Why does Olympus DAO work

DeFi bonding series part 1

Fig.1. Olympus DAO’s website image

Olympus DAO was arguably one of the hottest DeFi 2.0 projects in 2021. Since its inception, Onther has been actively looking at it as one of the possible answers to solving the mercenary liquidity problem (explained more in the next section) for Tokamak network. After many hours of internal research and discussions, instead of using Olympus DAO model as it is, we have decided to design an improved version of it. The result is presented in the following three-part posting series:

• This posting will summarize the building blocks of Olympus DAO and explain why it works.
• The second posting will define the DeFi bonding primitives, the building blocks of Olympus DAO’s bonding system, and generalize each part.
• Finally, the third posting will show an example of a new DeFi bonding model called the user owned liquidity (UOL) specifically designed for the launchpad platform.

Introduction

The automated market maker for ERC20 tokens, such as Uniswap, enabled the emergence of new DeFi systems that utilize user-friendly and gas-effective ways to swap tokens directly on Ethereum chain without using order books.

But, it also left homework of combatting mercenary liquidity providers that provide large liquidity for the short term to maximize their incentives and remove their liquidity once the incentives are gone.

Olympus DAO is one of the solutions to mercenary liquidity. It is a community-owned decentralized financial infrastructure to provide a global stablecoin (i.e., OHM) backed by crypto assets. Instead of actively combating mercenary liquidity, they propose a highly coupled system to maximize the liquidity owned by the protocol, also known as protocol owned liquidity (POL).

How does POL solve the mercenary liquidity? Suppose a protocol owns the majority of the liquidity. In that case, it has no reason to provide additional incentives to liquidity providers, and without the incentives, it also eliminates the reason for mercenary liquidity providers to provide liquidity.

But, POL comes with significant risks for users as they have to put all of their trust in the protocol. We will discuss this more in the later posts.

In the next section, we will describe the building blocks of Olympus DAO’s bonding system that enables a protocol to own the majority of the liquidity effectively.

Olympus DAO’s bonding system

Olympus DAO’s bonding system, which consists of “bonding” (treasury), SushiSwap & UniswapV2, and auto-staking, is the heart of the operation for the protocol to own the majority of the liquidity. We will describe each functionality and analyze how it contributes to the protocol owning most of the liquidity.

Fig.2. Olympus DAO’s bonding system (v2)

“Bonding”

“Bonding” is a capital injection mechanic that allows the exchange of various crypto assets for OHMs at a discount with a vesting period.

To be more precise,

1. First, the user deposits their crypto assets such as Ethereum, stablecoins, or liquidity provider (LP) token to the treasury.
2. The deposit’s worth is then assessed to determine how many OHM tokens will be given to the user after the vesting period is finished. Usually, a small discount is applied to the “bonding” price to entice the user to “bond”. For example, if the current OHM’s price is 40.1 DAI per OHM and the discount is 0.1 DAI per OHM, then the “bonding price” is 40.1–0.1 = 40 DAI per OHM. And if the deposit is worth 80 DAI, they will receive 80/40 = 2 OHMs at the end of the vesting period.
3. Then, Olympus DAO’s treasury mints 1 OHM per 1 DAI based on the deposit’s worth. For example, if the deposit is worth 80 DAI, 80 OHMs are minted.
4. After the vesting period is passed, newly minted OHMs are given to the user based on the “bonding” price. Continuing with the previous two examples, 2 OHMs are given to the user for the 80 DAI deposit. Note that any excess OHMs from minting are reserved for another usage (example: rewards for staking, income for DAO, rewards for front-end operator, etc)

There are three things to highlight for “bonding”. First, the protocol only increases its liquidity position on the liquidity pool when a user gives up their LP token for “bonding”. The second is that the discount offered on OHM when “bonding” rewards users for exposing themselves to significant risks from the price of OHM possibly decreasing after the vesting period. If the price of OHM is increased or stays the same by the end of the vesting period, “bonding” is a cheaper way to get OHM. Finally, the third is that treasury mints OHMs equivalent to the asset’s worth, which effectively creates backing for every minted OHM.

Fig.3. Sushiswap image from binance academy.

SushiSwap & UniswapV2

SushiSwap & UniswapV2 are non-custodial ERC20 token swap dapps that provide a decentralized way to swap between ERC20 tokens and OHMs.

In a centralized exchange or order book based DEX, buyers and sellers must coordinate the swap ratios and the number of tokens to be swapped. But, in SushiSwap & UniswapV2, users can directly swap their tokens on Ethereum, and the swap ratio is automatically determined using the constant product formula.

The constant product formula determines the swap ratio between the two tokens in the liquidity pool. It is designed to fix the product of the number of each token in the liquidity pool to be a constant. For example, if there are 200 Ethereum and 10 OHMs in the liquidity pool, then swap should be performed such that it maintains the product of the number of each token, i.e., 200 x 10 = 2,000. Paradigm Research provides an excellent explanation in their posting if you want to learn more about the constant product formula.

Unlike “bonding”, swaps do not inject new capital to Olympus DAO. Still, OHM’s relative price rises when a user swaps for OHMs due to the constant product formula; OHMs become more valuable, effectively increasing the total values of OHMs.

Auto-staking

Auto-staking is the glue that holds “bonding” and SushiSwap & UniswapV2 together to form a single system to maximize the liquidity owned by the protocol.

In Olympus DAO, OHM staking reward incentivizes users to lock their OHMs to protect the relatively high OHM price in SushiSwap & Uniswapv2 liquidity pools. To further incentivize the users from selling their OHMs, the staking rewards are given every 8 hours and the rewards are automatically auto compounded. This means that staked OHMs are auto compounded three times a day!

Auto-staking is a feature that further improves the user experience of “bonding” and staking. The OHMs from “bonding” are automatically staked so that the user won’t miss out on the rewards from staking during the vesting period. Thus, users don’t have to worry about diluting their OHMs during “bonding”.

Why does Olympus DAO work?

To summarize, Olympus DAO’s bonding system is designed such that

1. Olympus DAO owns most of its liquidity: Auto-staking feature combined with “bonding”, provides an attractive way to trade LP tokens for OHM at a discounted price.
2. Providing liquidity is disadvantageous: Staking offers a higher profit than the expected profit of providing liquidity. This effectively dissuades mercenary liquidity providers from joining the liquidity pool.
3. Owning unstaked OHMs for a long time is disadvantageous: Auto-staking OHM is the best way to prevent loss from dilution. Since OHMs in the liquidity pool are technically unstaked, Olympus DAO holds the most unstaked OHMs and stands to lose the most from dilution as time passes.
4. Most OHMs are locked out of circulation: Auto-staking locks OHMs until they are unstaked, and Olympus DAO owns most OHMs in the liquidity pool. Consequently, OHM’s price can quickly rise above its backing value of 1 DAI.

In theory, POLs like Olympus DAO benefits immensely from the reduced risk of mercenary liquidity and the high valuation of their native token. But, they also expose significant risks to the users because they control most of the liquidity.

For example, it is easy for the protocol to manipulate the market by reducing the liquidity or even using treasury assets to change the swap ratio in the liquidity pool. Obviously, DAO should be designed to prevent such behaviors. Still, in the current Ethereum environment — where the gas fee is expensive and voting can be manipulated by massing governance tokens by the rich- users are exposed to the risks associated with POL.

So, how do we design something better or even different from Olympus DAO’s bonding system? Stay subscribed for the next medium post that will break down Olympus DAO’s bonding system into DeFi bonding primitives that can be used to build a system with a different goal.

Acknowledgment

This posting series is the result of research and analysis of Onther team members (names are listed in alphabetical order):

• The main idea for improving Olympus DAO: Kevin.J
• Community building research and analysis: Kevin.J, Oscar.K, Richard.N
• Smart contract design and research: Harvey.J, Kevin.J, Praveen.S, Suah.K, Zena.P
• Model feasibility analysis: Gongpil.C, Jamie.J, Kevin.J, Oscar.K, Suah.K, Zena.P
• Olympus DAO’s model analysis: Harvey.J, Jamie.J, Kevin.J, Praveen.S, Suah.K