Fluidity University Part 2: Distributing yield through the Transfer Reward Function
In the first part of Fluidity University, we’ve seen how the protocol protects itself against fraudulent transactions by piggy backing off transaction fees. In this entry, we want to study what lies at the heart of Fluidity’s economics design and is responsible for the distribution of rewards: The Transfer Reward Function (or TRF for short).
The TRF derives its name from transfer functions in control theory, as we took inspiration from closed feedback loop systems in designing our own economics modelling. Traditional financial tools only provide a limited range for stimulating certain outcomes. The applications of these tools describe open loop systems. Decentralised Finance however gives us a wide range of tools for incentives and disincentives, which we’ve considered in designing our economics.
The TRF mainly works by taking the annual percentage yield (APY) the protocol receives from lending out the underlying assets, the number of Fluid Transfers per block and the fees for a single transaction and plugging them into the function. The output gives a vector for the payouts in order to randomly distribute the yield according to the percentages. This process occurs according to a drawing mechanism that is similar to a lottery drawing, with the exception that it is continuous and instantaneous instead of having drawing periods. The TRF is called for every transaction with Fluid Assets. The values of the function can range from a few cents every 3rd transaction, up to millions of dollars in rewards at least once every 3 months.
So how does it work? We have designed the TRF to be modular and composable for various use cases. If we break it down, its simplest component reveals a power function of the form 1/x. Plotting this function we can see the following graph, where x is the probability of a payout, and y is the associated payout.
We can see that y is inversely proportional to x, i.e. if the probability increases, the payouts become lower. This is the first building block.
Now to get closer to the actual TRF, we introduce two coefficients to this power function, named a and c. a is used for distributing the yield based on the number of Fluid Transfers and includes the Optimistic Solution. As a reminder, this is used to prevent spam attacks if the expected value becomes greater than the transaction fee (as discussed in our previous entry to this series).
c is reserved for the Elastic Sigmoid Curve (ESC), which fractionalizes the payouts y according to the law of diminishing returns [link]. Its function is to incentivise good behaviour. It does so by boosting rewards and providing diminishing reward multipliers for normal payouts and the distribution of governance tokens through Utility Mining. The value of the ESC can be increased for an individual user depending on how they interact with the protocol, e.g. locking governance tokens or providing liquidity and holding onto Fluid Assets.
The values of x will then be calculated according to the hypergeometric distribution. This way the TRF becomes non-continuous and its values are given out in vectors with length M, which stands for the number of reward tiers (the first one being the lowest and the corresponding reward to M being the largest, which is paid out once every 3 months).
Combining all of the above gives you the following:
https://colab.research.google.com/drive/1ZToXmG8YOyZz_IyHwQUH3W_Jl2Ws_NI1?usp=sharing
We recommend you to play around with it a bit, and come up with your own scenarios for using Fluidity.
As we will see, the TRF is highly versatile and can be used in many applications, not just for distributing yield, but also for distributing tokens in a fair way or finding optimal token emission rates. We will talk more about this later.
You can learn more about the Transfer Reward Function and other token mechanisms in our Economics Whitepaper. If you’ve made it this far through this post, we highly encourage you to read it yourself! And as always, feel free to reach out to us if you have any questions.
In the next part of Fluidity University we look at Utility Mining, a novel and fair strategy for distributing tokens via the Transfer Reward Function. Be sure to not miss new entries in this series by subscribing to our newsletter and social media channels:
https://www.fluidity.money/
https://t.me/fluiditymoney
https://discord.gg/w9DVhGDR
https://twitter.com/fluiditymoney
