UMA: Modern Incentive Structures

Introduction

A blockchain oracle is a third party data-feed that serves as a bridge between blockchains and data that exists off-chain. In our last piece, we talked about what blockchain oracles are, the problem they’re looking to solve, and why they’re a crucial part of the reliability and scalability of smart contracts. The immense growth of blockchain adoption over the past few years, coupled with the flexible nature of blockchain oracles, enables an environment where multiple oracle solutions have the potential to capture value among different sectors. 

UMA

Universal Market Access (UMA) was founded by Hart Lambur and Allison Lu, who have previously worked a trading floor together at Goldman Sachs. UMA is being developed by a company called Risk Labs. Risk Labs was founded in 2018 and raised $3.9 million in venture capital by ZBS Capital and Fintech Collective. Hart Lambur mentioned that he believes oracles remain the biggest problem in DeFi today, due to the damage that vulnerabilities within these networks can create (e.g., Wormehold’s guardian attack). With that in mind, UMA aims to minimize the use of oracle networks by creating an innovative “priceless” contract design with their Optimistic Oracle (OO). 

The Optimistic Oracle was created with two generalized purposes in mind:

  1. to enable the creation of priceless synthetic tokens 
  2.  to create a decentralized oracle service by utilizing UMA’s Data Verification Mechanism(DVM).

Priceless Synthetic Assets

Synthetic assets are collateral-backed tokens that act as tokenized derivatives, enabling on-chain participants to access synthetic cross-chain tokens (e.g., wrapped bitcoin), and traditional financial products (e.g., commodities, equities, fx). The value of synthetic assets is generally derived from the index they’re referencing, which is continuously measured and relayed through an on-chain price feed. This is where the word ‘priceless’ comes into play; UMA’s priceless smart contract architecture was built with incentivized structures in place for synthetic token issuers to fully collateralize their position, without the need for a continuously updating on-chain price feed.

Priceless contracts are highly modular. As UMA matures as a protocol and integrations start to increase, more use-cases and new templates will emerge. As of this time of writing, there are three main templates being utilized:

KPI Options 

Key Performance Indicator (KPI) options utilize UMAs LSP contract template and the Optimistic Oracle to create collateralized synthetic token options which are redeemable for a governance-approved token at a predetermined maturation date or event. KPI options were first introduced in February of 2021 when UMA tested out this new incentivization structure design by airdropping a uTVL KPI option to different DAO communities (collateralized with $UMA tokens). More information about the airdrop can be found here.

KPI options are essentially conditional payments as the success of a chosen KPI determines its redemption value. The KPIs selected can be any combination of off-chain or on-chain performance metrics (e.g., TVL, Volume, dApp usage, etc.). As long as the metric can be identified and measured by the optimistic oracle, it can be used.  

KPI options were explicitly designed to align incentives between projects and their communities. Projects are effectively hiring the community to help them reach a specific goal, whether that be a simple TVL KPI or a combination of different metrics. This design type aligns incentives between projects and their communities asa cost-efficient way for projects to grow their treasury; Implementation of a tiered structure reduces financial risk of using the contract design, as projects only pay out what they’ve budgeted for each milestone.  A recent upgrade to UMA’s architecture includedthe introduction of the Optimism Rewarder contract, which enables users to create highly modular ERC721 tokens (NFTs). The Optimistic Rewarder acts as a receipt,and tracks all future actions associated with the NFT. Rulesets can then be implemented to create novel incentive structures that weren’t possible with the KPI design. 

UMA’s co-founder, Hart Lambur, talked about the Optimistic rewarder program, and how it can be used to recreate DeFi models during a talk at Eth Denver. The link to the talk can be found here

Further details regarding KPI options, including developer-related information for token issuers, can be found here

Success Tokens

Success tokens are similar to Range tokens, aiming to align the interests of VCs, projects & communities. With range tokens, projects are able to sell their tokens at an upfront discount. Success tokens incentivize investor interest by wrapping the token with a covered call option that is backed by a set amount of the token. By selling a covered call option alongside the token, projects effectively sell their tokens at a higher valuation, and VCs increase their potential upside- everyone’s happy. 

The experience and capital that comes from VC investors can be game-changing for start-up projects. In theory, the interest generated from success tokens should come from investors with genuine conviction in the project. In addition, the covered call option is only valuable if the project does well, incentivizing VCs to work with the project to boost their chances of success.

Further reading on success tokens can be found here

Range Tokens

Range Tokens were first introduced in June of 2021, and first implemented as a proof of concept experiment by the Risk Labs team in July of 2021. Range Tokens allow  DAOs to diversify their treasury by borrowing against their native token without sell pressure or liquidation risk- akin to convertible debt in traditional finance. 

The range token primitive creates a distinctive risk profile for investors with long-term conviction in a project. Range token holders are incentivized to take on the project’s downside / liquidation risk through compensation with a call option on the native token. 

An example of how this works:

  • A new blockchain start-up called ‘We’re Definitely Not Dogs Finance” is looking to raise capital and diversify its treasury by issuing Range Tokens, collateralized by their native token ($BISKT) and with an expiration date of 10 months.
  • WDNDF sets its contract terms. They issue 1 $BISKT token as collateral for each range token with a lower bound of $10 and an upper bound of $20 (1 $BISKT is the maximum amount of tokens an investor will receive, regardless of price) 
  • To generate interest from investors, WDNDF decided to offer its range tokens at a discounted price of $0.92 USDC, which reflects a 10% APY over ten months. 
  • Being the opportunistic investor that he is, John decides that these range tokens are a deal that he absolutely can not pass up and quickly pulls the trigger on his investment. 
  • If the price of $BISKT falls below the lower bound price of $10, John has exposure to a $10 put option and consequently earns an APY lower than 10%
  • If the price of $BISKT rises above $20, John has exposure to a $12 call option and consequently earns an APY greater than 10%  

Details on a Proof of Concept Risk Labs did to pilot the range token model can be found here.

Use Cases

Treasury Diversification
Figure A. Cumulative treasury AUM over 12 months. source

Despite all of the adverse price action over the past year, projects have continued to build, and value has continued to accrue into treasuries. The total AUM of treasuries has grown by $9.6 bln since June 2021 (+1690%)

As someone who has been involved in DeFi for several years, it’s incredible to think that billions of dollars are being controlled by decentralized groups in the Web 3 ecosystem. With that said, these new financial constructs are currently facing a major challenge- the lack of diversification in treasury funds. 

Figure B. -Top 20 DAO Total Treasury Balance vs. Native Token Treasury Balance. Source

Figure B shows that many of the top 20 treasuries (by AUM) mainly consist of their protocol’s native token. When everything is moving up following the index (BTC/ETH), a lack of diversification isn’t much of an issue. 

Figure C. Relative Performance of Altcoins vs. $BTC and $ETH

The problem is that crypto is a volatile market, with high beta correlations across the board. In FIgure C, we compare a basket of the top DeFi tokens, including governance tokens held by some of the DAOs mentioned in Figure B (marked in red), and examine their relative performance to BTC/ETH in the past month(marked in orange). The importance of diversifying treasury assets was first emphasized in 2017 during the ‘market crash.’ As a result, projects were forced to sell their treasury funds at historically low prices, causing many of them to shut down. 

We’ve recently experienced a modern taste of that over the past month. While BTC and ETH have experienced about a 28% drop in price, many altcoins have gone down much more. The recent market crash has resulted in total DAO treasury AUM falling by 1.2Bln, a figure that could have been much lower had risk amongst treasury assets been diversified into stablecoins and benchmark assets. Figure C further supports this claim. It shows a visual representation of top DeFi tokens including governance tokens held by some of the DAOs mentioned in Figure B (marked in red)– and their relative performance to BTC/ETH over the past month (marked in orange).

Let’s look at an example of how UMAs Range Tokens can be used to help treasury diversification while keeping both investor and community interests in mind. 

  • DX DAO currently has a liquid treasury balance of 76.9M, of which 39.6M is held in their native governance token $DXD

Despite DX DAO having one of the largest treasuries in DeFi, the governance token has very low liquidity, making it difficult for larger players to invest in their token and participate in governance without incurring slippage.

We’ll come back to the liquidity issue later. For now, let’s say DX DAO decides to be more capital efficient and diversified with their treasury holdings. The problem here is that there isn’t enough demand to absorb sell pressure, hurting price, and consequently affecting governance token holders. 

This is where Range Tokens could serve as a helpful alternative. By issuing a range token, DX DAO would be able to effectively borrow against their native token, without any liquidation risk.  DX Dao would be leveraging UMAs tech to postpone the sale of its native tokens- which would simultaneously occur at a higher valuation than going through an OTC desk or a Dex.In doing so they now have secured free capital to use, in the form of stablecoins. 

The counterargument here is that a sale is a sale, regardless if it’s happening now or the future; at some point, there will be sell pressure. While this is undoubtedly true, it would be up to DX Dao to utilize their new free capital to build value. For example, they could take the stablecoins from the range token campaign and execute different yield strategies to generate revenue for the treasury. The profits from the yield could be used for $DXD buybacks, which could then be distributed to native token LPs as an incentive to generate more liquidity. An increased APY would create more interest for LPs to provide liquidity. This results in a positive feedback loop where it will be more appealing for LPs to compound their rewards to capture a larger share of the pool rewards. 

Liquidity Mining 2.0

Liquidity mining refers to protocol-provided incentive programs that reward market participants for providing liquidity. These programs became mainstream when protocols like YFI used them as a form of fair launch, making many rich in the process.

The mercenary stalking its prey. 

Then Fall came, and the ‘.finance’ fork epidemic came with it. Every other week a new liquidity mining program launched, promising unheard of APYs (sound familiar?) Before long, mercenary capital started to creep iin. And, by creep in, I mean kick down the door and ransack the place.

The issue remains that projects tunnel vision on the bootstrapping phase,  at the expense of bringing long-term value to the network.  The moment that yields start to collapse, mercenaries withdraw their LP positions, sell their rewards, and move on to the next highest yield opportunity. 

UMA’s treasurer, Kevin Chan, proposed a novel solution by redesigning LM programs. The idea tries to reward liquidity providers with KPI options whose redemption value is directly tied to the program’s success through a combination of time and volume-based metrics. It’s important to note that while this doesn’t stop mercenary capital completely (rewards can still eventually be sold at maturation), it does delay the process, consequently buying protocols time to prove that their ability to capture value extends beyond liquidity incentives. 

Let’s walk through a short case study on how this would work using Sandclock, which is gearing up to launch its Liquidity Mining program soon with KPI options. However, before we get into it, note that this case study does not contain any factual information regarding Sandclock’s LM program; it is entirely hypothetical. 

Case study:

Sandclock intends to launch liquidity mining incentives in the near future by utilizing UMAs KPI options. Unfortunately, apart from an announcement, there isn’t much information regarding how they plan on structuring the program, including distribution. However, we know that as part of its tokenomics, Sandclock has 30% of its 100M total supply of $QUARTZ set aside for liquidity incentives & protocol development(30M).

 (Step 1) The first thing we need to do is decide on a total token allocation and expiration period. Let’s say of the 30M total distribution intended for liquidity incentives, 5M will be allocated to a quarter-long uQUARTZ KPI pilot(illustrated below showing a weekly summary). 

(Step 2) The next step would be to identify what KPI metric. Metrics can be anything measurable. Some options include volume, fees, users, social media followers, TVL, and supply staked. For Sandclock case study, let’s go a step further and use TWAP liquidity using the following formula to measure the TWAP daily for 3 months, and then take the mean to get our results:

TWAP Liquidity=mean(Open+Close+High+Low)

(Step 3) Now that a KPI metric has been identified, the next thing to do is  define a token payout structure based on the KPI metrics’ performance. The following parameters need to be identified in order to structure the payout:

  1. Floor Payout — the minimum number of $QUARTZ the KPI option will payout. – Identified in this case study as 0.5 $QUARTZ
  2. Maximum Payout — the maximum number of $QUARTZ the KPI option will payout – Identified in this case study as 5M $QUARTZ
  3. Floor TWAP — the minimum quarterly TWAP  that would reward the floor payout –  Identified in this case study as ≥10M
  4. Maximum TWAP — the maximum quarterly TWAP  goal that would reward the maximum payout – Identified in this case study as ≥50M
  5. Base Payout — Equivalent to the payout that would be received from a regular LM program (1:1 redemption)
  6. Base TWAP — The liquidity TWAP that would be expected under normal circumstances with a standard LM program – Identified in this case study as ≥30M
  7. Expiry — The date when the quarterly TWAP liquidity is measured – Identified in this case study as the end of the quarter following the start of the program.

With all of the required parameters defined, we can now run a simple simulation to get a picture of how payouts would occur in different end-of-quarter scenarios. As illustrated above, LPs receive a more significant payout when the liquidity TWAP is higher. When liquidity TWAP is lower, payments go down, ultimately saving Sandclock’s treasury 3.3M $QUARTZ in the worst-case scenario. By design, communities and LPs are now incentivized to educate themselves on Sandclocks protocol, allowing Sandclock to showcase its utility as a protocol and consequently aligning incentives for performance and growth.    

Decentralized Oracle Service

As mentioned earlier in this paper, Hart Lambur stated that general vulnerability risks that come with oracle networks are the biggest problem DeFi is facing. This is primarily due to the lack of crypto-economic security within current oracle models. Although Oracle solutions such as Chainlink and API3 plan to implement their version of crypto-economic guarantees, those plans are just promises that have yet to be fulfilled. Meanwhile, UMA’s oracle service was developed with built-in incentive structures from the start to ensure economic guarantees, theoretically making it one of the more secure oracle solutions available today. 

UMAs oracle system consists of two components working in conjunction with one another:

Optimistic Oracle

In my research on UMA, if there is one strikingly obvious theme, it is that they are big believers in incentive systems. Hart Lambur recently spoke at Eth Denver on the topic, arguing that all of DeFi is essentially just one big incentive game. Liquidity mining, consensus mechanisms, data verification, lending, andall of the other constructs that make up our decentralized financial systems exist only to serve the infamous question, “what’s in it for me?”. 

With that said, using incentives to achieve a goal is a battle-testedsystem that works, and is the foundation upon which UMAs Optimistic Oracle was built.

Optimistic Oracle Work Flow. Source
The Optimistic Oracle consists of three components:
  1. Requestors – Any smart contract or system that submits a request for data. The requestor posts a reward offer which is then accepted and fulfilled by a Proposer. Incentive: Data service
  2. Proposer – An off-chain actor who proposes an answer to a data request. For a proposer to accept a request, they’re required to stake a proposal bond as a form of a security deposit to ensure the legitimacy of the data. Once proposed, there is a pre-determined grace period during which the requestor can dispute the proposal’s validity. If the proposal were proven invalid, the proposer would lose its staked security deposit. Incentive: An honest actor has nothing to worry about and is incentivized by the requestor’s reward offer.
  3. Disputer –  An off-chain actor who has the power to dispute proposals. The ideal and most common scenario is when a proposal is created, the dispute period comes and goes without the disputer needing to do anything. In the event where a disputer disagrees with the proposal, the disputer matches the proposer’s bond with a bond of their own, and the case gets escalated to the DVM (mentioned below). Incentive: If a disputer can prove that the data relayed by a proposer is wrong, they’re rewarded with the proposer’s bond. 

For further reading on the Optimistic Oracle, refer here

Data Verification Mechanism (DVM)
DVM Work Flow. Source

UMAs DVM is the next element of crypto-economic security provided by UMAs oracle service. The purpose of the DVM is to handle situations where proposers and disputers disagree on the legitimacy of a proposal or when a contract liquidation occurs. The workflow for how DVM handles disputes can be seen in Figure 2 and can be generalized in three main steps:

  1. UMA’s DVM aims to create a consistent environment where the Cost of Corruption (CoC) is always greater than the Profit from Corruption (PfC). The first step in achieving this is by identifying and measuring the CoC.  When a dispute gets escalated to the DVM, $UMA token holders reference the price identifier’s UMIP to calculate and vote upon the data they believe to be correct. The vote is taken place on UMA’s Voter dApp. By utilizing a governance-based approach for disputes, the CoC consequently becomes the cost of buying a majority share of $UMA tokens for voting (51% minimum).
  2. The next step is identifying how much a malicious actor could gain from corrupting a data feed.UMA requires all project integrations that use the optimistic oracle to register with the DVM and report the maximum potential loss they could face from an attack. The DVM then aggregates all individual values and measures the total potential profit from corruption. 
  3. Once the DVM calculates the CoC & potential profit from corruption, the logical next step is to establish a mechanism that continuously measures the cost to profit ratio and enforces a higher cost. The DVM does this with a variable-fee policy used to consistently rebalance the ratio through  $UMA token burns and buybacks as its market cap fluctuates. 

This is a very high-level overview of the Data Verification Mechanism. I would highly suggest reading through their documentation for more details regarding the DVM, including its math. 

Tokenomics

$UMA Token Distribution (0x04fa0d235c4abf4bcf4787af4cf447de572ef828)

  • (35 million) tokens are currently held by individuals on the Risk Labs team.
  • (10 million) are held in the Risk Labs treasury.
  • (15 million) are held by other investors.
  • (3 million) are deployed as Grants/Liquidity Mining
  • (2 million) are deployed through AMM and market making distribution.
  • (35 million) are held by the UMA DAO.
  • Inflation Rate: The total supply increases by 0.05% every time there is a DVM vote. 
  • Current total supply of 107M tokens with a circulating supply of 66M (62%)
$UMA Governance

UMA is an ERC-20 with its main use-cases revolving around governance. $UMA tokens are used to vote on oracle and liquidation disputes. By participating in governance, voters receive inflation rewards and earn about 40% APY on tokens, not including gas reimbursements. Note that rewards are paid out only if you vote for the correct answer

Final Thoughts & Additional Notes

  • It’s important to note that UMA’s current live products serve more as a proof of concept to showcase its potential. 
  • UMA is a relatively new protocol with a complex architecture that will take time to mature and find its market fit as it becomes more understood. 
  • Due to UMA’s flexible architecture, a number of interesting use-cases are emerging, such as Opium using UMAs Optimistic Oracle to offer insurance for space X launches.
  • UMA is looking to expand beyond the scope of creating accessible financial instruments. There are currently three other live products, including Across, a cross-chain bridge protocol. More information on these products can be found here

This report is not investment or trading advice. Please conduct your own research before making any investment decisions. Past performance of an asset is not indicative of future results. The Author may be holding the cryptocurrencies or using the strategies mentioned in this report.

Stay up to date with our research

Sign up to receive an email when we release a new post