Fedimint[1] is a privacy and technical innovation coming to the bitcoin ecosystem.
As with anything, there are trade-offs. With Fedimint, in exchange for privacy and (presumably, eventually) ease-of-use, users will give up the ability to verify the total supply of e-cash tokens which are issued by the mint. Users also give up custody of their sats, at least temporarily.
Mint guardians have a consensus protocol among themselves to regularly balance the books where Total Assets = Total Liabilities[2].
However, there is also the term structure of assets and the term structure of liabilities to content with. If all liabilities are due on demand, and all assets are 30 days out from being liquid, then such an entity is insolvent. An obvious way to reduce that risk is to simply keep all deposited sats in reserve (e.g. the full-reserve model) and this is what the current fedimint developments seem to encode.[2] This is a natural, conservative route to get up and running first.
On the other hand, if a mint has made it clear to its users/members that their deposits are not "demand deposits" but are more akin to savings bonds of the olden days, then this allows such a mint to:
  1. Be profit-seeking while still staying solvent. Users could partake in both the upside and downside.
  2. Privately aggregate and deploy capital for service of its community/members.
In essence, such a mint is more akin to an investment club or, depending on its goals, might take the form of a microfinance charity[3].
What we might see play out is a new era of free banking[4] for better or for worse.
What do we think about this?
The edit window closed for the original post. Here are a couple more thoughts:
Fedimint, as it is currently being built and explained seems to be mostly advocating for many small mints versus few large mints. Of course there is no technical reason why a mint needs to limit its size, but there may be some good social/economic reasons for doing so.
Nevertheless, it does seem reasonable to think that there will be some mints which will be created solely to serve (the federations of) other, perhaps smaller, mints.
reply
I have installed the custodial LN wallet, Wallet Of Satoshi. I use it occasionally as it for certain things (e.g., for bond payments when using RoboSats).
Is WoS fractional reserve? I have no idea. That's why I don't hold any amount of significance on WoS. Even if felt I could trust them (which I'm nearly willing to do as there have never been withdrawal issues that I'm aware of, and no other red flags exist, for example), I know that there are hacks that could happen and/or their wallet could get seized by the state.
The few times I have received any payments of size to WoS I've nearly instantaneously moved those funds off of WoS. That's somewhat fine -- the duration of my exposure to counterparty risk with WoS is measured in minutes.
But can FediMint be used in the same manner? As far as I understand it, the interoperability with LN is thanks to the federation's gateway. But that requires liquidity, so I can imagine there will be times when I need greater liquidity than is immediately available from the gateway. This might be an occurrence that is not all that uncommon but that wouldn't necessarily mean there is increased risk of insolvency / fractional reserve occurring with the mint. Whereas if I were to start seeing withdrawal issues with my custodial wallet (e.g., Wallet of Satoshi), that would be the last time I would consider using it.
But if the federation was not committed to full reserve, I would not use them.
if a mint has made it clear to its users/members that their deposits are not "demand deposits" but are more akin to savings bonds of the olden days, then this allows such a mint to:
That is confusing to me. Is the minted token meant to be interoperable with lightning (i.e., functioning gateaway)? If so, there's almost a guarantee this will fail then when the bitcoin exchange rate rises and those who understand the risk of this being fractional reserve withdraw.
What am I missing?
reply
Thank you for your reply. I mostly agree with what you said. Unfortunately I am not yet comfortable enough with the technical details of fedimint to know exactly how the federation gateway(s) with LN work. I remember reading somewhere in the docs that there can be multiple gateways offered to a federation, some of which may be run by non-federation members, so perhaps this would somewhat mitigate the concern you bring up regading liquidity, but maybe not.
That is confusing to me. Is the minted token meant to be interoperable with lightning...What am I missing?
Thanks for asking for clarification. Maybe it would be easier to think about the following scenario:
  1. a mint starts with the as-currently-being-coded full reserve policy where all sats are redeemable on demand
  2. somehow the guardians/federation come across one or more "investment opportunities."
  3. the opportunites would probably be in the form of debt (e.g. lending fixed amount of sats for a fixed term length and known return). Let us assume the opportunity is for 30 days.
  4. the mint somehow notifies its members/depositors of the opportunity, and lets anybody redeem their on-demand e-cash-sats for some new 30-day-locked-e-cash-sats.
  5. These new 30-day-locked-ecash sats are basically like savings bonds.
  6. Once enough savings bonds have been "sold" to the members, then the mint federation/guardians go ahead and fund the deal (via regardular lightning or on-chain sats).
  7. When the 30 days is over, the deal either returned adequatlely or it did not. If it did not, the mint buys back the savings bond from the bond hodlers at a discount. If it did perform, the mint buys back the savings bond at par value and furnishes the bond hodlers with their portion of the interest, etc.
In this regard, the guardians/federation are acting in a capacity more similar to a fund manager -- filtering deals/opportunities.
As a side note, the savings bonds themselves would, presumably, be e-cash (e.g. transferrable/tradeable) as well, similar to treasury bonds. Savings bonds issued by the same mint with the same par value and same maturity would be fungible with one another.
guarantee this will fail then when the bitcoin exchange rate rises and those who understand the risk of this being fractional reserve withdraw.
Everyone would know that this is how this particular mint operates, so the bitcoin exchange risk is information any would-be participants can factor in. Of course they are also factoring whether the mint guardians/federation are going to be able to source/underwrite deals which payoff. On the other hand, if no deals are sourced, then the status quo would be the usual full-reserve-minting.
What I tried to describe here would somewhat insulate member/depositors who do not want to buy any savings bonds (subject to trust in the guardians/federation, but that trust/risk is there regardless). They just use the mint like normal.
Anyway, there are obviously a lot of details that would need to be worked out to even make something like this work. What I am simply trying to show here is that it might still be possible for the mint to stay solvent, in expectation, by matching the term structures of its assets with those of its liabilties.
It is subtle, and I have probably failed to fully explain it, but it is not quite the same as fractional reserve.
reply
the mint somehow notifies its members/depositors of the opportunity, and lets anybody redeem their on-demand e-cash-sats for some new 30-day-locked-e-cash-sats.
There's no capability within FediMint, as far as I know, to lock minted tokens as you describe.
And I see no benefit to trying to make Fedimint work this way. Why not just ask interested investors to spend their eCash to purchase a share or bond or whatever, which is issued outside of Fedimint? It's the same net result, and the Federation can remain full reserve.
reply
And I see no benefit to trying to make Fedimint work this way. Why not just ask interested investors to spend their eCash to purchase a share or bond or whatever, which is issued outside of Fedimint?
There is increased privacy and convenience for investors, but at the cost of, presumably, reduced returns and additional trust in the guardians to deploy the proceeds from the sale of these savings bonds.
reply
Fedimint has to be the most interesting edge case development for Bitcoin this halving.
As someone who advocates for sovereignty through self-custody of ones keys - I don't understand the resistance from much of the Bitcoin space, seems ignorant to me.
This is not being promoted as a preferable solution to self custody - rather an alternate custody option with far greater improvements over the easily captured centralised 3rd party exchanges where the incentives of those in control do not align with those who use them - which unfortunately is a large portion of the Bitcoin community.
Fact of the matter is, you probably already include a layer of trust (on a personal level at least) in your self-custody/backup solution already.
I could see Fediment's being a useful resource for those who self-custody as a way to move Bitcoin privately, but predominantly as a means to enable safer access to Bitcoin and Bitcoin based fiduciary for the late majority.
reply
Thank you for your thoughtful reply.
I don't understand the resistance from much of the Bitcoin space, seems ignorant to me.
Yes, I guess I can understand and share in the desire to protect noobs from losing their presious sats via a custodial rug pull.
I could see Fediment's being a useful resource for those who self-custody as a way to move Bitcoin privately, but predominantly as a means to enable safer access to Bitcoin and Bitcoin based fiduciary for the late majority.
I think I agree mostly with this.
reply
We support the markdown extension for footnotes. 1
Footnotes
reply
Good to know. Edit window for my post closed though.
reply
I fully expect mints to invest in Curve and adjust their Curve position as necessary. I also expect them to not fully pass the yield to the users.
And why not? Lots of people use Tether that keeps money in treasuries but doesn't pass any yield to the users.
reply
Interesting, so you are saying that there may already exist at least some mechanism (this Curve thing you speak of?) that mints could use to source some yield which then can be partially passed back to their users?
reply
Sure. Use the Ren Bridge to lock Bitcoin into renBTC tokens on, say, Polygon. Deposit it into Curve's Ren pool on Polygon to get about 1% yield (currently) paid out in CRV. Withdraw renBTC and bridge them back when you're done.
There's a bunch of risks with this, the most prominent of which is WBTC token minted by BitGo, but see also the Curve interface hack. Saylor has said that he considers all risks like to be not worth it and therefore he just holds Bitcoin. That said, Fedimint's risk dwarfs all of this risk IMO.
reply
You're gonna lose all your Bitcoin if you use this lmao. Fuck freebanking
reply
Obviously there are risks with fedimint or any custodial wallet/service. However, such risks do not imply that all fedimints will be malevolent.
reply
How is this any better than using an exchange like kraken that has proof-of-reserves?
reply
No more risks than using it with an exchange
reply
Pfft yeah lmao. Not quite sure what the point you're making here is lol.
reply
Why trust people when you have the blockchain
EVERY bank has exit scammed
And in the future once a few of these literally named FEDimints exit scam, regulators will come after them with their own KYC agenda and womp womp now you have to use KYC to use this anonymous "free" bank
This is an unnecessary injection of middlemen in bitcoin that bitcoin was designed to eliminate
reply