2 ms·

I had this same idea a few months ago: https://mobile.twitter.com/AlanJay1/status/1418786319199780865 https://mobile.twitter.com/AlanJay1/status/141878631919978

I had this same idea a few months ago: https://mobile.twitter.com/AlanJay1/status/1418786319199780865 https://mobile.twitter.com/AlanJay1/status/14187863191997808... I'd bet on USDT failing in 3 years. But I'd want to understand how the close-out works under the assumption that the non-stablecoin collateral in Aave crashes and the liquidation process isn't able to preserve the value of the pool. I haven't studied that in depth. My guess is there's non-negligble risk of not being able to get that return on the last leg of the trade, even if the hypothesis is right.

c7b 4y ago
I guess most people who looked at crypto had that idea, and most people who thought about for more than a short moment probably came to similar conclusions as the OP (if you do it, do it through DeFi, and ideally only with what's probably play money for them). One thing they didn't mention that someone who just wants to play around should keep in mind is that you might incur tax reporting obligations. On the technical point: USDT isn't used as collateral in the protocol, apparently, so the main risk is that other assets in the collateral pool like Ether drop so sharply that the liquidation mechanism can't keep up / that the price oracles get messed up. That's a known unknown, there are also unknown unknown, eg bugs/unexpected behavior in the protocol. There's also the question whether/how you'd be able to convert your crypto holdings back into fiat dollars at that point (most centralized exchanges might be in trouble, with USDC you need to create a business account with Circle, but still seems like a feasible off-ramp).