The ETF record belongs to three funds, not the shelf

Three funds are carrying the ETF industry's record year, a fact that makes the record narrower than the headline suggests, and the $2 trillion forecast for 2026 is Q4 seasonality applied to this year's run rate—a projection leaning as much on the calendar as on demand. Underneath it, the intake keeps routing to the same few tickers, while the rest of the shelf competes for what the biggest names leave behind.

Extrapolations built that way have a specific failure mode: they assume the fourth quarter resembles the quarters that came before it and treat the current run rate as a floor rather than a peak. Composition matters more than the total, and the composition is lopsided—the industry can post a record year while most of its products gather nothing of consequence.

The funds doing the carrying are the cheap, liquid ones, and price is the entire pitch. Broad beta is a solved product, so an issuer's pipeline has to answer a question the price war already settled: what do you sell a client who owns the beta, holds it in the cheapest wrapper available, and has no reason to move?

This week's product news reads as three answers to that question, and they point in one direction: BNY's ETF suite is at $20 billion and mostly free beta, Amplify filed a sports ETF with a 15% private allocation, and three funds filed as value under three different index rule books and finished the year 1,100 basis points apart. That is new shelf space in each case; none of it moves where the money is going.

The BNY milestone measures distribution rather than revenue, and the suite runs five funds, so the mix carries the whole story: the cheapest exposures gather the assets, while the fee-bearing corners of the lineup have to earn against them. A $20 billion book whose bulk does not bill is a license to sit on the platforms where allocation decisions get made—a gathering machine first and a revenue line a distant second.

Read the milestone's audience correctly and it makes more sense: a book that size is a credential presented to the gatekeepers who decide what gets a slot, and a free-beta core is proof that the issuer can compete at the commodity end of the market. The fee-paying part of the lineup is the actual business, and it has to be sold on something the free fund cannot offer.

The trade is still worth making because presence compounds. An issuer sitting on the platform sees the next mandate, the next model change, the next conversation, none of which shows up in the fee on the cheapest fund. But $20 billion gathered in exposure that charges nothing is also a price signal every rival can read: a new broad-market launch cannot win on cost when the cost is already at the floor.

That leaves structure—a screen, a sleeve, a story, a wrapper that does something the cheap fund cannot, and that is precisely what the rest of the week's pipeline offers. The industry is adding complexity at the moment its customers are rewarding simplicity most clearly, and the reason is not that anyone has misjudged the client. The simple product no longer pays for the distribution required to sell it.

The arithmetic cuts in two directions. Shelf space is cheap to create and expensive to fill, so issuers keep filling it with products that charge more than beta, while the dollars stay where the price is lowest. The market is telling issuers where a fee can still live: in whatever a client cannot already get for free.

Amplify's sports ETF is the most consequential of the three because the filing carries a number most thematic products avoid: a 15% private allocation. Fandom sells the pitch—leagues, rights, the media economics of sport—but the sleeve is the part that will decide whether the fund works. A daily-priced wrapper holding private assets has to publish a quote every day for a book that gets appraised on a slower clock, and the gap between the two shows up as a premium or discount whenever the market's mood and the last mark disagree.

That mismatch is the product's public face, not a footnote in the prospectus. The sleeve is where valuation cadence stops being a compliance question and starts being the thing that determines whether the ETF trades at its net asset value, and the first divergence between the tape and the appraisal will be read as a statement about the fund rather than about the calendar the assets are priced on.

Plainly, this is a filing—not a launch and not assets. The registration sets the boundary at 15%; the fund will be judged on whether it holds that line once the sleeve has to be marked and the shares have to trade.

Fifteen percent, not the sports theme, is the number the industry should be watching. If sleeves of that size become a habit in daily-priced wrappers, the ETF structure is being asked to do a job slower-pricing vehicles handle more comfortably; the funds that do it well will look boring, and the ones that do not will have a discount tell the story before the manager gets a chance to.

The value funds show what happens when a category is asked to carry information a category cannot hold: three funds file as value, follow different index rules, and finished the year 1,100 basis points apart. Screening on a category rather than a methodology costs roughly the width of that gap, and the label itself gives the client no warning about which rule book sits behind it.

Value covers whatever each index says it covers, which is why two funds can share a label and hold almost nothing in common. The dispersion is not a market inefficiency waiting to be arbitraged; it is a word doing what words do, and 1,100 points is what the word costs a client over a year when nobody reads past it.

Labels stay sticky because they are the language platforms and clients already use. A model portfolio needs a value sleeve, so a value sleeve gets filled, and the methodology question gets deferred to whoever bothers to ask it. There is a fee argument buried in the gap as well: an expense ratio is defensible against a rule book and nearly impossible to defend against a category name, and the spread between the best and worst readings of one word is wide enough to outweigh the fee itself.

The category screen survives on convenience. That is the cost of buying a word instead of a methodology, and the buyer pays it annually.

The mid-year scorecard arrives at the same conclusion from the other direction. Breadth widened, and 67% of large-cap funds still lost to their benchmarks—exactly the environment in which stock pickers were supposed to earn their fees, with more names leading the market, more chances to be right, and less of the index's return concentrated in a handful of mega-caps.

That wider market still left two-thirds of large-cap funds behind, which turns the active-management case into an argument about wrapper pricing rather than stock picking. If more opportunity doesn't lift the majority past the benchmark, the justification for a more expensive or more complicated product has to rest on something a client can observe, not on the market supplying more opportunity. The private sleeve and the value label run into the same wall; complexity is easy to file and hard to get paid for.

The private markets, meanwhile, keep holding the assets that would make some of these listed products easy to sell. Brookfield and ACME announced a $600 million energy deal, one line in a day that also carried clean-energy and data-center announcements across two continents. That timing sits awkwardly for issuers who listed exposure to the theme without the underlying assets, as this publication's coverage made clear this week.

The week's agenda is narrower than the flow forecast. Amplify's 15% cap is the first thing to watch: if a second issuer files a private sleeve of real size in a daily-priced wrapper, what looks like one registration becomes a structural question. BNY's suite is the second, because a free-beta book that size either attracts fee-bearing siblings or stays a gathering asset, and the mix will say which. The concentration is the third, and it is the number that actually summarizes the year: the share of the intake still landing in the same three funds. The next flow print will show whether that is still true; if it is, the record holds without the shelf's newest products contributing anything to it.