When Sats Become The Standart
Core developers design Bitcoin to sustain pressure and censorship. In many places of the world, it works despite various oppressive measures. These measures normally hinder adoption only in so-called casual situations. It is forbidden to import European cheese in Russia but skillful individuals could find gorgonzola with a solid premium on a black market in Moscow. Bitcoin is even easier to hide from the corrupted government. It is not legal in Russia hence it is not wise to put a “Bitcoin accepted here” in front of the store. Like a cheese, people normally hide facts about using Bitcoin. The opposite is happening in any country where Bitcoin has semi-legal status and of course it is happening in El Salvador. This Latin American republic leads in the adoption of the Lightning Network and Bitcoin generally. Every small business in El Salvador becomes a beta-tester of the Bitcoin scaling solutions.
In the previous post, I’ve slightly touched the idea of fractional reserve banking on top of the Lightning Network and calculated an economical effect of the so-called Hosted Channels. In the present post, I will discuss an idea of the Fiat Channels and make a deeper dive into LN native fractional reserve. Fiat channels are obviously useless in the hyperbitcoinized world or for long-term hodlers who may be considered as rich individuals especially when they have access to cheap fiat credit and collateralized loans. For poorer users or small business owners in developing countries, they may deliver real value with similar opportunity costs and risks as if they were using centralized exchanges or lending services.
There was lesser-known local experiment called Bitcoin Beach in El Salvador, and here we will consider some of the identified problems in bootstrapping of Lightning Network in the republic. Among mainstream media, in summer 2021 Bloomberg reported about Bitcoin Beach and the article contained several peculiar observations about the volatility of Bitcoin price. Remember, in May 2021 mining in China was completely outlawed and it might partially cause a twofold Bitcoin price crash.
Hugo Conteras, a shirtless twentysomething, stands on the shore with a long lens, photographing them (surfers). Later, he offers to sell them a series of the best shots for about $20. He tells me surfers sometimes ask if he’ll take Bitcoin. He’s taken it on a few occasions, but the dips in price burned him. “Now I tell them it’s $25 if they want to pay in Bitcoin,” he tells me. “You don’t know when it’s going to go down.” (Bloomberg)
In autumn 2021 before the Adopting Bitcoin conference, the project released a report Bitcoin Banking for Communities: Lessons Learned from Bitcoin Beach. Their recipe for bearing volatility is to educate users.
Help people expect and plan for short term volatility. Set expectations so your community can be prepared for drops in the Bitcoin price. Consider a program that compensates merchants whose account balance drops due to price fluctuations for an initial onboarding period until their comfort with Bitcoin is established.
Preparing people for Bitcoin price volatility is important but we propose a technology that, while addressing the volatility issue of the Bitcoin price specifically, maybe generalized on many types of assets and forms a basis for satoshis-denominated derivatives with instant LN settlement. The history potentially could make a weird twist if Fiat channels will lead to fiat monies as Synthetic Commodity Monies backed by satoshis and issued by various Bitcoin banks. George Selgin in his Bitcoin critics couldn’t anticipate this.
Custodial, Private, Natively Settled
There is no clear use-case of “algorithmic stablecoins” for the real world economy where transactions tend to be small. Numerous examples of Ethereum DeFi algorithmic stablecoins actually couldn’t work with Lightning Network for several reasons besides of free American option argument for monoasset LN. Bob McElrath in On The (in)Stability of Stablecoins argues that the only way stablecoins could efficiently work is via centrally issued tokens. “Classic” stablecoin is inherently custodial and by making them less custodial designers make them less efficient. High relative efficiency practically allows centralized stablecoins to be protocol-agnostic.
Therefore the notion that an algorithmic stablecoin will work with an amount of collateral in the underlying less than the full market cap of the pegged currency is categorically false. If you’re holding the entire market cap in reserve however, this is “fully collateralized,” and no buying/selling is required, rather one only needs to issue new pegged units in concert with acquisition of the underlying. (Bob McElrath)
Fiat and Hosted channels are working with a dedicated Eclair node plugin. It is open-source therefore anybody running Eclair is able to load it along with LN node and provide this kind of service to anyone. Eventually, we may consider an old routing LN node as a firm entity: it has a record of successful operation, well-known peers, and operates for profit. Single entity design avoids coordination overhead which takes place in FediMint project. Wise users can search for information about particular nodes supporting specific features using independent sources. Some of them may be even locally maintained by the community. Hosted channels definitely require trust but generally, low-income individual may be very unfortunate in selecting the firtst peer on LN trustlessly and it may lead to losses although a different kind.
Privacy guarantees of Hosted channels are important. To some extent, they are similar to Chaumian cash with one exception that the Host knows the balances of the Client. The Host does not know who the Client is and has no metadata on Client’s payments because Client’s application does route selection, onion formation, and preimage generation for incoming payments. As long as Fiat and Hosted channels function as open protocols we can imagine new end-user applications specifically designed for privacy use-cases. For example, when they re-create identities and channels after a couple of payments.
Fiat channels protocol is a modification of Hosted channel protocol which supports constant nominal value in fiat currency while being programmatically backed by the corresponding amount of sats. The most advantageous part of the proposal is that payments from/to such fiat channels will be transferred on top of LN network original nodes w/o any limitations. Technically, a client-side app generates standard LN messages / packets and host-side, after settling client-host fiat relations. Then the Host provides routing of payments in the network according to the protocol rules. Constant Fiat channel capacity is provided by the equation
Sats*Rate = const,
where Rate denominates prices of some third assets in sats. Each time the Client and the Host interact about the new channel state they must agree on the additional Rate component of the state, while the balance of the channel in Sats allows client software to natively interact with the rest of the LN network. Permissionlessness and easier bootstrapping (nodes do not need to communicate with anybody else) of Fiat channels is a trade-off with their less reliable security model as opposed to in FediMint proposition.
Node Assets and Liabilities
Hosted Channels and Public Hosted Channels
Let’s consider how HC and PHC work on the top level. Public hosted channels also do not have an onchain transaction and represent trust relations on LN. The difference to private HC is that PHCs are being put into the augmented graph and can be used by SBW Bitcoin wallets to route payments.
By its properties, Hosted channels are monoasset channels and do not involve any price risk. Their sole purpose is in providing fast inbound liquidity for peers. Turns out they may also allow fractional reserve banking on LN. Despite known bad records in XX century, fractional reserve banking may be beneficial for node operators since it will reduce their security risks because Lightning Nodes work are always having their private keys online.
Without special measures, the Host does not work on the fractional reserve. It accepts payments from the outer network and “stores” user liquidity until they need it again for conducting payments. By adding one simple operation for swapping out funds from unbalanced channels the Host may become a fractional reserve LN bank according to the scheme in the Figure.
As history tells us, there were few bankruptcies in the free-banking era in Scotland and Canada and banks were normally operating with as much as 1% or less gold in their branches redeemable for issued banknotes. Given that LN is generally still early and we have situations when major implementations may suddenly become incompatible with a new release, Hosts with a larger userbase may simply have more liquidity online and less in their Cold storage.
Another important difference of Bitcoin if compared to gold is that cold storage funds may be verifable for a given Host. We have no working solutions for LN nodes specifically but at least two exchanges Coinfloor and Bitmex have implementations of Proof-of-reserve protocol.
Fiat channels
Full LN compatibility and native payments require for Fiat channel’s Host to always have liquidity in the normal channels. On the other hand, preserving user’s purchasing power is the purpose of the very existence of Fiat channels. Node liabilities demand an active market position of the Host and may entail some third-party security risks.
The figure illustrates the situation when, if the FC user wants to pay somebody on the network 1 USD, the Host will need to use an additional 1 sat of its own assets to compensate for price increase from 1 sat to 2 sat per 1 USD. It is possible that the Host may somehow provide the opposite market position when the user deposits satoshis into the FC but let’s consider the situation of the responsible Host which does simply provides a synthetic asset. How FC Host can do that? Via hedging.
There is a good article written about hedging by Kollider team “Hedging, Yield and Lightning Network Nodes”. The exchange works on LN and provides instant deposits and withdrawals. This exchange sets an example of the LN Host providing a payment solution and an exchange where the speculating activity takes place. Using LN exchanges may be beneficial for the Host since it may provide liquidity directly to the payment node and the flow may be automated. It may be either Kollider’s perpetual swaps or LN Markets CDP (however, they proxying are tracking Bitmex prices).
An Eclair plugin may calculate net FC position and create a corresponding short position or resize it accordingly. When more merchants are closing their FC and drain funds into normal channels or swapping onto cold storage, additional funds may be withdrawn from the exchange and refill node outbound liquidity. Many centralized exchanges have quite low funding rates because of high volume and market competition. Since these costs must be imposed onto end-users of FC, they will incentivize clients to store only working capital in these channels. We expect that 1X short or 1X long positions (in case the Host accepts stablecoin payments or regular bank transfer some way) shouldn’t be prohibitively expensive for relatively short timeframes from weeks to months.
May LN node be using numerous speculative instruments provided by DeFi? As we mentioned third-party risks, may be beneficial for node operators. But there is no clear path to it. Borrowing rates on DeFi aren’t low and these numbers do not include all required onchain activity LN node would need to conduct to refill its channels.
Conclusion
Fiat channels proposition targets specific pain point of Bitcoin adoption in El Salvador. High volatility may hurt low-income individuals or small business owners in a way they would need to sell their Bitcoin savings. For some of them, Fiat channels may deliver a unique value proposition to fix purchasing power in a short term timeframe and use remaining funds for long-term savings later by draining the Fiat channel or even closing it and re-opening later when they need it.
More broadly Fiat channels actually propose a way to bridge so-called “cryto-assets” into the Lightning network in a form of synthetic assets with instant settlement in satoshis. For many of such assets, the actual settlement in millisatoshis maybe even more convenient than tracking prices in BTC. These “channels” may allow the existence of the LN-native cryptocurrency exchanges in a way proponents of Ethereum DeFi couldn’t dream about.
If you like this article, please consider visiting project page and voting for it. Also, please share it with #BitcoinBankathon hashtag.
Or, you may send satoshis via LN using this LNURL QR code below. If you submit a corresponding message it will be accounted for donation on the Standard Sats project.
