Credit risks in Bitcoin
Bitcoin is trading at 2020 prices again, and credit risk has everything to do with it. Everyone and their dog has written about the details regarding Luna Terra, Celsius, and Three Arrows Capital, so I will not bore you with the details. However, it is important to note that the “cryptocurrency” industry tends to go through these cyclical boom-and-bust cycles every few years, due to the buildup of leverage and credit in the system.
As is apparent today, bad credit can lead to industry-wide contraction. Even if your defi protocol is over-collateralized, bad credit that exists externally can cause cascading liquidations that threaten even the more cautious investors. This is a phenomenon that appears to be impossible to prevent, and perhaps should even be embraced as a natural process inherent to free markets.
Indeed, Bitcoin was created in part as a response to fiat bailing out irresponsible creditors, as evidenced by the message embedded in the genesis block. But sure, while there is no ‘Bitcoin Central Bank’ to bail out bad debt, you cannot prevent the creation of debt in the first place, as debt is also an uncensorable peer-to-peer concept.
It is understandable that many in the Bitcoin community are reluctant to accept the existence of credit in a Bitcoin economy, but it is inevitable. And I would like to add that credit itself also has many benefits - for example, if properly managed and allocated, it can accelerate good businesses or eradicate bad ones. Thus, the focus should not be on preventing credit, but on mitigating systemic risks.
Credit and Lightning
Recently Alex Bosworth commented on the zero-conf channels feature bit to be implemented soon in Lightning. Zero-conf channels allow consenting peers to consider a channel ‘open’ before the transaction has confirmed a sufficient number of times on-chain, and for other nodes to route transactions over such credit-based channels.
An interesting aspect to the upcoming LN protocol change to support channels without block confirmations is that you could start spending sats on LN that you don't really have from the perspective of the Blockchain
— Alex Bosworth (@alexbosworth) June 20, 2022
This could drive a UX where you onboard instantly but on credit
For what it’s worth, I don’t agree with all of Alex’s opinions, but I was taken aback by the amount of pushback he received for welcoming credit on Lightning. The majority of replies seem to be wary of systemic risk; however, in the case of zero-conf channels, the only parties at risk are the nodes who share this channel. (i.e., you are not a direct counterpart of zero-conf channels unless you explicitly consent to it, and your funds in traditional channels will not be affected by failure of zero-conf channels.)
There is massive room for credit in the Lightning Network, due to its P2P nature being very complementary to credit, and I believe this is one of the directions Lightning will take to provide better UX, more functionality, and cheaper fees.
In principle, a good question to ask when evaluating projects that introduce credit might be: What is the worst that can happen? In Lightning channels, the worst that can happen is theft of channel funds or more generally closure in incorrect state, but since your Lightning channel is a fully-collateralized affair between just you and your counterpart, these issues themselves do not have contagious effects (with the exception offorced closures due to stuck HTLCs going on-chain, but this is an LN-wide problem regardless of introducing credit or other features).
As financial use cases of Lightning are explored, there will inevitably be products that introduce some systemic risk to other such products, but I struggle to think of examples where regular lightning channels go bust for this reason. Perhaps a rush for liquidity will cause channel closures en masse? I’d love to hear any examples where the introduction of credit and a subsequent stress event can cause issues with Lightning’s ability to facilitate payments.