Risk analysis of the Terra, UST and Anchor protocol

May 12, 2022

Terra Blockchain

Terra is a blockchain with the aim of providing a platform for stablecoins, cryptocurrencies  with values pegged to real world currencies. The value of  stablecoins on Terra are stabilised algorithmically rather than by maintaining an underlying reserve of the fiat currencies (such as USDC for example). This approach lends itself to decentralisation, but  has a higher risk of the stablecoin becoming “depegged”, where the value drifts from the targeted fiat currency.

LUNA token

Terra operates a bonded proof of stake consensus mechanism to ensure the validity of transactions on the network. A number of servers (called nodes) containing copies of the full Terra blockchain act as the validators of transactions and participate in the creation of new blocks. The native staking and governance token of Terra is called LUNA, holders of LUNA are able to bond their tokens to a specific validator to effectively vote for them to take part in consensus. The 130 validators with the most LUNA bonded to them take part in the actual validation. Rewards are generated from verifying transactions and these are shared between the validators and stakers.

When a user stakes LUNA they will receive the equivalent amount of bonded  LUNA  tokens  (bLUNA). The original LUNA tokens cannot be traded or used in other protocols while bonded, but the bLUNA tokens can.

The process of exchanging bLUNA back for LUNA takes 21 days to complete, during  which time neither token can be used and staking rewards are not generated.


Another use of the LUNA token is to maintain the stability of the various stablecoins on Terra. Here we will focus on the coinpegged to the US dollar, the UST, as this is the currency used in both the Anchor and Ozone protocols. 1 USD worth of LUNA can always be traded for 1 UST onTerra, so if the value of UST fluctuates up LUNA holders are incentivised to trade LUNA for UST to make a profit and if the value fluctuates down users can buy UST with USD and then trade for 1 USD of LUNA. This mechanism has the effect of shifting most of the volatility on to the LUNA token while keepingUST at a stable price.

Anchor protocol

The Anchor protocol is built on the Terra blockchain and designed to facilitate the lending of stablecoins using bonded staking tokens as collateral.At the time of this report, Anchor supports UST as the lending currency and bLUNA and bETH as the bonded staking tokens. bETH tokens themselves are wrapped stETH tokens (staked ETH) which are designed to match the price of ETH and provide staking rewards of around 5% (although Ethereum is in a transition period between proof-of-work and proof-of-stake, it is possible to stake ETH already). In terms of value, the current ratio of bLUNA to bETH used as collateral is currently 6.6:1.

The use of bETH does introduce two additional points of failure, ETH to stETH and stETH to bETH, however the benefits of diversifying the collateral likely outweigh any additional risks. There is also a danger that switching to PoS will greatly increase the popularity of Ethereum and hence the price of ETH and make it more worthwhile to hold on to ETH than use it as collateral. The transition will likely take a couple more years to be fully completed so there should be other bonded assets available on Anchor before this is an issue.

The unique selling point of the Anchor protocol is that staking rewards from the bonded tokens are used to stabilise the interest rate for the lenders. Most decentralised exchanges have rather variable interest rates, partially due to big swings in cryptocurrency value and partially to underutilised liquidity. The target interest rate for investors is called theAnchor rate and is currently set at 20.07%.

When an investor deposits UST in the protocol they will receive Anchor UST tokens (aUST) representing their share of the pool that can later be redeemed for their original deposit + interest.

The staking rewards from the bonded asset collateral are shared between the lenders and the borrowers. The share varies depending on the utilisation of deposited funds, more of the share is diverted to lenders if the utilisation drops in order to maintain a stable interest rate.

Loans of UST must be over collateralised according to a loan-to-value(LVR) ratio, the weighted average collateralisation ratio for the currently deposited bLUNA and bETH is 2.56. If the value of the collateral drops below a minimumLVR then the collateral can be liquidated. The liquidation prevents loans from becoming under collateralised in situations where the value of the collateral tokens drops and protects the deposits of lenders.

During liquidation, bonded assets are purchased at a discounted rate for UST by executing a number of standing orders when the minimum LVR is passed. Liquidators are required to lock their UST in these standing orders and must wait a month if they wish to retract their order. As well as discounted bonded assets, liquidators also receive a variable interest rate on their UST set by the number of standing orders present. The borrowers must also fund the interest rate that the liquidators earn. If there are not enough standing orders to fulfil the liquidation requirements, the assets can be effectively auctioned off on the open market. This could result in a loss of value in theAnchor pool reserves and potentially eat into the interest of lenders.

An oracle is required to set the accurate and up-to-date prices of the bonded assets. This is a critical part of the protocol because an inaccurate price could result in the liquidation of borrowers assets and the oracle does not operate trustlessly like the rest of the protocol. This has caused an issue previously and will be discussed further in the later analysis.

The interest rate offered to borrowers is variable and operates more like most DeFi lending pools. As the liquidity of the pool increases, borrowing becomes cheaper to incentivise greater usage of the liquidity. As the liquidity of the pool decreases, borrowing becomes more expensive as there is more demand for loans. Normally, the interest rate for lenders would increase and encourage more liquidity, but in Anchor the distribution of staking rewards is shifted to maintain the Anchor rate. This has the effect of subsidising the borrowing interest rate and making it cheaper to borrow when utilisation is high.

The other side of this is that when the liquidity decreases, the distribution of staking rewards will be shifted in favour of the lender, making it more expensive to borrow than in a traditional DeFi pool. The stability of this system will be investigated in more detail later but it has the potential to cause a runaway effect if not enough liquidity is utilised.The borrowers are paying for the interest rate for the lenders through the borrowing rate and lost staking rewards, if the utilisation ratio (total borrowed/total lent) of UST becomes too high it will become prohibitively expensive to borrow and the Anchor rate could collapse.

If the actual Earn interest rate is higher than theAnchor rate, excess funds are shifted into the yield reserve. Conversely, the yield reserve can also be used to subsidise the Anchor rate when the actual Earn rate is too low. The increasing popularity of the system and healthy rewards meant that the initial interest was around 25-30% and the yield reserve was filled up. The current actual Earn rate is around 14% and the yield reserve is being rapidly deplenished with an influx of lenders. It is predicted that without external intervention the reserve will run out towards the end of February and this could result in a big drop in Anchor and LUNA value.

Anchor token (ANC)

Anchor also has a native token used both for the governance of the protocol and as an incentive for participants in the protocol. As the borrowing rate decreases the ANC incentive to borrowers is increased. A portion of the interest generated by the pool is used to purchase more ANC tokens to be used as an incentive. This mechanism requires holders of ANC to provide liquidity for ANC-UST swaps so that the pool can purchase ANC with the UST it obtains from the staking rewards.

Again, this means that as the borrowing rate decreases borrowers will have to pay more in order for the pool to purchase ANC tokens to encourage more borrowing. It is likely the net effect of this is zero (or even negative for borrowers because of the fees accrued when purchasing and swapping tokens),but this needs further investigation.

Currently ANC tokens are still being minted, with ANC tokens being created over a 4 year period (from the launch of Anchor) for distribution amongst borrowers, lenders and others. As Anchor has grown in popularity and the value of ANC has increased the newly minted ANC has been sufficient to incentivise borrowing and keep the Anchor rate stable. It will be interesting to see how the protocol operates once the minting dries up, but the analysis section of this report will try to predict that.

Risk Harbor Ozone

Ozone is an insurance protocol on the Terra blockchain that was acquired by Risk Harbor, a DeFi risk management marketplace. It is designed to protect providers of liquidity against smart contract risks and attacks. The protocol allows a user of Anchor to purchase protection for their aUST. Once protected, if the value of aUST drops below a certain threshold relative to UST they can file a claim by sending their aUST to the protocol. If the claim is deemed valid, they will be paid out in UST according to a predefined ratio, currently 1.14UST to 1 aUST.

Ozone is a Terra version of Risk Harbor Core version 1 (V1),an EVM based insurance protocol that has been deprecated in favour of version 2(V2) of the protocol. Core V2 currently covers 4 different protocol tokens (including a UST) with USDC. There is currently no protection remaining for aUST on Ozone and around 16k aUST worth of protection on Core V2. The Core protocol has the added benefit of protecting against a UST depeg as USDC is independent of the Terra blockchain.

In the Core V1 protocol underwriters transfer funds to an underwriting pool that is configured to provide protection for a specific protocol, with a certain interest rate and expiration date. Users of the protected protocol can purchase protection by paying the total premium calculated to the expiration date up front. This premium is distributed amongst the under writers over time proportional to their share in the underwriter pool.

The official documentation on the Risk Harbour protocols is fairly sparse and the smart contracts are currently closed source, although there do appear to be plans to change that. There is no way to provide UST to be used as protection on the Ozone web application so it appears that all of the initial underwriting funds are from the Terra community pool. Ozone offers protection to Anchor users and, as Risk Harbour mention in their own whitepaper, it makes no financial sense to provide protection for a single protocol as you would be exposed to the same risk and better rewards by investing in the protocol.

The Ozone whitepaper suggests that it will move towards a similar model to Core V2, but provides no timelines for doing so. The V2protocol spreads the risk amongst several protocols in order for it to make financial sense to provide underwriting funds. A key component of this is determining correlated risks between protocols. The V2 whitepaper reports that the Risk Harbor team has done a lot of research into calculating these risks, but does not cite any other resources or references. There is due to be an announcement in the coming weeks so this document will be updated when this is released.

Depegging risk

UST vs USD price history

Notable events

December 2020 downwards depeg and January 2021 upwards depegs

There is little public information available about the depeg in December 2020, it was a significant drop, but was limited to a singleday rather than the week of the later depeg. This was in the early days ofTerra and the UST supply was about 1% of what it is today. The mechanisms behind Terra have been tweaked on several occasions since this, some of whichwill have been in response to this early instability, so the later depeg is more relevant to the current risk of depegging.

May 2021 downwards depeg

The extended depeg in may was due to a sharp decline in the price of LUNA, partially caused by a large number of liquidations on Anchor.Terra is designed to have a soft cap of $20 million of redemptions of UST toLUNA with a 2% spread, but redemptions exceeded $80 million with a 7-8% spread which caused UST to trade at a discount. Since the depeg the soft cap has been raised to $100 million with a 0.5% spread.

As described before, when the UST peg fails downwards it incentivises people to convert UST to LUNA and then sell the LUNA for USD, then buy back UST for a profit. As the peg was decreasing rapidly, the amount of profit that a user could expect from this process was overshadowed by the drop in UST value over the time it would take them to complete the transaction.This had the potential to result in a run on the market type situation as the motivation for restoring the peg was reduced. However, Terra is designed to handle this type of situation by increasing the cost of transactions and increasing the rewards to LUNA stakers. The in-built defence mechanism proved to work well and the peg was restored after a short while. The initial drop occurred over a period of 3 days, followed by a sharper drop over a day, after which the peg was restored gradually over another 2-3 days.

LUNA vs USD price history

Notable events

Initial price increase

The price of LUNA was very consistent for a long period of time. The jump in LUNA value corresponds to a large increase in the total supply of UST and is in general related to an influx of users to the Terra ecosystem.

First drop

This corresponds to the May 2021 depeg of UST (and is in fact the cause of it). The drop in LUNA price was the result of a cascading liquidation effect  in  Anchor  triggered  by  extreme  volatility  in  crypto  markets.  A large number of liquidations, where a liquidator buys bLUNA with UST can cause the price of bLUNA/LUNA to drop.

This drop in price causes the collateral value of borrowers to decrease which could cause their loans to drop below the minimum loan-to-value ratio and result in more liquidations.

Second price increase

The value of LUNA stabilised relatively quickly due to the recovery of the UST peg. The price then began to increase significantly because of increased interest and upgrades to the system. One of the upgrades resulted in a significant amount of LUNA ($832 million)being burned from the community pool, causing the value of LUNA to deflate.Another modification was to burn the LUNA that is swapped for UST, previously it went to the community pool. This means that the total supply of LUNA is decreasing over time as the amount of UST increases. These changes were all communicated far in advance and voted for by LUNA holders according to the governance rules.

Second drop

The current decline in LUNA value is related to the general trend in the cryptocurrency market due to increased regulatory and tax pressure in some of the largest markets such asChina and the US. The price of most cryptocurrencies (including LUNA) does appear to be stabilising, but a prediction of the movement of cryptocurrencies in general is out of scope for this report.

ANC vs USD price history

Notable events

Initial jump in price

The Anchor protocol immediately generated a lot of interest and there was high demand for ANC tokens when the protocol launched.This was followed by a settling period while the “true” market value was found.

First drop

The sharp drop is again related to the UST depeg caused by Anchor liquidations. The value of the ANC tokens is derived from the popularity of the system and trust was degraded by Anchor causing the depeg. This highlights a potential issue that Anchor may be too big and have too much influence on the Terra ecosystem as a whole.10It is clear from this event that a significant failure in Anchor could result in a failure of the entire Terra system if a situation arises that has not been accounted for in either protocol.

Second drop

The current drop is also related to the downturn in the global cryptocurrency market.


The UST peg and the Terra blockchain as a whole appears to be fairly fault tolerant and has recovered from some fairly significant stress tests. The risk of UST losing its peg because of a flaw in the design of Terra seems to be fairly low. The main concern is how tightly coupled Terra and Anchor are. A major fault in Anchor would likely cause another depeg of UST and the loss to the value of LUNA could cause another cascading loss. It would therefore beprudent to find insurance that covers both Anchor and a UST depeg.

Terra destaking delay risk

The process of withdrawing LUNA from the proof of stake protocol, converting bLUNA back to LUNA, takes 21 days to complete. In this time neither the bLUNA or LUNA can be traded. There was a fear that this could lead to a delayed effect where stakers will attempt to withdraw their LUNA in order to sell it off as soon as possible when the value of LUNA drops.

The amount of staked LUNA has stayed fairly constant over time, both in terms of the percentage of LUNA staked and the total number of staked tokens. Large swings in the price of LUNA do not appear to correlate with large increases or decreases of staked LUNA. This suggests that there isn’t usually a delayed effect from people destaking LUNA in order to sell it when the price drops. However, there have only been two significant drops in the price of LUNA and one is ongoing so firm conclusions cannot be drawn.

The 21 day destaking period may actually bring stability to the system as it reduces the damage that short term volatility can have on the system. Staked LUNA holders will be less prone to split second sell decisions as the market will almost certainly change during the destaking period. This does also mean that stakers will shoulder a large amount of the losses in the case of a complete collapse of Terra. In terms of the Anchor protocol, stakedLUNA is the collateral for loans and so the risk of the price of LUNA collapsing may fall on the investors.

Initially liquidators will exchange UST for bLUNA, but if the price keeps dropping the appetite for liquidation may dry up and the Anchor protocol may have to sell bLUNA at a big loss. Since the depeg, Anchor have introduced and implemented several proposals to limit cascading liquidations.

Liquidity utilisation risk

Utilisation ratio comparison

Anchor is compared to some of the most popular lending protocols, Aave, Compound and Venus. All follow a similar model where a lender deposits collateral tokens in order to take out loans of other tokens provided by lenders. Most of the protocols allow borrowing of stablecoins as well as other cryptocurrencies, but the utilisation ratio is consistently higher for stablecoins and can be used asa direct comparison for Anchor which only lends UST.

There are a few different motivations for borrowing.Firstly, it can be a general bet that the crypto market as a whole will go up.Profit can be made if the value of collateralised assets and assets bought with borrowed stablecoins both go up. Secondly, it is easier to borrow fiat currencies with stablecoins and tax can be avoided by borrowing rather than selling.

The utilisation ratio (UR) is defined as the ratio between the number of borrowed tokens and the number of tokens supplied to the pool by lenders. A high UR is desirable as it increases the interest generated, although it is beneficial for it to not be too high so investors are able to withdraw funds at will. A low UR means either low interest for investors, or increasing collateralisation ratios and borrowing costs.


Aave is a lending protocol on the Ethereum network with several pools for Ethereum based assets. As with all of these protocols, the loans must be overcollateralised and the collateral can be liquidated if the borrowed value exceeds a predefined collateralisation ratio. Aave has variable interest rates for both lenders and borrowers. A lower utilisation ratio means borrowing is cheaperand returns for lenders are lower, higher utilisation means the opposite, borrowing is more expensive and returns for lenders are higher.

The official data provided by Aave is very noisy because flash loans are possible on their system, making the plot difficult to follow. The pattern for stablecoins is that the UR is consistently above 50%for most coins meaning that over half of the deposited funds are out on loan to borrowers. The UR does not include the collateral that the depositors provide.


Compound is very similar to Aave, both are based on Ethereum and follow the same lending mechanisms. Compound has been around for longer but it doesn’t offer flash loans and typically requires more collateral for loans. Stablecoin UR on compound is much more stable as a result and is consistently above50%, with the average closer to 60-70%.


The Venus protocol is a lending and synthetic stablecoin (like UST) protocol on the Binance smart chain. Again, it requires borrowers to over collateralise their loans with other tokens. The current URs for the most popular stablecoins are:

USDT = 82%

BUSD = 80%

USDC = 76%


The Anchor protocol is described in the first interim report, it differs from the other protocols by providing a fixed interest rate to lenders subsidised by the collateral staking tokens provided by borrowers. Anchor maintained a UR of around 70% from inception up to mid May 2021, after which there was a large drop in borrowed UST and the UR fell to 20%. The UR then gradually climbed up to around 60% by the end of November but the deposited amount has been increasing at a faster rate than the borrowed amount since then. This has resulted in the UR steadily dropping down to the 25% seen today.

The trend at the moment is that the deposited amount is still increasing (aside from a recent sharp drop that can most likely be attributed to a single investor) while the borrowed amount has started to decrease.Borrowing rates are not competitive with other protocols, but have been subsidised by generous ANC incentives to keep borrowers on the system.

This will result in more financial stress on the remaining borrowers as they support the interest rate of the investors.


The underutilisation of liquidity poses the most significant risk to the long term feasibility of Anchor. Most protocols consistently provide a utilisation ratio above 50% for stablecoins and use variable borrow and lending rates to balance this. Anchor will need to adjust their lending and borrowing parameters in order to become a sustainable system.

Smart contract risk

Smart contract risk refers to bugs in the contract code that could result in a loss of some or all of the initial investment. The smart contracts that make up the Anchor protocol have been audited by Cryptonics. It’s worth noting that the official Cryptonics website generates a security error in the browser because of an expired certificate. Either they are now out of business or their credentials as security professionals should be taken with a pinch of salt.

Smart contract risk is difficult to quantify, asecurity audit will reduce that risk but most of the large exploits have occurred on systems that have been audited. A security audit should be viewed asa baseline requirement rather than proof that the contracts are secure. It is beyond the scope of this report to perform a full security audit of the Anchor smart contracts so it will have to be assumed that the Cryptonics audit was up to standard.

The audit identified 2 critical, 2 major, 7 minor and 6informational issues. All of the critical and major issues were fixed. 3 minor and2 informational issues were acknowledged but not fixed and the rest were fixed.Both Terra and Anchor have been operating for a relatively long period of time so it is likely that the smart contract risk is in line with the majority of existing DeFi protocols (AKA fairly but not extremely high).

Protocol design risk

Protocol design risk refers to purposeful features of the system that may cause a loss of value when the system experiences stress.

The two main suspect features of the Anchor protocol are how it handles an increasing gap between lent funds and borrowed funds, and how it handles a run on the market type situation where many lenders try to withdraw at once.

A key part of the anchor protocol is the yield reserve, it is designed to hold excesses of funds and make up for shortfalls to bring more stability to the platform. Anchor supplements the lending APY with the yield reserve when the Anchor rate does not match the interest charged to borrowers. The yield reserve is decreasing at a steady rate and will drop to zero within the next couple of weeks without intervention. It does seem however, that Terraform Labs plans to deposit a large amount of money ($100-300 million) in the reserve to prevent this from happening.

Without intervention the APY will be forced to drop and if it falls significantly enough it could result in a large withdrawal of funds from Earn. This will either bring the lending rate closer to the borrowing rate and stabilise the system, or, if enough liquidity is withdrawn it could trigger a run on Anchor and cause a UST depeg.

The pool should always have the funds (through collateral, unused liquidity and staking rewards) for investors to close their lending positions and collect the interest accrued. However, if investors are withdrawing their funds because they think Anchor/Terra is too high risk they will also likely want to trade their UST for USD (or a stablecoin on another blockchain).A high demand for selling UST would likely cause the peg to slip, and any investors who are too slow will lose out.

Oracle risk

Anchor is highly dependent on the Oracle that tracks the price of LUNA and bETH on other exchanges. An error in the price of LUNA could trigger liquidations. This has occurred previously (December 9th) and caused a loss of $37 million distributed over 239 borrowers. There are proposals to refund the borrowers from the community pool but the incident highlights how important an accurate Oracle is to both Anchor and Terra.

If the Oracle was to fail significantly (or was manipulated)it could trigger cascading liquidations like those seen in the May depeg. There are plans to integrate with Chainlink Oracles to spread the risk more evenly, but any import of off-chain data poses a risk because it introduces a point of trust in a trustless system.

Bridge risk

There are two bridges fromTerra to EVM based chains, the Terra Shuttle bridge and the wormhole/portal bridge. The Solana to Ethereum wormhole bridge was the subject of the recent $300+million hack. Bridges are usually very complex and pose a significant risk of hacking.

The bridges are mostly used for transferring UST to and from Terra for use in protocols on other chains. There are hundreds of millions of dollars worth of UST locked up in the two bridges and so a hack could pose a big problem for Terra. If the entire holdings of one of the bridges was lost it would almost certainly trigger a depeg, but it is hard to predict if/how Terra would recover. There are/were plans to migrate all of the funds from theShuttle bridge to the wormhole bridge, but it is too early to say if these plans have changed in light of the hack.

Regulatory risk

According to new guidelines from the UK tax regulator, if tokens put up as collateral are used by the protocol to generate income then the beneficial ownership of the tokens has been transferred. This means that the collateral would be treated as a disposal and would occur capital gains tax.Anchor would fall in this category as staking rewards are transferred to the pool, and if other regulators follow suit it would make the borrowing position even less favourable.

Coincidentally, Anchor already has a proposal to change the system so that staking rewards are still paid to the borrower and they just pay a higher total interest.

This would likely allow them to circumvent this rule.The proposal was not raised in response to regulatory issues, rather to make the onboarding of new bonded assets easier. If regulatory pressure did increase it seems almost certain that they would move to this model. The downside is that this model makes it more obvious to the borrower how bad a deal they are getting.

Insurance providers

Alternative insurance providers are compared using a number of different metrics.

Covered tokens: The tokens that will be insured. A claim will take the form of transferring the given tokens to the insurance protocol in return for a set amount of underwriting tokens. aUST are Anchor tokens and UST are the Terra USD stablecoin tokens.

Premium: The annual interest charged in order to provide cover.

Provided cover: The amount of cover provided to other users so far.

Available cover: The amount of cover available for new users.

Default  ratio:  The  value  that  the  covered  token  must  drop  below  in order for a claim to be considered. For example, a default ratio of 1 UST < 0.87 USD means that the UST must be trading at less than 0.87 USD on exchanges  before  the  insurance  protocol  can  be  triggered.  This  is designed  to  protect  against  short  term  depegs  where  the  system  can recover.

Payout ratio: The number of underwriting tokens that will be paid out for every covered token.

Risk Harbor Ozone

As described in the first report, Risk Harbor Ozone is aTerra-native insurance protocol for Anchor, providing cover for a drop in a USTvalue.

Covered tokens: aUST

Premium: Not available

Provided cover: $8.81 million

Available cover: $0

Default ratio: 1 aUST < 1 UST

Payout ratio: 1 aUST : 1.14 UST

Risk Harbor Core

Risk Harbor Core provides multi-protocol insurance and is based on Ethereum.

Covered tokens: aUST, UST

Premium: 2.15%

Provided cover: Not available

Available cover: 5.52k aUST

Default ratio: 1 aUST < 1.1 UST, 1 UST < 0.95 USD

Payout ratio: 1 aUST : 1.12 USD

Unslashed finance

Unslashed finance operates in a similar way to Risk Harbor Core, underwriters provide funds to insure many different protocols, spreading outt heir risk. When a user wishes to submit a claim  it is made visible to all users of Unslashed to view and potentially challenge, if it is unchallenged the claim will be paid out. Challenged claims are assessed through the Kleros decentralised dispute resolution protocol, where independent assessors review the claims and vote on the result.

There doesn’t appear to be any vetting process forJurors on Kleros, it operates through a simple staking mechanism, a would-bejuror buys and stakes PNK tokens and the probability of being selected as aJuror is proportional to their stake. Jurors are presented with the evidence and must vote “Yes”, “No” or “Abstain”. If a Juror votes with the majority theywill earn rewards, if they vote with the minority they will lose their stake.

Jurors are motivated to try to predict what other jurors will vote for rather than objectively examining the evidence, making this a horrible system for a real world court but hopefully fairly stable for clearcut protocol hacking/depegging insurance claims. The main risk would be if there’s a significant overlap in Unslashed underwriters and Kleros jurors and they were able to communicate and collaborate with one another, or a single party with multiple accounts stakes a significant amount of PNK when a jury is beings elected.

These risks are mitigated if the user base of Kleros is sufficiently large. However, there appear to only be around 700 active jurorson Kleros, which means that the system would be wide open to manipulation.However, It does not appear that any Unslashed claims have been challenged and sent to Kleros as of yet.

In order to cover a certain token for an amount of time, a user must pay a certain premium. Premiums and claims are paid in ETH, upon payment of the premium the user will receive policy tokens representing the amount of ETH they have covered. As the price of ETH is very variable compared to USD, it is recommended by Unslashed that users reduce or increase their insured value every two weeks.

The amount of available cover for aUST on unslashed is high because the funds are shared amongst several different protocols, but capped at a maximum share for each protocol in order to manage risk. There isn’t any easily accessible data available on previous Unslashed finance payouts, so it may be that the system is untested.

The insurance covers a UST depeg if:

  • UST is trading below 0.87 USD on trusted centralised exchanges
  • The time weighted average price for UST is below 0.87 USDover a two week span
  • The loss occurred during the policy period

The insurance covers an issue with the Anchor network if:

  • The loss is related to Anchor smart contracts
  • The loss was caused by unauthorised, malicious or criminal activity exploiting code vulnerabilities and/or due to errors or omissions in code implementation or inability to access the smart contracts
  • The loss occurred during the policy period

You are not covered if:

  • The loss is due to any malicious activity where theTerra and Anchor protocols behaved as expected (e.g. an external bridge beinghacked)
  • It is related to a vulnerability which was disclosed before the coverage period
  •  It is related to an attack vector which was disclosed in any official documentation
  • It is related to a smart contract generated for the sole purpose of submitting a claim and getting cover

Covered tokens: aUST, UST

Premium: 7.172%

Provided cover: Not available

Available cover: 3473 ETH ~ $ 9.7 million

Default ratio: 1 UST < 0.87 USD, loss related to Anchor unspecified

Payout ratio: Not available


InsureAce provides multi-pool protection and allows under writers to provide cover in multiple different tokens. The claiming process is similar to Unslashed as it is not automatic and relies on community voting.

Covered tokens: aUST, UST

Premium: 4.2%

Provided cover: Not available

Available cover: 0 aUST

Default ratio: 1 UST < 0.88 USD, Anchor loss unspecified

Payout ratio: 1 UST : 1 USD

Enzyme finance

Enzyme is an asset management system for DeFi, it aggregates protocols to spread out risk. It is effectively like being the underwriter for several protocols in the above insurance pools. In fact you could almost certainly construct matching Enzyme finance pools/vaults with the same risk profile as the insurance pools but with higher yields.

Enzyme aggregates several different strategies including yield farming, lending  and  borrowing,  derivatives,  swapping,  AMM  pools, programmable risk management and bots. It may be a lower risk strategy than investing in a single protocol but evaluating the risk associated would be incredibly complicated.

As with the insurance protocols and any other DeFi native way of managing risk it does add another potential point of failure. Most things in DeFi are unregulated and uninsured so there’s no guarantee that you could recover funds if Enzyme failed.

Ethereum protection: Futures

As the Unslashed insurance pays out in ETH and the value of ETH relative to the underlying UST asset is likely to shift significantly over any reasonable investment time, it would make sense to open a futures position on Ethereum. There are several platforms that would allow this kind of trade, such as Binance.

It might be possible to take out a policy on Unslashed for a value of ETH that matches the amount of UST invested and then take out a futures contract to sell that amount of ETH for the value of USD originally insured. If the value of ETH increases then the underlying UST is covered and there’s no need to worry. If the value of ETH decreases then the underlying UST wouldn’t be fully covered, when filing a claim it would only be possible to withdraw the amount of ETH initially covered.

If a futures contract on the ETH was taken out for that amount previously, then the claimed ETH could be traded for the amount originally covered. If a claim is not made then some or all of the investment will likely have to be withdrawn in order to fulfil the futures contract.

However, the Unslashed documentation is not clear on this point, but one would assume that they will only pay out up to the value ofUST/aUST you can prove that you own. This means that if ETH increases in price the payout will be capped at the actual value of UST rather than the originally insured amount of ETH. Without this restriction it would be easy for people to take out premiums on far higher values of underlying assets than they actually own and the system would not function. Therefore, if the value of ETH goes up then the paid out ETH will be lower than the futures contract demands and money will be lost, so this strategy won’t work.


The only viable insurance option identified so far is Unslashed finance. More research is required into the protocol but there couldbe issues with the community voting claim mechanism. Another potential issue is that coverage is given in ETH and the value of ETH is unstable, requiring constant modification of coverage.