Morgan Stanley’s MSBT Is the Cheapest Bitcoin ETF. The Fee Is Not the Point.

Conclusion first: Morgan Stanley’s MSBT charges 0.14 percent a year, the lowest fee in the US spot bitcoin ETF market, and the fee is the least interesting number in the fund. What changed in April 2026 is not the price of holding bitcoin in a brokerage account. It is who is allowed to sell it to you, and through which channel the sale arrives.

Morgan Stanley Bitcoin Trust listed on NYSE Arca on 8 April 2026. It holds bitcoin rather than futures, prices off the CoinDesk Bitcoin Benchmark 4 PM NY Settlement Rate, keeps the coins with Coinbase Custody in cold storage, and runs cash custody, administration and transfer agency through BNY Mellon. It arrived at 0.14 percent, undercutting every rival on the board, and it is the first US spot bitcoin ETF issued under the name of a major American bank.

Those are three sentences of description, and none of them explains why the fund exists. The explanation runs through the fee, the custodian and the advisor list, in that order, which is the reverse of the order in which those three things get reported.

0.14 percent is a marketing budget, not a cost advantage

Here is the fee board in the US spot bitcoin market as the product sheets print it. The gap between the cheapest fund and the largest one is 11 basis points.

Run the arithmetic on a normal account before you feel anything about that spread. Ten thousand dollars at 0.14 percent costs 14 dollars a year; the same ten thousand at 0.25 percent costs 25. The saving is 11 dollars, which is about 92 cents a month. At 100,000 dollars the gap is 110 dollars a year. Anyone choosing between these funds specifically to capture that difference is optimising a variable that the trading spread eats on the day they buy: on a fund trading tens of millions of dollars a day, half a percent of slippage on entry is four times the annual fee difference on a ten-thousand-dollar position.

So why does a bank price a product below its competitors and keep it there? Because for the issuer the expense ratio is not really revenue. It is customer acquisition cost, and the acquisition target is not a bitcoin buyer. Morgan Stanley’s advisory business runs about 16,000 financial advisors and, by the firm’s own reported figures, more than $6.2 trillion in client assets. A fund that keeps a household inside that relationship for a decade is worth far more in advisory and platform fees than the fund’s own management fee, and a fund that pioneers the category under the bank’s own name is a marketing asset that does not show up in the expense ratio at all.

Note what this means for the fee war narrative. A price war is a fight over a margin that the participants have already decided is not where their money comes from. The firm cutting to 0.14 percent is not the firm taking a loss to win share; it is the firm with the largest non-fee way to monetise the same client, pricing the door open.

A bank issued the fund and still will not hold the coins

The structural change is smaller than it sounds once you read the service providers. Morgan Stanley is the sponsor and the brand. The bitcoin sits with Coinbase Custody, in cold storage, with transfers through Coinbase Prime. Cash custody, fund administration and transfer agency sit with BNY Mellon. The bank sells the product, calculates the fee and carries the name, and does not take custody of a single coin.

That is a deliberate boundary rather than an oversight. Holding crypto for clients used to be expensive for a US bank for an accounting reason: staff accounting bulletin 121, issued in 2022, required a bank that held digital assets for customers to recognise both the asset and a matching liability on its own balance sheet, which turned a custody service into a capital and disclosure problem. The SEC withdrew that guidance in January 2025 with bulletin 122, and bank interest in digital asset services followed. A bank can now build the product and the distribution without becoming the entity that answers for the coins, and that is exactly what this fund does.

The custodian arrangement also tells you where the industry’s risk still sits. Twenty months after the category opened, the asset side of the largest bank-branded bitcoin fund in the market is operated by a crypto-native custodian, and the traditional bank is doing what banks are best at: cash, administration, paperwork and distribution. Anyone expecting a Wall Street balance sheet to stand behind the coins should read the prospectus instead of the press release.

The channel that has not started spending yet

Here is the number that decides what this fund becomes. Inflows so far have come mainly from self-directed clients who found MSBT on their own, not from the advisory channel that gives this bank its scale. That is not a small detail: the largest distribution network in the retail brokerage business has not begun recommending the fund in volume, which means the current asset base is a floor rather than a ceiling. Bloomberg’s Eric Balchunas put the fund on a path to $5 billion in assets in its first year on the strength of that channel.

The path from here is arithmetic. One percent of $6.2 trillion is $62 billion, which is roughly one hundred times the fund’s current size. The bank does not need the whole network to engage, or even a tenth of it, for MSBT to pass the largest funds launched in 2024. It needs a fraction of the advisors to put a fraction of each client’s allocation into a product with the lowest fee on the shelf and the bank’s own name on the cover, and it needs the suitability process to clear.

That is why the 0.14 percent is not a discount. It is the price of entry into a channel that no crypto-native issuer can rent at any fee, because the channel is not a website or a trading app. It is a room of advisors with existing client relationships and a compliance department.

8,000 BTC and a streak that needs a denominator

The fund’s growth record is real and worth stating precisely.

Now the caveat that the streak needs. A run of 20 sessions with zero net outflow days sounds like relentless demand, and part of what it measures is the absence of sellers. A newly launched fund gathers assets before it faces redemptions, and the flows inside those weeks were concentrated in a handful of large trading days rather than spread evenly across the run. The streak is a fact about the ledger; demand is an inference from it, and the two are not the same claim.

Scale puts the record in proportion as well. At roughly $640 million, MSBT is a rounding error next to the largest funds in the category, which hold tens of billions of dollars. The interesting quantity is not the level. It is the slope, and whether the slope changes when the advisors switch on.

What the fund actually tells you

Strip the marketing and three things are true. The fee went to 0.14 percent because a bank with $6.2 trillion in client assets can afford to treat the management fee as a marketing line, and the category will keep repricing as long as the distribution is the real prize. The asset is still custodied outside the banking system, because a bank-branded bitcoin fund does not have to mean a bank-held bitcoin. And the demand so far is a test of the product page, not of the product’s channel, because the 16,000 advisors who make this fund strategically different have not started allocating.

For anyone choosing between these funds, the practical ranking is unchanged by any of this: check the fee, then check the tracking and the spread, and then ask who is recommending it. The last question is the one the marketing pages do not answer, and for this fund it happens to be the whole story.