## Reassessing Restaking Risks - ### How Institutional Stakers and Node Operators Can Evaluate AVSs and LRT Daemon Day @TOKEN2049 Singapore - September 2024 --- ## Freddy Zwanzger ### Ethereum Ecosystem Lead @ Blockdaemon ● Co-Founder & Chief Data Officer (2018 – 2021) at **Anyblock Analytics GmbH** → acquired by **Blockdaemon** (since Dec 2021) --- # What to expect 1. Why are we here? - State of Restaking 2. What is Restaking? - Brief Recap 3. Stacking of Considerations & Risks 4. Comparing to Bonds 5. Takeaways --- ## Why are we here? - State of Restaking Quickly developing narrative! → in crypto historically means to better proceed with caution… | Protocols | Chain | TVL | | --- | --- | --- | | Live on Ethereum | | | | EigenLayer | Ethereum | $17.645 B | | Karak | Ethereum | $915.49 M | | Symbiotic | Ethereum | $316.5 M | | Live on Other Ecosystems | | | | Solayer | Solana | $74.11 M | | Merlin | Bitcoin | - | | Pell Network | Bitcoin | $133.84 M | | Allstake | Solana&NEAR | $12.72 M | | Parasail | Filecoin | $60.46 M | | Octopus Network | NEAR | $2.84 M | --- | State of Restaking - Room to Grow | | | --- | --- | | STAKING | RESTAKING | LIQUID STAKING | LIQUID RESTAKING | | ~28% $ETH is in STAKING of which ~41% is in LIQUID STAKING ~16% is in RESTAKING ~12% is in LIQUID RESTAKING | | | Data as of Sept 2024, sources: | Ultrasound Money, Beaconcha.in, Blocklytics on Dune, DefiLlama (1, 2) | --- ## What is Restaking? - Brief Recap Restaking generally means **reusing** **staked collateral** to provide economic security to projects ● *for stakers:* additional reward opportunities for an increased level of risk ● *for AVSs (Actively Validated Services):* “security as a service” tailored to their needs **Note:** All restaking protocols now *allow many ERC-20 tokens* beyond ETH/LSTs as collateral. This can add utility to tokens of secured protocols and actually outsource their security as staking-as-a-service (so not just RE-staking anymore). --- ## Stacking of Considerations ● Protocol Selection (continuous) ● + Asset Selection (continuous) ● + AVS Portfolio Selection (cont.) ● Node Operator Selection (1x) --- ## Risks in (Re)Staking 01 **Slashing** (maliciously or stupid) 02 **Downtime** (offline penalty) 03 **Liquidity** (exit/withdrawal queue, peg/depth for LSTs/LRTs) ## Correlations are the real danger! ● Staking: **client bugs** ● LST / LRT / Restaking platforms: **smart contract bugs** ● AVSs: **correlated slashings** --- ## Modeling of Restaking Risks is Evolving But early and mostly highly mathematical… --- ## Comparing to Bonds ### Non-liquid vs. liquid positions --- ## Risk vs. Return of Restaking ● Restaking is new -> risk will be higher than potential return at the moment ● 6-12 months should see: ○ Risk reduced as software matures ○ Regulation is made clearer ○ Rewards begin to be earned in full on networks - just as we saw with traditional ETH staking --- # Staking options for institutions as of now: | Staking | $ eq$~3.5% native ETH rewards $ eq$Lowest risk (“Internet bond”) | AVAILABLE NOW | | --- | --- | --- | | Liquid Staking | $ eq$Liquidity (swap in/out without delay) $ eq$Utility of LST (leverage, Defi, ...) $ eq$No 32 ETH increments needed* | AVAILABLE NOW (Permissioned for Institutions: LIQUID COLLECTIVE) | | Restaking | $ eq$Developing opportunity for additional rewards (vs. risks) $ eq$Delegate to Blockdaemon or dedicated white-label set of AVSs | Ecosystem developing | | Liquid Restaking | $ eq$Institutions need to understand risks and rewards | Ecosystem developing | ## APPENDIX / BACKUP --- ## Stacking of Considerations: Checklist ### Staking - Liquid Staking - Restaking - Liquid Restaking ● Is the node operator a known and trusted entity, with reputation at stake? ● What is the node operator’s track record of performance? ● How ‘good’ of a job do they do operating infrastructure on the network, and receiving network rewards for doing so? ● What kinds of risk mitigations does the node operator have in place, to reduce the risk of slashing, mitigate operational or security risks, etc.? ## Liquid Staking - Considerations of the LSP include: ● Are the liquid staking protocol’s smart contracts audited? Has the code deployed to mainnet received security reviews and/or code audits? ● How long does deposited ETH pass through the protocol before it is programmatically staked via Ethereum’s deposit contract? ● Has the liquid staking protocol’s code been made public, so that external security researchers and participants can evaluate it? ## Restaking - Considerations of the AVSs and operators include: ● How long has the LSP been active without a meaningful security incident or loss of staked funds? ● What are the slashing risks and conditions associated with the AVS? Under what circumstances can staked tokens be slashed, due to the behavior of the restaking operators, etc? ● As with staking and liquid staking, are the AVS operators reputable and experienced? ● Do they have slashing mitigations in place that meet the slashing considerations of the AVS? ● What is the economic model of the AVS, and how does the AVS and the operators being delegated to manage economic incentives? ● Does the AVS have any governance mechanism? ● Does the AVS introduce any external risks due to the services it is offering, such as additional attack vectors or liquidation risks? ## Liquid Restaking - Considerations of the LRTs include: ● All of the above, plus ● Diligence process of selecting and monitoring operators and AVSs (e.g. correlations, risk-reward trade-offs, liquidity) ● Compliance considerations --- | Attribute | Sovereign bonds | | L1 staking positions | | AVS restaking positions | | --- | --- | --- | --- | --- | --- | | Liquidity/Leverage | Illiquid as individual assets, not easy to borrow against | | | | | | Yield | Source: Sovereign treasury Denomination: Sovereign currency | | Source: L1 protocol Denomination: L1 token | | Source: AVS protocol Denomination: AVS specified | | Duration | Fixed maturity | | Withdrawals rate limited by the L1 protocol | | Withdrawals rate limited by the AVS, the restaking protocol, and the L1 protocol | | Default | No default because the Sovereign controls the money supply | | L1 slashing where some or all of principal can be destroyed | | Both AVS slashing and L1 slashing where some or all of the principal can be destroyed | | Portfolio construction | Single asset portfolio | | | | | | Attribute | Bond funds | LSTs | | LRTs | | Liquidity/Leverage | Liquid because of fungibility, easy to borrow against | | | | | Yield | Source: Sovereign treasury Denomination: Sov. currency Aggregation over: Many interest rates and durations | Source: L1 protocol Denomination: L1 token Aggregation over: Many node operators | Source: AVS protocol Denomination: AVS specified Aggregation over: Many AVSs and node operators | | | Duration | Many different maturities | Withdrawals rate limited by the LST protocol and the L1 protocol | | Withdrawals rate limited by the LRT protocol, the restaking protocol, each underlying AVS, and the L1 protocol | | Default | Duration mismatch (e.g., SVB collapse), or illiquid redemption market (e.g., Sept.2019 Repo Crash) | One/some of the L1 node operators getting slashed | | One/some of the AVS node operators or one/some of the L1 node operators getting slashed | | Portfolio construction | Low complexity: balancing many low risk assets | Medium complexity: selecting L1 node operators and/or collateral ratios | | High complexity: choosing AVSs, AVS node operators, different interest denominations, L1 node operators and/or collateral |