Is custodial bitcoin really worth anything?

13.6k sats \ 65 comments \ @Scoresby 26 Mar bitcoin

People are willing to pay for it

Normies have spent more than $25 billion buying custodial bitcoin vouchers via the ETFs.

Additionally, people have left 1.8 million bitcoin sitting on exchanges. If these people felt that their bitcoin sitting in accounts at exchanges was worth less than the market rate, presumably they would withdraw it. So it would seem they think their bitcoin vouchers are worth a total of $126 billion (using a btc price of $70k).

Based on the actions of the people who buy ETFs and the people who leave bitcoin on exchanges we can conclude that bitcoin vouchers are worth a total of $150 billion.

What is the bitcoin value proposition if you don't have the keys to a utxo?

Clearly, the people who have bought such vouchers are not holding their own keys. What I want to think about is whether they get any value from these bitcoin vouchers or if they are simply buying them for speculation.

Here are Bitcoin's value propositions as I understand them:

Censorship-resistance
Borderlessness
Digitally native money
No inflation
Seizure-resistance
No middlemen

If you don't control a utxo, how many of these value propisitions hold up?

No inflation

And even this one is pretty dubious. There have been enough "paper bitcoin" scandals (FTX being the most recent) that it seems clear that whatever fixed supply guarantees Bitcoin offers are not duplicated by trusting someone else to hold the bitcoin for you.

Is it possible to use bitcoin without using a utxo?

I don't really care about the ETF people. What I'm interested in is the conversation around bitcoin and increasing adoption.

There's lots of seemingly simple math¹ that demonstrates how bitcoin doesn't scale to 8 billion utxo-owning users. You can look at how long it would take to create 8 billion new utxos or how few transactions bitcoin can accommodate (even with lightning) or how inevitably rising fees lead to a higher dust threshold making it harder for poor people to own a utxo.²

In response, smart people have been building things like Fedimint and Cashu and Liquid and Mercury and the list goes on. But all of these solutions involve giving the user something other than the keys to a utxo.

So my question is:

Can you get any value from bitcoin if you don't hold your own keys?

You should check my math, but I actually think you could create ~500 billion utxos per year if you batched transactions: 52,560 blocks per year * 3000 transactions per block * 3223 max outputs per transaction = 508,202,640,000 new utxos per year. ↩
Again, check my math: if you subtract the number of lost bitcoin (~5 million) from the total supply (21 million) and then divide this by world population (~8 billion) you get an average utxo size of 200k satoshis. Is a 200k sat utxo going to be that useful in a high fee (>300 sats/vbyte) environment? ↩

view all related items

303 sats \ 7 replies \ @Murch 26 Mar freebie

Your calculation is off. If all blockspace were used to create outputs, you could make about 1.2 billion new UTXOs per year.

There are only 4 million weight units allowed per block, so to create 9m UTXOs you would need about 400 blocks. https://bitcoin.stackexchange.com/a/115451/5406

17 sats \ 4 replies \ @OT 26 Mar

The link says we can create around 23k UTXO'S per block (P2TR). Has anyone ever done this on mainnet? I think the the most I've seen is around 4k outputs

118 sats \ 0 replies \ @Murch 26 Mar

Note that this figure is for P2TR outputs (because I restricted myself to the same input and output script type). It looks like 26,908 outputs were created in block 826,052.

0 sats \ 2 replies \ @Scoresby OP 26 Mar

This 2018 Coinmetrics article claims the record then was 13,000 outputs, but they don't link to a transaction, so it's not that useful.

527 sats \ 1 reply \ @Murch 26 Mar

The record before 2018 was block 409,069 in which 23,642 outputs had been created.

I found these blocks by sorting blocks per their output counts on blockchair

0 sats \ 0 replies \ @Scoresby OP 26 Mar

This is awesome!

21 sats \ 1 reply \ @Scoresby OP 26 Mar freebie

This is excellent. I think I had even read it before but couldn't find it when writing this.

My error was in figuring out max outputs per transaction and then assuming max transactions per block. Should've caught that.

34 sats \ 0 replies \ @Murch 26 Mar

Yeah, thanks :)

155 sats \ 12 replies \ @DarthCoin 26 Mar

Is it possible to use bitcoin without using a utxo?

Yes totally yes. Many good bitcoiners could run "uncle Jim" scenario for mini-LN banks, for their family and friends. It is nothing wrong with that and we already have many solutions for that. And many will come.

That's why is important to educate now as many users we can to be able to manage correctly those mini-LN banks.

I run myself such a mini-LN bank for some people in my community. And works perfectly fine. And I wrote several guides about how to do it.

110 sats \ 11 replies \ @Scoresby OP 26 Mar

The uncle Jim use case is good and I neglected to mention it.

In the uncle Jim scenario, such people as the lucky friends of DarthCoin get to use LN sats without holding the keys to the utxo. They get most of the bitcoin properties.

Or maybe it is more accurate to say they get the advantages of bitcoin as long as they have a trustworthy uncle. This is good, and I can see it working with ecash, too. So that we transfer many of the advantages of bitcoin to people who aren't using bitcoin directly.

I wonder how it will play out though. With email hosting, it was too annoying or technical for most people to bother. And the uncle Jim's who could have run servers for friends and family didn't get enough benefit for the trouble. End result is you pretty much can't send email from your own host nowadays.

Why do we think bitcoin uncle jimming will turn out differently?

127 sats \ 10 replies \ @DarthCoin 26 Mar

Why do we think bitcoin uncle jimming will turn out differently?

Because those uncle jims will be forced to be "good guys" and do the things properly. They also have incentives to do it. If they run properly those mini-banks, they can earn enough sats from routing payments, or even they can charge fees for the services. LNbits for example, have that option to put a %fee on each tx from its internal accounts.

LND Accounts also have that option. Ecash also could have that option.

And this can be applied not only for families and friends, but also to small companies, running a mini-bank for its employees (for those without "uncle Jims").

Is just about education in the end. More strong educated bitcoiners could run future mini banks. Absolutely nothing wrong with that.

Here people could read those "uncle Jim" guides:

10 sats \ 9 replies \ @Scoresby OP 26 Mar

I wasn't saying that uncle Jim's will do a bad or poor job.

Just that it takes effort and we all have limited time. I think my email hosting example has a lot of flaws, but it makes me wonder if the uncle Jim way will ever be a very large slice of the payments pie.

If you go on vacation and your server crashes, all of a sudden the people relying on you can't make payments.

Obviously they should pay for the service. At the moment, Wallet of Satoshi is better at providing a user experience than you.

On the other hand, I live in the States and cannot use WoS without a VPN, so maybe they don't provide such great user experience.

I think that uncle jimming is only scalable as a business and once it's a business it quickly ceases to be uncle jimming.

17 sats \ 8 replies \ @DarthCoin 26 Mar

Not really have to be a business. You can host very easily your node on a VPS, if you are afraid about redundance. Nowadays Voltage and NODL are doing great job for that.

You can run the core + LN node on a VPS and can have all other apps on your local servers too. Are possible many scenarios. Even this one about running a nodeless LNBits for example.

Greenlight also is coming in force and could offer great options. Mutiny Server also could be a good option. You run as uncle Jim a Mutiny Server on a VPS or at home (it doesn't really matter) and with few resources, you can have each user running his own mobile node.

So many options are out there... people just have to test, improve and utilize them.

33 sats \ 7 replies \ @Scoresby OP 26 Mar

Your optimism does not disappoint.

It seems I have some work to do.

(Today I learned from @ek that I can use NWC with my own LN node to use my own wallet with SN. Then I fell down a rabbit hole full of docker thorns trying to use NWC with Phoenixd cause wouldn't that be a cool way to get to self-custodial LN on stacker news? I didn't manage to get it working, but if I keep banging my head at it, it will stay in my mind and then I'll notice when someone who actually knows what they are doing makes it work.)

54 sats \ 0 replies \ @DarthCoin 26 Mar

I am optimist about LN because I tested so many apps, nodes, solutions and create various scenarios with them, based on what people near me, IRL asked me how to use BTC in x scenario.

All my guides are real life scenarios, that I tested myself before I wrote those guides.

I do not like to talk about things I do not know or control. For example you will never see me into a discussion about writing code. That is my weak point, I know shit about coding.

But I know very well, economics, business, accounting, IT, networks, servers because all my life I was into these, I was the connection between end users and those building hardware and software. Usually developers don't bother too much listening what end users want or do not understand all the processes into a business. But my experience of many years in this field helped me see very clearly where each application can fit.

0 sats \ 2 replies \ @DarthCoin 26 Mar

Phoenixd is very good. The only thing I do not like at it is that works only with ACINQ node, all channels are with the same LSP.

43 sats \ 1 reply \ @Scoresby OP 26 Mar

Im interested in it for the easiest possible way to run self-custody LN. If it gets to the point where you run a command or two and click connect in a few apps and you have a self-custody LN wallet, that's something I can convince friends and merchants to use.

view replies

0 sats \ 2 replies \ @ek 26 Mar freebie

What is "Phoenix drive"? Do you mean Phoenix Server?

34 sats \ 3 replies \ @HardMoney 26 Mar

ETFs improve 10x when the custodial model is multi-institutional multisig.

Obv still less advantageous than real bitcoin but there are certain benefits that make holding bitcoin an option for institutions and HNW people unable to manage keys.

Nightmare scenario is Coinbase fucking up and losing access to keys or getting hacked. Multisig negates this honey pot immensely.

It’s inevitable that fiat rails get connected into bitcoin via instruments like ETFs. A way better top of funnel entry vs shitcoin casinos like Coinbase.

Hopefully many etf investors learn more overtime and decide to buy real bitcoin.

0 sats \ 2 replies \ @Scoresby OP 26 Mar

Why do you think it is way better than shitcoin casinos?

I don't think etfs will ever allow you to exit to a utxo. So at best you have a legal claim on the bitcoin that the fund owns. Probably though all you have is a "share" of the fund which is worth in dollars whatever proportional amount it represents of the fund.

I accept your point about institutions and hnw people.

27 sats \ 1 reply \ @HardMoney 26 Mar

Better than shitcoin casinos because Coinbase basically incentivizes anyone new to start trading shitcoins.

If Coinbase is the first interaction a new coiner has they will likely “diversify” into shitcoins at some point.

If an etf is the first interaction a new coiner will start with it as 1-5% of their portfolio, see it grow as a % overtime. And may decide to learn more about bitcoin via podcasts or books.

ETF investors won’t rotate into shitcoins as easily as a first time Coinbase user

0 sats \ 0 replies \ @Scoresby OP 26 Mar

Fair point. I hadn't considered that.

34 sats \ 2 replies \ @grayruby 26 Mar

It shouldn't be worthless but it should be worth less than bitcoin you control the keys to.

50 sats \ 1 reply \ @Scoresby OP 26 Mar

Right. To some degree all of us who hold our own keys believe this is true.

17 sats \ 0 replies \ @grayruby 26 Mar

I want to liken it to the fact that there always has been a premium to buy physical gold over paper gold. My intellectual bias says that makes sense because physical is better than a paper promise but it probably has to do more with the transport, storage, retailing costs that put a premium on it. But people buy it so I suppose they do value it more or they would buy the paper promises.

34 sats \ 6 replies \ @StillStackinAfterAllTheseYears 26 Mar

The simple answer is that it's worth something until it's not. So the sats I've got in a custodial lightning wallet that I exchange for a steam key on Joltfun are worth something because I was able to use the sats in exchange for a good¹.

Similarly, if I've got BTC on an exchange and send them to someone's wallet in exchange for a good, that's still worth. As would be selling the BTC on the exchange for fiat (stupid, but there's an explicit value of sorts).

Of course, these both rely on trusting other entities, and as you note, that "worth" can go away at any time. And if you modified your question to "long-term worth," I think the answer becomes closer to a definitive "no."

I do feel that ETFs and Exchange BTCs are different beasts, though. Folks using ETFs are simply treating Bitcoin like any other investment like a stock or bond, and only care about how much it's worth in fiat. Folks with BTC on the exchanges might be doing this, but might also think they'll take the BTC into a wallet some day, or exchange it for something other than fiat.

Yes there's a legit secondary question about Steam keys being goods we "own." ↩

21 sats \ 5 replies \ @Scoresby OP 26 Mar

Good point about the difference between exchange BTC and ETF vouchers.

Let's think about sats on SN. None of us have our keys here. But we can withdraw at any time and so until that changes, we treat SN sats like real sats.

When using external wallets on SN gets built (note my effortless confidence in the SN devs) we will all be using Lightning and so no problem.

But what if external wallets turns out to be more trouble (expensive) than it's worth for the SN use case?

What if an ecash mint makes more sense? In that case, would it be different than using a database (which is what I assume SN does now)? That difference us what curious about.

44 sats \ 4 replies \ @ek 26 Mar

When using external wallets on SN gets built

You can already attach external wallets via NWC. The UX does not match the custodial UX as much as we'd like yet (it's one of our priorities right now) but it's something that's already possible.

Not sure if you already knew, your comment didn't sound like it.

0 sats \ 3 replies \ @Scoresby OP 26 Mar

Haven't used nwc much. Does this mean balance gets held on the external wallet and zaps come from there?

61 sats \ 2 replies \ @ek 26 Mar

Yes, I use this from @benthecarman

10 sats \ 1 reply \ @Scoresby OP 26 Mar

That's pretty cool! I thought we were still just at auto withdraw above a threshold and that's it. Man, you guys are awesome!

27 sats \ 0 replies \ @ek 26 Mar

NWC was released last month

Man, you guys are awesome!

Thanks :)

34 sats \ 2 replies \ @030a29f333 26 Mar

NGU is a strong value prop, even if custodial. 99% of people aren't concerned about being rugged of their funds. How much money is in banks today? Brokerage accounts? Property? How many people continue to keep their sats on exchanges despite the numerous blowups? That value all has custodial risk and yet here we are.

0 sats \ 1 reply \ @Scoresby OP 26 Mar

You are completely right. Outside the context if bitcoin custody is so normalized that most people don't notice it. And following from that, things like censorship resistance and seizure resistance don't even appear to be needed. It's really amazing how bitcoin changes how you think.

But I'm curious what you think the value prop of things like ecash and liquid are. Does any of the usefulness of bitcoin extend to things that claim to peg to it?

34 sats \ 0 replies \ @030a29f333 26 Mar

I'm quite bullish on ecash overall:

-Cashu is a strict improvement on custodial lightning (massive privacy boost, no INDIVIDUAL censorship/rug risk, can selectively choose mints to trust), which is the preferred way for people to interact with LN today.

-Fedimint is quite interesting for communities. The "proof of punch" model where you know who to punch if funds are tampered with will be fascinating to watch.

We'll have to see how the value of custodial sats evolves over time. Rugs will keep happening. Each rug brings more appreciation and awareness for those properties of self custodial bitcoin you mentioned. Ultimately what matters most, IMO, is that we continue to minimize trust where possible and have ever-improving fallback options to transact freely when needed.

Lyn Alden described this recently on WBD. Im not gonna walk around my neighborhood with a bulletproof vest because I trust the people around me. If I'm going to a wartorn area, I'll put it on. If you experience frictions using a cheap and easy system, you have the freedom tech tools necessary to continue. Everything is about tradeoffs.

34 sats \ 2 replies \ @kr 26 Mar

yes, so far the market has told us that custodial bitcoin is worth just as much as self-custodial bitcoin

this may change in the future, but i have no reason to believe it is not true today

0 sats \ 1 reply \ @Scoresby OP 26 Mar

But is it as useful?

17 sats \ 0 replies \ @kr 26 Mar

if your use case is storing value, i’d say yes. if your use case is exchanging goods, i’d say no.

34 sats \ 3 replies \ @0xbitcoiner 26 Mar

The only value I see is the potential capital gains from a future sale. But this is not taking advantage of the true benefits of Bitcoin.

32 sats \ 2 replies \ @Scoresby OP 26 Mar freebie

Right, so it's just speculation.

But if we talk about fedimint (also a situation where you don't have the keys to a utxo) are there any advantages of bitcoin that remain?

Ecash is still digitally native, so there is that. If ecash could prove it's backing or demonstrate strong inflation protections, maybe it could retain the no inflation advantage.

And maybe the censorship resistance can be retained through the privacy guarantees of ecash.

34 sats \ 1 reply \ @0xbitcoiner 26 Mar

I just remembered that ETFs can be used in retirement savings plans, and I see an advantage in this application. As for fedimint and ecach, I'm not yet fully convinced, as we have to trust the issuing entity. I'm not sure about ecash, but with fedimint it's not possible to verify the token supply.

32 sats \ 0 replies \ @Scoresby OP 26 Mar

Retirement is an interesting point. However, there aren't any guarantees that IRA and 401k rules will stay the same. I guess that may come down to how far off retirement is.

With fedimint and cashu I think people talk about a system of automated withdrawals (automated bank runs) which are much easier to achieve than real bank runs exerting a pressure on mints to stay honest. But it's not as certain as bitcoin's halving schedule, that's for sure.

17 sats \ 0 replies \ @PlebeiusG 26 Mar

Sword of Damocles comes to mind.

"...nothing is happy for him over whom terror always looms?"

To elaborate, your world (read: wealth) could come crashing down at any moment with custodial bitcoin.

Be unruggable.

17 sats \ 0 replies \ @BitcoinIsTheFuture 26 Mar

Pretty much a paper promise

17 sats \ 0 replies \ @riberet19 26 Mar

for me Bitcoin doesn't make any sense if I don't self-custody, what is the difference between having my money in the bank or having Bitcoin on an exchange, and let's not talk about ETFs...

17 sats \ 6 replies \ @Coinsreporter 26 Mar

It was just 2 months ago when I used to have custodial wallets, thanks to SN, I have removed any custodial wallets straight away except one just for emergency funds. Getting value ? Depends, if that custodial doesnt go bankrupt and remain working, no hacks or mishandling of funds, the list is long. So value is another proposition, first we should think about saving the theft of our sats. Self custody makes it easier and only to one condition, just you can't lose your keys.

11 sats \ 5 replies \ @Scoresby OP 26 Mar

But do you think any of the things that bitcoin is good at are available to someone who owns custodial bitcoin?

17 sats \ 4 replies \ @Coinsreporter 26 Mar

If they are holding it on a CEx, it would be worth of a few of them. But holding them as ETFs or ETNs don't worth any of the inherent values of Bitcoin at all.

0 sats \ 3 replies \ @Scoresby OP 26 Mar

Why do you feel bitcoin at an exchange retains some of bitcoin's value props? Also which ones?

34 sats \ 2 replies \ @Coinsreporter 26 Mar

Borderlessness Digitally native money No inflation

I think for these. But, yeah I don't believe that anyo e must be hodling there. CEXs are for trading only.

11 sats \ 1 reply \ @Scoresby OP 26 Mar

Digitally native is a good point.

The other two: I am not so sure about.

34 sats \ 0 replies \ @Coinsreporter 26 Mar

I must admit, I'm not well versed to answer this but for the sake of trying, here's my reply-

Borderlessness because, we can send or recieve payments there as well.

No inflation because the value is written same on custodial or non custodial wallet.

17 sats \ 2 replies \ @draoi 26 Mar

Custodial bitcoin is arguably inflationary because it enables fractional reserves.

0 sats \ 1 reply \ @Scoresby OP 26 Mar

Yeah, so then what is left? Why is custodial bitcoin worth anything?

34 sats \ 0 replies \ @south_korea_ln 26 Mar

Twitter cloud.

17 sats \ 0 replies \ @bzzzt 26 Mar freebie

only, and only if you can exit for something with value. True for all fiat products.

20 sats \ 0 replies \ @justin_shocknet 26 Mar

Custodied coin has one killer feature that's a key milestone on Bitcoin's path to world reserve status... Collateral.

Collateral kicks the speculative attack into high gear, its how normies can load up to the tune of billions or trillions without any impact to their fiat lifestyle, by simply leveraging margin.

As long as fiat gains outpace your living expenses, you literally never need to sell ... just borrow against the gains tax free.

If you think you'll need to sell any coin to live over the next 4-10 years, you should do it now and put it in the ETF in a margin enabled brokerage account. Doing this effectively ensures your gains from here on out are tax free since you'll never need to realize them and just borrow fiat as needed.

0 sats \ 0 replies \ @02bcd3eeb0 27 Mar

You're missing the part where states and institutions fork the chain and you end up with whitelist coins and blacklist coins.

17 sats \ 3 replies \ @Darthcoin_Cash 26 Mar

Buying ETF bitcoin is a sure sign the person buying it is a complete moron. Like the people who took the "vaccine". Zero capability to think for themselves. Would be interesting to have a correlation between vaccinations and boosters and willingness to buy a bitcoin ETF :)

People are willing to pay for it

What is the bitcoin value proposition if you don't have the keys to a utxo?

Is it possible to use bitcoin without using a utxo?

Can you get any value from bitcoin if you don't hold your own keys?

Footnotes

Footnotes