A friend in Zurich asked me which bitcoin ETF to buy, and the product his broker pulled up was an ETN. Same coin underneath, similar fee, nearly identical chart, completely different legal animal. This comes up all the time, because the same bitcoin exposure trades as an ETF in the US, as an ETN or ETC across most of Europe, and as a closed-end trust in a few other corners of the market, and almost everyone treats the three as interchangeable. They are, right up until something breaks, and then the wrapper is the only thing that matters.
The way I think about it, the three structures differ on three axes. Can you redeem, what actually backs the paper, and whose credit you are exposed to. Fees, spreads, and tracking error are all real considerations, but they are rounding errors next to those three questions in any scenario where you would badly want the answers.
ETFs, and why the redemption plumbing does the work
An ETF holds its price near the value of its assets because of a piece of plumbing most holders never look at. Large trading firms called authorized participants can create new shares by delivering assets or cash to the fund, and can destroy shares by redeeming them. If the fund trades rich against its holdings, APs create shares and sell them until the gap closes. If it trades cheap, they buy shares and redeem. The price discipline comes from market makers chasing an arbitrage profit rather than from anyone acting in good faith, and greed has historically been the more dependable force.
One nuance that surprises people. The US spot bitcoin products are marketed as ETFs but are legally trusts registered under the Securities Act, structural cousins of the big physical gold products rather than of a normal index fund. They do not sit under the Investment Company Act, which means fewer of the classic fund governance protections. What makes them behave like proper ETFs anyway is the creation and redemption loop, plus actual coins sitting with a custodian. At launch the regulator only permitted cash creations rather than in-kind ones, which added friction and a little cost, and that restriction was relaxed later. Canada, for what it is worth, listed genuine spot bitcoin ETFs before the US did. The label on the ticker page has always been looser than the legal documents underneath it.
ETNs are debt, even the well-built ones
Most of Europe cannot wrap a single asset as a normal fund, because UCITS rules require diversification, so the industry routes around the restriction with notes. An ETN is a debt obligation of an issuer that promises to pay you the return of bitcoin. When you buy a fund you own a slice of a pool of assets. When you buy a note you own a promise, and the promise is only as good as the collateral behind it and the balance sheet that made it.
The good modern crypto ETNs, often labeled ETCs on German exchanges, are physically collateralized. The issuer holds actual coins with a custodian, an independent trustee holds a security interest over that collateral for the noteholders, and institutional holders can usually redeem notes for the underlying. Built that way, a note behaves close enough to a fund that the difference rarely shows up in daily trading. But not every note is built that way. Some, particularly older ones, are plain unsecured debt where the issuer hedges the exposure on its own book, and your claim if that issuer fails is the claim of a general creditor. The cautionary tale is Lehman Brothers, which had ETNs outstanding when it collapsed. The indexes those notes tracked kept ticking along, while the notes themselves became claims in a bankruptcy and holders waited years for partial recoveries.
Closed-end trusts and the premium trap
The third structure is the closed-end trust, and the famous specimen is the big US bitcoin trust in its original form, before it converted into an ETF. Shares were created through private placements at net asset value, typically with a lockup of several months, and then traded on the secondary market. The missing piece was redemption. Nobody could hand shares back and receive coins, so the arbitrage loop only worked in one direction, and the share price was free to detach from the value of the bitcoin inside for years at a stretch.
And detach it did, in both directions. For a long period the trust traded at a fat premium, because it was one of the only ways to get bitcoin exposure into a normal brokerage or retirement account, and buyers paid well over coin value for the convenience. A cottage industry of funds ran the obvious trade, minting shares at net asset value, waiting out the lockup, and selling into the premium. Then the premium flipped to a discount, the discount kept widening until it went past 40 percent at the worst point, and several of the leveraged funds running the mint-and-flip trade were destroyed by it. People who thought they held a bitcoin proxy learned they were also short a redemption right they had never priced. The gap only closed when the trust converted to an ETF and redemptions finally existed. I cannot think of a cleaner real-world demonstration of what redemption plumbing is worth.
How I check a wrapped product before buying
My rule of thumb is that the ticker tells you almost nothing and the prospectus tells you almost everything, and ten minutes with the documents covers most of it. The sequence I actually run:
- Find the legal form in the first pages of the prospectus or the key information document. Fund, note, or trust. That one word predicts most of the behavior under stress.
- Check whether the product is physically backed or synthetic. If synthetic, identify the swap counterparty, because your real exposure runs through that firm before it ever touches bitcoin.
- Look for the redemption mechanism. Can authorized participants create and redeem daily, and is it in-kind or cash only? If there is no redemption at all, assume the price can wander a long way from net asset value and size the position accordingly.
- For a note, read the collateral section. You want segregated coins, an independent trustee, and a security interest in favor of holders. If the collateral language is vague or missing, you are lending to the issuer unsecured, whatever the marketing says.
- Check the custodian, then check it across everything you hold. A striking share of the large products use the same few custodians, so three different tickers can quietly collapse into one custody exposure.
- Pull the premium and discount history. A product that has already strayed far from net asset value in calm markets will stray further in a stressed one.
None of this shows up in the fee tables people use to compare these products, which is why so many comparisons treat the wrappers as interchangeable when they are anything but. It also helps to keep the baseline in view, which is holding spot yourself. Self-custody has real failure modes of its own, lost keys being the classic, but it carries no issuer credit, no premium, and no redemption question. That tradeoff is part of why we built trade execution on Blockcircle to be non-custodial, because once you have watched a wrapper trade at a deep discount to the coins inside it, keeping your own keys stops sounding paranoid.
Wrappers are still fine tools. Retirement accounts often cannot hold spot, mandates require listed instruments, and for plenty of people the operational risk of self-custody is a bigger danger than any discount will ever be. Just know which of the three animals you own before a stressed market makes the differences obvious, because by that point switching is expensive, and the answer was sitting in the prospectus the whole time.