Thread by @el33th4xor

DSD, an algorithmic stablecoin, up trading around $0.39.

Here are my thoughts on why and what to expect, in the hope they may be of interest or inspiration to DeFiers.

https://twitter.com/jp_koning/status/1342826994258620418

DSD, derived from ESD, follows a fairly standard design, where the amount-in-circulation of the coin is adjusted algorithmically to maintain the peg at $1. The (simplified) idea is to mint S when price is above $1 and to withdraw S below $1, to keep the value of S pegged to $1.

There are some design choices in these coins: the period at which adjustments are done, the oracles used for the price information, their granularity, the amplitude of the adjustments, and so forth. What you want is a dampening control loop.

For instance, here's the control loop that the DSD designers had in mind.

Because smart contracts are passive, a secondary, designed-to-be-volatile coin is typically used to provide the right incentives to active market participants to step in and help the system do the right thing, that is, mint more coins or pull them out of circulation.

There are many wonky, technical, complicated discussions to be had about these design choices. Which oracles? Which time granularity? Which response function? etc.

I'm not going to get into them. There's no need.

There is a big problem that many people tend to miss about these coins, and I want to point that out.

There are really two potential challenges such a coin must solve: what to do when the coin is trading above its peg, and what to do when it is under.

It should be amply clear from basic supply-demand laws that if the coin is trading above $1, and you mint more of it and give it to people, its price will eventually drop. This is the easy problem. (h/t @jaesf)

The real problem is what happens when the token is trading below its peg of $1.

Who will step in?

The designers of the coin will give you a lot of reasoning about all the incentive structures they have in place. Push those aside for a minute.

The point I want to make today is that **these coins may indeed have a peg, but that peg may very well be stable at a value other than $1**.

To use technical jargon, there may indeed be a Schelling point, but that point may reside somewhere other than the designer's intended $1. Let me illustrate.

The market tells us that DSD is trading at $0.38 right now. Not enough DeFiers are actively engaging with the protocol to boost its price to $1.

But there's probably some price at which it becomes worthwhile for them to do so.

Perhaps that's $0.37. Perhaps it's $0.0000001. Whatever it is, it's below the current point.

And those people will sell when they think they have earned enough, and when they think DSD price is above a level where it's worth it for others to step in, which is evidently >$0.38.

So the protocol, its incentives **and the expectations of the users from other users** altogether form a series of differential equations that set the expected price action around a peg.

But that peg is an emergent value. It may be $1. But it may be lower.

If the stablecoin is consistently seeing usage demand, then it's likely to be stuck to its designed peg. I believe that the successful algo stablecoins (e.g. @MakerDAO) are enjoying such a dynamic. This is indeed a stable Schelling point.

But just because there is a gradient around a designed price (that is, minting above $1, withdrawal below $1) doesn't mean that users will follow that gradient to its conclusion. They will travel as far as what they think other users will do.

Just my $0.02 about why one might encounter a stablecoin designed for $1 trading at a value lower than $1.

If you're interested in stablecoins, this paper that we wrote provides a classification framework for them:
files.avalabs.org/papers/stablec…

And needless to say, we love everything related to DeFi and stablecoins at Avalanche ($AVAX).

Just to be super clear: I don't think there's anything inherently wrong with DSD and nothing I said here constitutes an expectation that it will continue to trade below $1. It's just an interesting case study in econ/CS/game theory.

And because the price is an emergent property dependent on expectations people have of others' expectations, it can probably be restored by a whale who buys enough coins to boost price to $1.