Episode 8: Living on Borrowed DAI
Leverage, markets, and the double spend problem no one warned you about
One of the key innovations of bitcoin is that it solves something called the double spend problem. This means that every bitcoin is provably scarce — it can’t exist in two peoples’ wallets at the same time. This is a big deal because it’s actually pretty hard to make sure digital files don’t get reproduced and propogated. Just think of what Napster and Limewire did with digital music files or what Pirate Bay and Popcorn Time did with movies.
But just because cryptocurrency can’t be “double spent” on a technical level doesn’t mean it can’t be in two places at once — used to create leverage.
In this episode we dive into consumer credit. We run through a brief history of lending, credit’s components, and innovation in both legacy and crypto credit markets.
Cryptocurrency markets have recently seen products like Maker DAO, Compound, Dharma, dYdX, Marble and others develop around lending. What are the implications? What are the problems here that remained unsolved?
Consumer credit is big business, with roughly 4 trillion in debt currently outstanding in the United States. And now, with cryptocurrency projects getting into the game, making gains off of other peoples’ leverage isn’t just for bankers anymore.
But we’ve got a few big outstanding questions for credit in crypto.
- The first question is that of information sharing. Loans very often depend on some shared information between lender and borrower (a credit score). Are there ways to manage and share credit scoring information without making use of a centralized third party (or third parties)? We are accustomed to trusted third parties like Equifax (lol) or Standard and Poors (lol again) managing this. Can this be done in a decentralized manner?
- The second question is around recourse. Generally lenders and borrowers know each other and have terms surrounding their agreement. If the borrower fails to pay back the lender, the lender has some form of recourse. In the absence of these relationships (legal or otherwise) in decentralized (sometimes pseudonymous!) crypto lending, all loans must be over collateralized. Can we break past this?
- Our third and final question is on risk management. Given how nascent these markets are, we have yet to develop risk models with any sophistication. For a really rich example of what we think about here, give Dan Elitzer’s post (cited just below) a read. What happens when collateral that is locked up to generate a loan gets lent out? How can we start to understand and more accurately price the risks that inherently lie below the surface, now that we’ve introduced leverage?