Nima's notes

Logo

Nima Kamoosi's public notes

View on GitHub nimakam/nimakam-notes

Ecosystem

The stable-coin system lives on the public blockchain, however the players interacting with it directly or indirectly, all live in the real world and will follow strategies and courses of action, based on their incentives and the changing ecosystem. In order to ensure the success of the stable-coin system, and the money product that it represents, we need to fully understand the ecosystem subsuming it and mitigate any foreseeable issues that may threaten the product’s success.

Actors

We start with the various actors involved in the greater ecosystem:

Money user

Money users are not deeply involved with the game theoretic aspects of the ecosystem. They rather just want to use a dependable form of money for everyday economic transactions. They need that money be accepted in as many contexts as possible, and are very much interested in receiving interest on it when not in use.

They of course need to trust that the value of their money is protected, and any sign of unpredictability and volatility will drive them away from using the money.

Loan taker

Given the typically large investments made by loan takers, as well as their intermediate to high understanding of finance, loan takers do get deeply involved in the game theoretic aspects of the ecosystem, in order to ensure their collateral value is kept safe and that they could have predictable and reasonable access to it when needed.

They need custody of owned ETH, assurance that it is kept safe, and is retrievable, such that the can safely and freely use the issued pegged currency from their loans (debt positions) to perform other profitable economic transactions. The need a basic understanding of how their loan works in common conditions and peace of mind that long term risks are minimal. They also would prefer to have assurances that they will be notified when action is needed, in order to protect their investment.

They are the key set of actors that contribute to key decisions around selection and compensation of price feed providers (oracles). They are also incentivized and engaged enough to have a higher likelihood of responding to major changes in the ecosystem conditions, such as a move towards lower trust or unpredictability.

Price feed provider

Having a healthy set price feed providers is crucial to the overall health of the ecosystem. After all they are the ones working actively everyday to maintain the system by providing accurate price information to the public blockchain contracts. Certain price feed providers can amass considerable power by building scale through brand recognition and relationships with key players, to influence and the likelihood of and capture loan allocations.

The default mode of interaction between price feed providers themselves is cooperative competition for share of loan (debt position) allocations, a decision that is determined by the sentiment of loan takers and in some cases loan taking aggregators (such as InstaDapp). In this default mode, all price providers provide accurate price information to the on-chain contracts, and try to differentiate their services to loan takers, partly through promotion of other activities beneficial to the ecosystem.

Price feed providers have advanced understanding of public blockchain, smart contracts, cryptoeconomics, business, marketing, traditional and decentralized finance. They need to be able to run a profitable business sustainably, and trust in the system’s correct operation. They prefer to reach a level of scale that results in high margins and thus a higher return on investment.

Foundation price feed provider

One can imagine the role of a special non-profit type of price feed provider, one that is as decentralized and transparent as possible, and one that promotes itself to loan takers as honest and a steward of the ecosystem. Amongst its uses if its income as part of the price feed provider allocation fees, could be promoting the use of the pegged currency amongst payment providers, as well as providing audits of most or all of the ecosystem actors.

Loan liquidator

Loan liquidators are incentivized by the profit they can make as a result of liquidating under-collateralized loans (debt positions), and depend on their ability to automate the execution of high quality liquidation requests faster than any other loan liquidator.

The system does impose a delay, and other restrictions, in cases where the system is in disputed state, which may hurt the loan liquidators bottom line, due to lowered chance of liquidation as well as capital lockup for longer periods of time.

Currency market arbitragers

Currency market arbitragers help the system through periods of high volatility, likely caused by changes in the market conditions around supply and demand of the pegged currency. During these periods they actually reduce volatility by intervening in the market. More competition amongst these arbitragers, would reduce volatility even further.

In order for currency market arbitragers to perform this service sustainably, they need to have faith in the system’s ability to supply supply and demand after these disruptions in a short amount of time, and they need to consistently be able to profit from performing such services.

An example of their services, is buying pegged currency in large volumes in the market when the price has gone lower than $1.00 due to poor demand, and selling it back into the market at profit after the system has incentivized demand for pegged currencies.

Payment network integrators

The pegged currency as a money product becomes really valuable when the payment networks and apps that consumers use everyday, provide the ability for merchants and buyers to transact using them.

Loan taking aggregators

As the UX for performing loan operations becomes aggregated amongst a few providers like Zerion and InstaDapp, it is entirely possible for them to provide their own default set of options to users around allocation of price feed providers on new loans. This can become a source of influence, that some price feed providers may decide to pursue with monetary or non-monetary propositions. This is of course an undesirable direction for the ecosystem to take, but it is entirely possible. We discuss possible mitigations in the game theory section.

Game theory

Strategic dynamic changes

A few possibilities around shifting of strategic power between groups of actors:

ToDo - Detail above dynamic changes along with mitigation steps considered by the system

Protocol changes

A few possible outcomes around changes to the protocol itself:

Perpetual operation (no-fork)

Summary: The same community will rely on the same protocol perpetually.

A successful general-purpose consumer money product is assumed to be held and used by large numbers of everyday people, which means the cost of altering the money during usage is extremely high to the individual people using it, as well as any other stakeholders. As a result, perpetual operation, meaning no disruptions to routine operation forever, is the ideal end-state of the system as designed, one that incurs the least costs in a large number of people having to switch from one protocol to another.

When the viability of the ecosystem are not in significant danger, this should be considered the default option. Of course, this is a subjective criteria, and is subject to the community’s interpretation.

Soft-fork

Summary: We end up with 2 sub-communities and 2 protocols adopted by each.

There is always a non-zero probability that beliefs about operation of the ecosystem, amongst its actors could increasingly diverge in time, in which case, a community split is possible. In such a case, those whose beliefs are more aligned with the existing operations and the original parameters of the system are likely to stay with it, and incur less costs due to not having to switch to a new version. The existing sub-community however will incur a lot more cost.

The cost incurred by the sub-community leaving the ecosystem, includes but is not limited to, the cost of money users exchanging out of the pegged currency, the loan takers’ cost of having to purchase sufficient pegged currency to cover loans (debt positions) and close them, as well as the cost of exiting price feed providers being slashed for the drops in their allocation numbers.

The original community will also experience a period of less certainty due to the large volume of pegged currency and deposits leaving the system, and incur other costs depending on size of the existing group.

Hard-fork

Summary: We end up with the same community moving to a new protocol. The old protocol will be transitioned and retired.

There is always a non-zero probability that the community as a whole may come to a conclusion that changes to the on-chain protocol are required in order for the ecosystem to remain viable. In such a case, the community as a whole would have to coordinate simultaneous transition to a new version of the system, one that uses parameters agreed upon by the community as a whole.

Given that the system is immutable and holds a large amount of state, in the form of value, transitioning out of it will be very costly, including but not limited to:

Community

Community’s potential points of common belief: