We have been working to continuously improve and refine the protocol and its governance. Over the years, the BZRX token model has evolved along with the industry, presenting fresh new ideas in the process.

2018: It was initially proposed that the protocol token, BZRX, be used to govern the protocol and as a medium of exchange. The token model was inspired by the earlier work of 0x on ZRX. As token economic theory advanced, it became clear that the medium of exchange model would not provide the kind of value capture required to align the incentives of protocol stakeholders.

2019: Proposed the first example of a non-extractive value capture mechanism. Under this model, BZRX tokens could be redeemed for a proportional percentage of the insurance fund. As protocol activity generates a larger insurance fund, the tokens become more valuable. This ensures that BZRX holders are aligned with securing the protocol lending parameters to prevent a loss to the insurance fund.

2020: We proposed the bZxDAO, a trias politicas model for governance, with participation incentivized through staking rewards. Under this model, active participation is rewarded with an ever increasing share of protocol ownership. Conversely, token holders not participating in protocol governance slowly bleed ownership of the protocol. Changes have been made to this model, like the removal of inflation rewards.

Today, we share two exciting additional proposals to align the incentives of stakers and users.

The Fee Sharing and Protocol Disbursement Initiatives

Fee Sharing

Each trade made, loan originated, and debt serviced generates fees for the protocol. There are three fees of which to be aware:

Fees are both paid and kept in their native assets, never requiring an extra swap. Since the protocol supports a large number of assets, the fee pool contains a large number of denominations of assets as well. When an individual stakes their BZRX to a representative, they are entitled to new fees generated by the protocol in proportion to their contribution to the staked supply.

Each time a protocol fee is generated, it is held in the protocol until the feeSweep() function is called, sweeping the fees into one of two Balancer pools containing the rest of the fee pool tokens. The feeSweep() function is permissionless and can be called by anyone, allowing rewards to be accessed in real-time. When fees are deposited into the Balancer pool, it automatically rebalances, emitting a token that represents ownership of the liquidity pool as a whole. Stakers receive the Balancer LP tokens minted during their staking period, proportional to their contribution to the staked supply. The rewards from fee sharing, i.e., the Balancer LP tokens, both accrue in real time and can be redeemed at any time for any other asset in the pool.

While the assets remain in the Balancer pool they earn fees from providing liquidity to users of Balancer. In the future it could even provide liquidity to traders on Fulcrum, allowing the protocol to source liquidity internally by providing AMM liquidity with its own assets. At the time of writing this article there will be two Balancer fee token pools, a stablecoin pool with no impermanent loss and a variable fee pool that holds more volatile assets such as ETH and other ERC20s.

The user experience of claiming fees and staking BZRX is as simple as can be with the introduction of our bZx Staking Portal. You can stake or unstake with a single click, and you can claim your rewards with a single click as well.

The staking dashboard also provides a history page to see your staking history and any fees collected during your staking period.

Protocol Disbursement Program

We are distributing ownership of the protocol to its users. We have allocated BZRX tokens to reward users of the protocol commensurate with their contribution to the value captured by the protocol. In total, 20% of the token supply has been allocated towards rewarding the users of the protocol. These tokens are split into two suballocations:

This means that anyone who executes a trade, opens a loan, or keeps a loan open, will begin to accrue ownership of the protocol. Whether we reward lenders or borrowers, we believe a virtuous cycle will take hold; the only difference is the distribution of the tokens at the end of the cycle. Whether we initiate that virtuous cycle by lowering rates, increasing yields, or doing both will not matter too much to the net level of activity on the platform. The boost in activity is going to be largely driven by the size of the subsidy rather than its distribution. Lenders are very fickle and largely commoditized. We rather maximize our influence on borrowers because it is those users that we have the best chance of getting to stick to the platform.

How Does BZRX Capture Value?

When you stake your BZRX to a representative, BZRX exposes you to four different mechanisms of value capture.

  1. Fee Sharing: Staked BZRX tokens grant proportional ownership of any Balancer LP shares minted during the staking period. For example, if a single person stakes 1% of the BZRX supply but comprises 10% of staked BZRX, then they are entitled to 10% of all Balancer LP shares minted while they staked.
  2. Balancer Fees: The fees held in the Balancer pools can supply liquidity for both traders and the protocol. Those fees accrue to all participants holding shares of the fee pool. The pools are optimized to minimize impermanent loss and maximize yields.
  3. BAL Rewards: Since the fee pool will generate volume, it will also generate BAL tokens. These BAL tokens are reinvested in the Balancer pools, allowing BZRX stakers to seamlessly benefit.
  4. Insurance fund: each BZRX token can be redeemed for a proportional amount of the insurance fund not already denominated in BZRX. This gives the BZRX token a fundamental value by being both backed by current assets as well as future cash flows. The insurance fund receives half of origination fees, fees from interest payments, and fees generated via trading. The proportion of fees directed to the insurance fund can be determined by the DAO during a future governance vote.

Staking Rewards for AMMs

Joel Monegro presented the idea of Proof of Liquidity, a concept that rests on the observation that staking can be harmful to the creation of mature, liquid markets because it disrupts core components of the initial bootstrapping process. This bootstrapping process starts with AMMs and culminates in organic activity around the token from a variety of trading providers including CEXs. It is important that providers of AMM liquidity are not penalized by missing out on fee sharing rewards.

Credit: Joel Monegro, Placeholder Capital

We propose that the LP shares of the most liquid AMM pool be eligible for staking to earn fees. The address of the LP pool is a parameter that can be changed via protocol governance. Balancer LP tokens from the fee pools will be distributed to BZRX stakers proportionally to the circulating supply. The tokens outstanding during each call to feeSweep() will be distributed proportionally to those staking BZRX LP tokens. This means that staking LP tokens as opposed to BZRX tokens may result in earning a slightly higher share of the fees since there will be unstaked BZRX in circulation that is not within AMMs.

Let us work through an example:

This means that the 100MM LP tokens will receive just as many fees and be entitled to just as much governance power as the 200MM staked BZRX. This is because the BZRX LP tokens are treated as the entire unstaked supply.

Token Distribution

Crowdsales possess a number of undesirable properties and, relatedly, a negative reputation. A crowdsale requires a large amount of coordination, considerable after-sale maintenance to ensure a liquid secondary market, and a host of potential legal/securities concerns. We have decided to take a different path. The tokens that would have been allocated to the crowdsale will be placed in a 4 year vesting contract with a 6 month cliff, starting from the time that the token is unlocked.

* Vesting tokens have 50% voting power when they are in the vesting contract.

The security fund will be used to create liquidity incentives for the iETH pool, to compensate traders for previous downtime, to pay out bug bounties to security professionals, and to align the incentives of key community members.

The following is the projected fraction of tokens in circulation over time:

The Unlocking

This has been a long time coming. We will be unlocking the token in the coming weeks as part of the bZx 2.0 relaunch sequence. More announcements will follow soon.

A Quick Refresher On Governance

The Legislative is a variant of liquid democracy with representatives elected by token holders staking BZRX. The three representatives with the highest number of tokens staked to their address become the legislature. At any point, token holders can change the representative(s) to whom their tokens are staked, but they cannot directly vote on proposals. Proposals passed by this branch must pass by a majority vote.

The legislative branch approves upgrades to the protocol and sets the critical parameters:

The tokens can be distributed to representatives, token holders, or through grants. A consequence of there being only three representatives in a system requiring majority rule is that an attack on the DAO requires at least 1/2 of the active voting power. To pass a resolution, votes must be signaled twice, and between each signal at least 16 hours must elapse. This prevents a representative from making even a single finalized vote without the full consent of their stakers.

The Executive is composed of the two leads of the core development team. The members of the executive branch will not always be part of the original core development team. Representatives may submit proposals to elect new executives. However, since this requires the vote of the executives themselves, the model resembles a Web of Trust. Much like the executive branch in traditional political systems, the executive has no power to propose or pass proposals on its own. Instead, the executive is to simply act as a check on the legislative branch, vetoing malicious proposals and attempts by representatives to form cartels.

By restricting the powers of the executive to ratifying upgrade proposals and inflation reward distributions, regulatory risk is minimized. In the worst case scenario, a regulator could prevent further upgrades to the protocol by apprehending the executive administrator keys. To mitigate this risk, executives can be replaced by a unanimous vote of the legislative. Replacement of the executive does not require approval of the executive.

The executive has the ability to unilaterally pause the protocol for 48 hours, after which there is a 3 month refractory period before another pause can be invoked. If a serious security issue is found in the protocol, security researchers can disclose the vulnerability discretely to the executive, have the system paused, and then allow for the vulnerability to be disclosed to the legislature. If the legislature cannot mobilize a comprehensive response within that 48 hour period, the pause period can be extended through the normal governance process.

The Judicial is the smart contract code and the EVM. Both branches can only act within the constraints of the smart contracts governing them.

About the author
Kyle J Kistner
CVO @ bZx. Product, Protocol Design, & Token Economics.