Trust Me, Bro

allen farrington
14 min readJun 16, 2023


Trusts, ETFs, and Trojan Horses

photo from

A precocious child once asked me, “Allen, do you ever regret cornering the niche market for bitcoin/crypto twitter shitposting about the exact legal, financial, and ethical nature of “securities” in an effort to simultaneously shill Only The Strong Survive and act as possibly the only public Liquid bull in existence? You could have outsourced it all piecemeal to Preston Byrne, Adam Back, and Lisa Neigut. Have you even made a net positive contribution? Are you happy with your decisions? Are you happy with your life?

Yes,” I said, dismounting my stallion, “and besides, if you don’t believe me or don’t get it, I don’t have time to try to convince you, sorry.

That child grew up to be Satoshi Nakamoto, and on June 15th 2023, BlackRock filed with the SEC to launch a Bitcoin Shmoogleblop.

Pull up a chair and I’ll tell you a tale.

Securities Law, Baby

photo by Dylan Calluy, via Unsplash

What seems to have confused most commenters on the topic is that a lot of large, established, reputable, and uncontroversial ETFs are structured as trusts in the same way as is proposed for the BlackRock Shmooplebob, such as GLD, the ~$60bn SPDR Gold MiniShares Trust, to give it its proper title. What gives? Are these not mutually exclusive concepts after all? Can ETFs be trusts, and vice versa?

The super-duper pedantic answer is that, no, a “trust” and a “fund” are distinct. However, GLD is a perfect example of a trust that is exchange-traded, and, in fact, the mechanics of which are very, very similar to that for “real” funds (“40 Act” Exchange Traded products, to be introduced just below). Hence the idea that “ETFs can be trusts” is true enough in practice to be worth explaining rather than dismissing on the basis of definitions.

In the US, typically any collective investment vehicle will be considered to be subject to the requirements of the Investment Company Act of 1940 (“40 Act” hereafter) unless the managers can credibly make the case that it is specifically exempt. I won’t go through all such permutations of rationale, but there are essentially four good reasons that many investment vehicles are able to opt for the Securities Act of 1933 instead (“33 Act,” ditto). Keep in mind while pondering all this that 40 Act vehicles are much more heavily regulated than 33 Acts and so if you can find a plausible rationale to go for the latter, you do.

1) Tax: trusts are highly disadvantaged on capital flows. It’s more complex than this, but, in short, if there is income on the assets held in trust (dividends, interest, lease payments, liquidations, something more exotic, potentially) that creates an extremely inconvenient tax liability for the beneficiaries, as it is not taxed as corporate income like you might expect of an “investment vehicle,” and is the case for a fund/40 Act.

2) Issuance & Redemption: in a fund/40 Act, that is also open-ended as opposed to closed-ended, the manager is legally compelled to issue and redeem shares at Net Asset Value (“NAV” — the aggregate value the underlying assets are selling at individually) — not to the public, to be clear, but to an intermediary called an “Authorized Participant,” on which I will elaborate shortly. In a trust/33 Act, the manager is not so compelled (n.b. technically this role is called a “sponsor” in a trust, but I will keep using “manager” as it means the same thing).

oh shit, I actually read it.

3) Diversification: 40 Acts/funds pretty much assume diversification by virtue of their purpose, whereas 33 Acts/trusts tend not to allow it (again, more complex than this, but this is the gist).

4) Ownership: in a fund/40 Act, holders own shares in the company and are collectively engaged in an Investment Management Agreement with the manager. In a trust/33 Act, the holder owns the underlying assets, and the manager is the trustee.

So far, so incredibly arcane.

So why is GLD a trust?

Because i) it only holds one asset with no yield, so there are no tax concerns; ii) it is impossible to commit to continuous issuance and redemption because the underlying is not a security that can be purchased on the open market. Delivery in both directions must be physical, which takes actual time, money, and logistics, and has to be at the manager’s discretion in order for the vehicle to be mechanically viable in the first place; and, iii) it is definitionally not at all diversified (Point 4 above is not yet relevant but we will come back to it).

These details handily explain further why many commodities “ETFs” are technically trusts, not just GLD. If the manager is comfortable enough handling the issuance and redemption process (which is likely a function of i) the exchange qualities of the underlying commodity and, ii) the size of the vehicle, hence its operating budget) they can make a decent effort to issue and redeem as if it were an ETF. So technically they are not, but saying they are is often true enough in practice. A final thorn here is that the word “fund” is often colloquially used to mean something like “a generic investment vehicle,” inviting the assumption that it can be appended with “exchange-traded” as an adjective, even though the full expression “exchange-traded fund” has the more technical meaning just covered.

I will grant that none of this is really a big deal unless you are providing exactly this kind of infuriatingly pedantic analysis.

I also want to quickly clarify that I couldn’t care less about GBTC. Yes, it sucks. Yes, it’s a trust. Is this better? Maybe, but “not GBTC” is not synonymous with “good.”

So what about the BlackRock Shmurdlebloop? How pedantic do we need to be? Well, anon, how much do you care?

Do You Care?

Whether or not you care probably depends on your perspective. A potential holder of shares of this instrument probably could not care less. Ditto for the surrounding infrastructure in tradfi. With GLD as the obvious comparison once again, most holders have likely literally never once contemplated the exact legal structure of the issuing entity, or why that is better, worse, or different in any way from any other:

But let me divert your attention back to point 2 above: the managers of this 33 Act vehicle do not have to redeem. How about now, anon? Shitting yourself? No? What about point 4? Do you care if you own shares or the underlying assets? ANSWER ME, ANON! ARE YOU PAYING ATTENTION YET?

Let us return to this weird detail of “Authorized Participant.” What’s that all about?

Members of the public do not buy ETFs or ETF-like trusts directly from the issuer. A group of intermediaries called Authorized Participants has the right to both create/redeem shares in the vehicle by buying/selling the underlying assets on the open market and either handing them over to the issuer for shares or giving shares in exchange for the underlying assets, to then turn around and sell. Everybody else buys in the secondary market from Authorized Participants or each other.

The incentives that keep all this spinning are that if the secondary market price deviates much at all from the NAV, the Authorized Participants can make an arbitrage profit from the above arrangement. Hence their involvement ought to keep the secondary market price close to NAV, which is useful for price discovery — or more like, for avoiding weird, inaccurate price discovery emerging due to the circumstances of participants as opposed to anything related to the underlying assets.

So to return to point 2 above: in a 40 Act, the manager is compelled to issue and redeem, per the terms of their SEC regulation. In a 33 Act, they are not. This might seem bizarre on its own, but if we bring in point 4 and the reminder that 40 Acts are more heavily regulated than 33 Acts, it will all start to come together:

In a 40 Act fund, the manager is bound by their IMA. This has more complex relevance if the fund is actively managed, but for a passive fund (i.e. an “index”, like everything discussed so far) the mandate and the constraints of the IMA are very simple: mimic the price of the underlying, gross of fees. If the manager neglects to correct a massive discount to NAV, they are in violation. And, recall, the Authorized Participants are strongly incentivized to do this anyway, meaning there isn’t really even any effort required from the manager. They just need to not be incompetent or recalcitrant.

So, to be extra clear: the owner of a passive fund unit is contractually engaged with the fund manager to be provided with an issuance and redemption opportunity via the Authorized Participants that tracks the value of the underlying assets. That is what ownership of this asset means.

On the other hand, the owner of a passive trust owns the underlying assets. They literally own them directly. Hence it doesn’t matter if the manager is being a dick and won’t cooperate with the Authorized Participants because the owner is legally unaffected. They have no right to value, they have a right to assets, and the manager and trust sitting around doing nothing affect their rights not one bit.

BUT! They do not have the right to physical redemption. Only Authorized Participants do, and even then only at the discretion of the manager. Even before getting to why this is just really, truly, awful in our case, it’s worth clarifying that even if this were a real ETF, regular holders would not be able to redeem for Bitcoin. That’s just not how this works, even with regular ETFs and regular commodities.

Now for most (if not literally all, ever) trusts, refusing to redeem would still be extremely stupid behavior. The manager makes money from a management fee, meaning the more assets they manage, the more money they make. If ever even attempted, the kind of shenanigans mentioned above would immediately become publicly known, the manager’s reputation would be ruined, and no new assets would ever flow in. So, what am I so riled up about?

This Time Is Different?

photo by Adrik Langfield, via Unsplash

Look, maybe BlackRock are all perfectly nice guys and just want to support Bitcoin adoption, starting domestically but eventually rolling out to The Global South? Maybe we orange-pilled them? Maybe it was Bitcoin Is Venice that got them over the line?

Maybe I’m just paranoid? This isn’t entirely sarcastic; BlackRock is an asset manager after all, and Bitcoin is a valuable asset. But even this doesn’t provide much comfort given I would argue BlackRock’s entire purpose is to poorly solve a problem Bitcoin solves well. Even if more bumbling than conspiratorial, Bitcoin “succeeding” is not in BlackRock’s interest.

So, even if only for the sake of argument, I’m going to outline exactly what I worry BlackRock is going to try to do with this, and you can just thank me in 10 years when it fails due to what I suggest after to counter it.

First, if this is approved, you can guarantee it will attract a metric fuck ton (finance jargon) of inflows. That will clearly pump the Bitcoin price in both the short and medium term. That’s why all the high time preference Bitcoiners are happy. NgU bro! Trust me bro!

Second, BlackRock will be Coinbase’s biggest customer by some orders of magnitude such that Coinbase will become a subsidiary, for all intents and purposes. No alternative “ETF” (real or otherwise) will be able to use Coinbase. They’ll have to use Binance LOLOLOLOL — jokes, they just won’t be able to launch at all. Obviously, this will loop back to point 1 above.

Third, expect an onslaught of FUD in tradfi (not on Twitter, sadly) on “tainted coins” so as to make even the idea of a physically redeemable spot Bitcoin whatever completely beyond the pale. None of the “reasoning” will make any fucking sense whatsoever, but that also won’t matter. Remember “Blockchain not Bitcoin”? Remember “Strawberries on the Blockchain”?

Yeah baby, it’s highly regarded season again. BlackRock’s novel inclusion of market surveillance will serve as an easy wedge for the SEC to once again turn against everybody else:

Obviously, this will all loop back to points 1 and 2 above.

Fourth, expect the SEC to LAP. THIS. UP. UTXOs in the BlackRock Stack O’Sats™ will be definitionally “clean.” Coins not yet in it (“precoins,” if you will) will be suspect, but capable of salvation. Coins that have ever left this safe space (“nocoins”) will be blacklisted. Why are you trying to move your coins, anon? What are you, a money launderer or something? What’s wrong with our IOUs? Again, this reinforces everything else.

Speaking of which, fifth, expect BlackRock to turn around and shill this to banks, and in particular big banks, such that JPMorgan, for example, can itself turn around and offer “Bitcoin” to its own clients. Of course, it will be an IOU to the BlackRock Stack O’Sats™ that you can transfer to your friend’s Bank Of America Bitcoin IOU account at your leisure. In other words, in short order, the entirety of the Blue Banks’ asset base becomes a funnel into the matrix of the pseudo-Bitcoin narrative attack vector (2h 42m 5s if the timestamp doesn’t work):

Sixth, on the back of all the previous points, expect BlackRock and Coinbase to dominate flows and hence effectively control pricing.

Seventh, in a move that makes me question its conspiratorial chops after all, BlackRock has decided to openly admit that it might have to / want to / get to pull a NYA 2.0:

Before you laugh, and even keeping in mind that this kind of “hard fork risk” language appears to be standard in similar filings, keep in mind the impact of all previous points on points one and six in particular: a metric fuck ton of Bitcoin, with control over how that Bitcoin is priced, can create a metric fuck ton of malicious economic incentives, especially with “no guarantee that the Sponsor will choose the digital asset that is ultimately the most valuable fork.” Success may not be guaranteed, but some amount of fuckery seems to be.

Eighth (and probably not even “finally” but I’ve listed a lot and I want to move on) even assuming some brave soul manages to get through this wave of propagandistic bullshit, BlackRock simply won’t redeem BECAUSE SECURITIES LAW (see above, sections everything through everything else). They will buy back in the secondary market anybody trying to redeem and their Stack O’Sats™ will grow and grow until a NYA 2.0 seems worth a shot.

I’m sorry to break it to you, but this last point really is key. If it were a real ETF, refusing to redeem would trigger lawsuits immediately. Hence if any of points one through seven were suspected to be coming into force in a way that would be detrimental to shareholders, they would sell down (AP mechanics re: NAV and all) and go find a way to buy real Bitcoin, or at least something better than this, meaning BlackRock would have to sell its Stack O’Sats™ to meet redemptions. But it is clear from this setup — BECAUSE IT IS NOT A FUND — that BlackRock can just say no.

Which Way, Anon?

photo by Pexels, via Pixabay

I first want to preempt a criticism of this entire post and my Twitter shenanigans on Black Thursday: that the ETF-or-not issue is a MacGuffin. And actually, I think that’s right. It is. That’s not what either this post or my frustration is really about.

I am frustrated because, if you know the difference, if you have some understanding of securities law, and if you see that BlackRock is filing for a 33 Act Trust that everybody is calling a “spot ETF”, you should not be celebrating; you should be suspicious! MacGuffins aren’t non-sequiturs. They at least get the plot going, and you are smack dab in the middle of the plot now, anon, believe you me …

Anyway, let’s try to end on a positive note. One thing to immediately point out is that Bitcoin Doesn’t Care. All this garbage will fail on a long enough time horizon. As has also recently been picked up on, the US is rapidly losing not only financial prestige but even the financial power to prevent (natively digital) innovation from simply moving elsewhere:

But as ma boi Keynes knows, in the long run we are all dead, so it is still worth pointing out that a lot of Bitcoiners are going to care if BlackRock gets away with it.

It’s also worth pointing out that “self-custody” is not an answer. I’m sorry if that also triggers you, but the idea that self-custody is viable for literally anybody with any fiduciary responsibility (financial jargon … no, really) at all, never mind for the likes of a bank, a mutual fund, or a public company, for example, is laughable. All it says is that you have no experience whatsoever of financial services or really any kind of commerce at a scale beyond friends and family.

Obi Nwosu has a great anecdote to illustrate this at an even simpler and more personal level than is required to appreciate the much more complex environment of corporate finance (5m 10s if the timestamp doesn’t work):

If we want “the institutions” to “come to Bitcoin” (and hey, maybe we don’t?) we need better custody solutions, and better legal and audit schemes around these solutions, ideally utilizing cryptographic attestations rather than … well … trust.

This, of course, is the gap in the market that BlackRock is ostensibly trying to fill here, in the nonparanoid reading. If the bajillions of likely clients could just go to Coinbase (or Coinkite 🤪) they would. But they can’t, so they are waiting for an ETF, real or otherwise.

If I really am just paranoid, then this will all amount to a stepping stone on the way to better and better custody practices. Or if my paranoia is spot-on but people wake up and outflank them, we can create and promote these practices before it’s too late.

And so, apart from whining about securities law and BlackRock to no actionable end, I think the real “lesson” of this piece is the following:

The opposite of good third-party custody is not first-party custody, it is no custody at all, hence BlackRock.


Don’t get mad, get even.

Obi telling the story above is no coincidence given Fedi potentially provides one such technical solution, or at least a tool with which to build appropriate solutions. OP_VAULT is likely another, on a longer time horizon.

Unchained’s collaborative custody model is clearly a foundational commercial alternative to BlackRock, hence the likes of what OnRamp and AnchorWatch are bringing to market on the back of Unchained’s work will likely be key too. Deciding to trust others at all likewise necessitates viable Proof Of Reserves and Proof Of Liabilities schemes. There will be many more innovations to this end that I haven’t thought of, am not aware of, or that have not yet been invented.

Technicalities aside, what we are looking for is a path to third-party custody that enables as wide a base of asset ownership as possible, but prevents what I think BlackRock is up to: namely, a black hole out of which sats will never return.

WERE THIS A REAL ETF, there would at least be a legal means of preventing this, and that would certainly be a good start, but technical means are even better. I would even go as far to say that over the course of writing this post, I have come to believe that non-first-party custody with the technical option of redemption onchain may be the holy grail of mass adoption — that is, defined as widespread sovereignty over UTXOs, not the price of whitelisted paper Bitcoin.

What will not lead to mass adoption is bullshit pseudo-ETFs that pave the way for Bitcoin Uncle Sam’s Vision. And what will actively harm mass adoption is financially illiterate Bitcoiners cheering on the SEC and simping for BlackRock.

When Larry Fink drops a Bitcoin Shmoodleplonk and says, “trust me, bro,” don’t trust.


Thanks to Steven Lubka and Nic Carter for edits and contributions.

follow me on nostr: #npub1sfhflz2msx45rfzjyf5tyj0x35pv4qtq3hh4v2jf8nhrtl79cavsl2ymqt

and twitter @allenf32



allen farrington

I’m an investor. I think about things. I write some of it down.