in the Fidelity Fundamental Emerging Markets ETF (NYSEARCA:FFEM) on the last trading day of 2025 was worth roughly $13,000 at the late-May peak near $43.45, before a recent week of selling trimmed it back to about $12,200 by the June 5 close at $40.73. That puts the headline number where it should be. FFEM brushed up against a 30% year-to-date gain in late May and is sitting on a 22% YTD return after the pullback. Over the trailing twelve months the fund is up 52%.
The S&P 500 over the same YTD window? The SPDR S&P 500 ETF Trust (NYSEARCA:SPY) is up 8% year-to-date and 24% over the past year. FFEM has nearly tripled the broad U.S. market’s YTD return while almost nobody is talking about it. The Reddit chatter is essentially zero. The fund still sits in that quiet stretch of the Fidelity lineup where retail flows haven’t found it yet.
What actually drove the run
FFEM is an actively managed emerging markets fund, charging 60 basis points a year. The active part matters here because the fund’s top holdings are concentrated in exactly the names that have done the work in 2026. The top ten positions account for 43% of net assets, so a small group of stocks is driving the bus.
The biggest single bet is Taiwan Semiconductor at 14% of the portfolio, which is roughly a full position size on its own. Stack on Samsung Electronics at about 7% and SK Hynix at 3%, and the three semiconductor names alone make up about 24% of the fund. That is a deliberate concentration on the AI capex supply chain, parked inside an EM wrapper. When the AI buildout ran hot through the first half of 2026, FFEM caught it.
The second engine is Chinese internet and consumer tech. Tencent (about 7%), Alibaba (about 3%), and PDD Holdings (about 2%) together carry roughly an eighth of the portfolio. These were the names that spent years being de-rated by U.S. investors who simply wouldn’t own them, and they are the ones that snapped back hardest once the dollar weakened and global allocators began rotating out of the most concentrated U.S. positions. CATL at about 2% rounds out the China tech sleeve via the EV battery supply chain.
The third engine is India. HDFC Bank, Reliance, and Larsen & Toubro together carry about 5.6% of the fund, with HDFC giving exposure to credit growth and the other two providing industrial and infrastructure exposure. None of those names is doing the heavy lifting on their own, but they are the ballast that keeps the fund from being a pure China-plus-Taiwan trade.
Why nobody is watching
FFEM is small, young, and lives in the shadow of the indexed EM giants. Fidelity’s branding on the product is muted. The fact sheet reads like a sober fundamental research note rather than a marketing pitch. The fund tracks no benchmark mechanically, which means it doesn’t get the passive flow that arrives every time a new dollar lands in a target-date fund or a 401(k) default. It is the kind of product that gets noticed only after it has run, which is exactly the spot it is in now.
Reddit has nothing to say about it. The retail discovery cycle that turns a 30% YTD chart into an inflow surge hasn’t started. AUM data for the fund isn’t broadcasting either, which is part of the story. The price moved before the assets did.
What has to keep being true
The forward read here comes down to three things, and a reader can watch all of them.
First, the AI capex cycle has to keep feeding TSMC and the Korean memory complex. That is the single largest swing factor in the fund. If hyperscaler capex guidance softens in the second half of 2026, a quarter of FFEM’s weight gets re-rated downward in a hurry. The leading indicator is the capex commentary from the four U.S. hyperscalers on their next earnings calls.
Second, the dollar has to keep cooperating. J.P. Morgan’s 2026 outlook notes that the U.S. dollar is still about 10% overvalued versus fair value and that inflows into emerging markets have begun to inflect higher after muted interest the last few years. Franklin Templeton’s house view for 2026 is similar, calling for a weakening dollar that bodes well for emerging debt and equity markets. A reversal in DXY would pull the rug on the FX leg of the trade independent of what the underlying stocks do.
Third, China tech has to hold the regulatory and earnings progress that allowed Tencent and Alibaba to re-rate. The next two quarters of earnings out of those names are the indicator, and any return of the kind of regulatory headlines that defined 2021 to 2022 would compress multiples again.
The honest read is that FFEM’s run is real, the mechanism is identifiable, and roughly half of the YTD move came from a setup that other investors are only now starting to position for. The recent 6% one-week pullback is a useful reminder that EM concentration cuts both ways, and that a fund with a quarter of its book in three semiconductor names will trade like a semiconductor fund on the days the AI trade gets sold. The trade that worked into May is the same trade going forward, which means the upside from here depends on conditions that are visible but no longer cheap. Watch the hyperscaler capex guidance, watch the dollar, and watch what happens to the China internet earnings in the next reporting cycle. Those are the three things that determine whether the next leg looks anything like the first one.