Why are you using Uniswap as a price oracle? Isn't that ill advised especially with a TWAP methodology, an exchange with a low volume can easily be manipulated by someone with not-even-that-deep of pockets.
I think low-liquidity rather than low volume would be the issue, and TWAP helps smooth that out somewhat (a price spike due to a large trade wouldn't immediately change the TWAP, and arbitragers would then bring price back to parity with other pools).
It would be ill-advised to use low-liquidity tokens for the credit rating however, I'm not sure how this is addressed by the beanstalk protocol because I didn't do a thorough reading yet. But if you're establishing something similar to a credit score based on an account's crypto assets, it would be unusual to factor in their holdings of tokens which might have very low liquidity, things like NFTs, etc.
so when the price of beans is low -- i.e. when there's more bean supply than there is demand -- the protocol tries to spur demand by increasing bean-denominated yield for beans ("weather"). the downstream effect of this is to further increase supply. and so the supply/demand imbalance that led to lower-than-$1 value persists (or will occur later, as the supply grows over time).
there's a million different ways to look at a system like this, and tbh i'm sort of drowning in the metaphors ("pods", "stalk", "weather", etc). is there a more convincing argument for how the protocol deals with the bean value < $1USD scenario?
I think you're on the right track regarding increasing the "weather" when the price is below the dollar peg — however this isn't quite to spur demand, it is to remove beans from circulation. This is what happens when you "sow beans", the protocol burns them and in exchange you get a locked-in interest rate (in pods), and you get paid out when all the lenders in front of you (the pod line) get paid out.
(and again, this is an activity that is incentivized when bean price is below $1 in order to bring the price back up)
Admittedly, the metaphors can get dense but personally once I started to wrap my head around it they were quite helpful!
> This is what happens when you "sow beans", the protocol burns them and in exchange you get a locked-in interest rate (in pods), and you get paid out when all the lenders in front of you (the pod line) get paid out.
ohh, i misunderstood the lockup conditions around lending. so the protocol properly does decrease circulation. still at the expense of an increased maximum supply, but with the rate at which that new supply enters circulation controlled for independently (by the "bean supply").
i don't see any mechanism to decrease the maximum supply. over the very long term, it seems that the system becomes increasingly unstable the longer that the bean demand growth is below the Weather rate. the bean demand growth ought to have some resemblance to the general rate of return, so the uncertain thing here is: will lenders lend at advantageous rates to Beanstalk after it establishes itself as a good creditor? or will lenders demand an outsized return for lending to Beanstalk, on account of their money being locked up for an indeterminant duration? i see arguments for both, but it's pretty academic: it looks like there's enough runway in the system that these "very long term" preponderances are mostly irrelevant in this stage of defi.
> will lenders lend at advantageous rates to Beanstalk after it establishes itself as a good creditor? or will lenders demand an outsized return for lending to Beanstalk, on account of their money being locked up for an indeterminant duration?
One would think that once Beanstalk has an even more robust credit history, lenders with lower risk profiles would be even more inclined to lend to Beanstalk! Once you have a sufficient credit history I don't think the "indeterminate" lock up period will be a deterrent (that's what high weather helps mitigate in the near term).
The protocol is in a debt cycle right now (as indicated by the market cap being $50M while the pod line of lenders is $390M), but in the long run the debt level should be much more reasonable and the market cap will be significantly larger than the pod line.
> interesting project.
Agreed! There are a lot of new, interesting aspects to this specific algo stablecoin that I haven't seen before. And the best part is that it's working! All other non-collateral algo stablecoins failed within a month (like ESD, Basis Cash, etc), but Beanstalk has been chugging along for 4-5 months now. I think the Lindy effects are strong here.
I think Bean’s ambition to be a 0 or extremely low interest lender is what got me really interested. They’re going to get there through a decentralized liquidity pool then attract creditors.
18 comments
[ 7.1 ms ] story [ 55.6 ms ] threadIt would be ill-advised to use low-liquidity tokens for the credit rating however, I'm not sure how this is addressed by the beanstalk protocol because I didn't do a thorough reading yet. But if you're establishing something similar to a credit score based on an account's crypto assets, it would be unusual to factor in their holdings of tokens which might have very low liquidity, things like NFTs, etc.
there's a million different ways to look at a system like this, and tbh i'm sort of drowning in the metaphors ("pods", "stalk", "weather", etc). is there a more convincing argument for how the protocol deals with the bean value < $1USD scenario?
(and again, this is an activity that is incentivized when bean price is below $1 in order to bring the price back up)
Admittedly, the metaphors can get dense but personally once I started to wrap my head around it they were quite helpful!
ohh, i misunderstood the lockup conditions around lending. so the protocol properly does decrease circulation. still at the expense of an increased maximum supply, but with the rate at which that new supply enters circulation controlled for independently (by the "bean supply").
i don't see any mechanism to decrease the maximum supply. over the very long term, it seems that the system becomes increasingly unstable the longer that the bean demand growth is below the Weather rate. the bean demand growth ought to have some resemblance to the general rate of return, so the uncertain thing here is: will lenders lend at advantageous rates to Beanstalk after it establishes itself as a good creditor? or will lenders demand an outsized return for lending to Beanstalk, on account of their money being locked up for an indeterminant duration? i see arguments for both, but it's pretty academic: it looks like there's enough runway in the system that these "very long term" preponderances are mostly irrelevant in this stage of defi.
interesting project.
One would think that once Beanstalk has an even more robust credit history, lenders with lower risk profiles would be even more inclined to lend to Beanstalk! Once you have a sufficient credit history I don't think the "indeterminate" lock up period will be a deterrent (that's what high weather helps mitigate in the near term).
The protocol is in a debt cycle right now (as indicated by the market cap being $50M while the pod line of lenders is $390M), but in the long run the debt level should be much more reasonable and the market cap will be significantly larger than the pod line.
> interesting project.
Agreed! There are a lot of new, interesting aspects to this specific algo stablecoin that I haven't seen before. And the best part is that it's working! All other non-collateral algo stablecoins failed within a month (like ESD, Basis Cash, etc), but Beanstalk has been chugging along for 4-5 months now. I think the Lindy effects are strong here.