Bitwise launched the NEAR Protocol ETF — ticker NRR — on the NYSE Arca on September 29. First spot NEAR product in the U.S., directly holding the token, with a twist: it doesn’t just custody NEAR, it stakes it, aiming to pass the ~4.9% network yield through to shareholders after the 0.75% fee. First two sessions: roughly $50 million in. First week: closer to $58 million. NEAR itself ran up nearly 180% in September before the profit-taking set in.
The launch is a real milestone, and the price action is real money. But I’ve learned to distrust the moment a fund’s own marketing becomes the market’s analysis. So I did what I always do: opened the numbers and read them cold. What I found is a gap that the headlines are politely skipping — and it’s the only part that actually tells you where this goes.
The pitch: a staking ETF for the AI agent economy
Let me be fair to the bull case first, because it’s not stupid.
The story Bitwise CEO Hunter Horsley is selling to institutions goes like this: NEAR is the layer-1 built by a co-author of the transformer paper, positioned to move AI agents and capital across chains. Its cross-chain settlement layer, NEAR Intents, has pushed past $31 billion in cumulative volume across 30+ chains, with around $3 billion in the last month alone. When allocators look at AI startups commanding billion-dollar valuations with no revenue, Horsley’s line is that NEAR “might be the cheapest thing you’ve ever seen in your life.”
And the ETF itself has a genuinely novel feature: staking. NRR stakes essentially all of its NEAR, grossing about 4.87% and netting about 3.29% after the sponsor fee. That’s not a meme-coin wrapper — it’s a yield-bearing wrapper, and it gave the fund something the older spot Bitcoin ETFs never had: a reason to hold it besides price.
So far, so coherent. Now the other side of the ledger.
The math that doesn’t cooperate
The part everyone quotes is NEAR Intents’ volume. The part that matters is what that volume actually buys.
NEAR’s design routes transaction fees into token buybacks — the volume is supposed to be a direct lever on supply. And over the last 30 days, those buybacks totaled about $1.9 million. Meanwhile, over that same window, the network minted roughly $13 million worth of new NEAR.
7x — the gap between NEAR’s new issuance and its fee-funded buybacks — buybacks must grow sevenfold just to break even
Read that again. The buyback mechanism — the whole deflationary heart of the thesis — is running 7x short of new issuance. The token isn’t being burned faster than it’s printed. It’s being out-printed, comfortably, even while the “AI agent economy” narrative is at peak volume.
This isn’t a criticism of NEAR specifically. It’s a reality check on the entire “fee → buyback → scarcity” playbook that half the market has been selling since EIP-1559. Buybacks funded by fees are only deflationary if fees outrun issuance — and for NEAR, right now, they don’t.
The supply reform that’s really going on
Here’s where it gets interesting, because the market is quietly doing the honest thing.
NEAR’s community is debating a governance proposal to cut maximum annual issuance from 2.5% to 1.6%, phased in over two years. That would avoid roughly 66 million new tokens over six years — about $329 million at late-September prices. The tradeoff is written in plain sight: staking yields fall from about 5.4% to 3.5%, because those rewards are the newly issued tokens.
Think about what that proposal actually is. It’s an admission that the buyback math isn’t self-sufficient — that if you want scarcity, you don’t juice the buyback, you cut the faucet at the source. That’s the responsible move, and it’s also a tell: the network’s own governance is signaling that issuance, not buybacks, is the number that controls the supply story.
It also quietly reshapes the ETF’s pitch. A 4.87% gross staking yield is only that high because issuance is that high. Cut issuance to defend scarcity, and the staking yield NRR sells to shareholders drops with it. The two features — “staking yield” and “supply discipline” — are pulling in opposite directions, and the fund can only advertise one at a time.
The drama nobody priced in
Then there’s the security test, because no L1 story in 2026 is complete without one. NEAR Intents took a $3.8 million exploit in early October, through a bug in its Omni deposit-and-withdrawal system. The funds were recovered in full within about 24 hours — general manager Alex Shevchenko negotiated with the attacker and set a 48-hour deadline for return.
Full recovery is a good outcome, and I’ll give NEAR credit for it. But “recovered” is not the same as “didn’t happen.” A cross-chain settlement layer that a hostile party can reach into — even briefly, even reversibly — is a layer whose security margin is still being written. When the investment thesis is “this is the settlement rail for AI agents moving capital,” a $3.8 million exploit is the kind of footnote that deserves more than a footnote.
What I’d actually do with this
None of this means the NRR trade was wrong. It means the trade and the story have drifted apart, and you should know which one you’re holding.
Here’s what I’d watch before touching it either way:
- The buyback-to-issuance ratio, not the volume. NEAR Intents volume is a vanity metric if it doesn’t close a 7x gap. Track fees against minting, quarter by quarter.
- The issuance proposal’s fate. If 2.5% → 1.6% passes, the scarcity story gets real — and the staking-yield story gets smaller. Decide which one you’re actually buying.
- The next exploit’s recovery, or lack of it. One reversible $3.8 million drain is a scare. A second one, or a non-recovery, is a thesis change.
The NRR ETF is a genuinely interesting instrument — the first staking-enabled spot wrapper for an alt-L1, and a clean on-ramp for money that wants the “AI meets crypto” story without touching a wallet. But an on-ramp and a thesis are different things. The on-ramp is real and working. The thesis — that NEAR’s cross-chain volume will outrun its own printer — is still $11 million a month short.
(The End)
Bitwise NEAR Protocol ETF (NRR) launched Sept 29, 2026 on NYSE Arca; ~$50M first two sessions, ~$58M first week; 0.75% fee; Coinbase Custody; ~4.87% gross / ~3.29% net staking. NEAR Intents cumulative volume $31–33B across 30+ chains. Buybacks ~$1.9M vs ~$13M issuance over 30 days. Issuance proposal 2.5%→1.6% (avoid ~66M tokens over 6 yrs). $3.8M Omni exploit recovered within 24h. Figures per CryptoBriefing, MEXC, CoinMarketCal and Bitwise filings.
