Table of Contents:

• Why retrospective analysis is important and how it can be applied to new narratives?

• What conclusions can be drawn from the trajectory of LSDFi development after a year?

• Differences in the restaking narrative compared to the LSDFi narrative

• Brief overview of the current landscape around restaking

• Aspects of ecosystem development around Curve Finance that may be applicable to restaking and LRTFi

• Conclusion

Why is retrospective analysis important and how can it be applied to new narratives?

The most obvious answer to this question lies in the development dynamics of LSDFi and LRTFi. Projects like Lido, Rocket Pool, Stader, and others once offered users the ability to maintain liquidity for assets staked in Ethereum. For a long time, LSTs were primarily used in DEXs, lending platforms, and other DeFi protocols in fairly limited scenarios. The term LSDFi itself emerged around the 2022-2023 transition period when protocols aimed at utilizing LST began to appear. Looking at EigenLayer in this context, it resembles Ethereum – the protocol itself doesn't offer liquid tokens, so LRT protocols quickly emerged in large numbers, following a model that had already been tested with LSTs.

Thus, analyzing the development frameworks of previous major narratives and ecosystems allows us to speculate on the development vector of the direction under consideration. In particular, I examined the LSDFi industry landscape in mid-2023 ("Global overview of the LSDFi ecosystem," "A Comparative Overview of LST-Backed CDP Stablecoin Protocols"). A possible framework for assessing the potential development prospects of LSDFi was the analysis of the ecosystem development around Curve Finance ("The Saga of Curve Finance 2: Curve Ecosystem as a Template for LSDFi").

What conclusions can be drawn from the trajectory of LSDFi development after a year?

As observed in the articles mentioned above and based on extensive observations, the peak of LSDFi occurred in the spring to early summer of 2023. Almost a year later, the following observations can be confirmed:

• Early projects such as unshETH, LSDx, 0xAcid, and Agility faded within a few months due to broken tokenomics, characterized by a strong bias towards native token mining. This led to continuous declines in these tokens' value and disillusionment within their communities.

• LST-backed CDP protocols like Raft, Gravita, and eventually Lybra (many of which were based on Liquidity) faced arbitrage mechanism imbalances. Borrowers encountered situations where they had to forfeit additional profits due to changes in the stablecoin's CDP protocol exchange rate and collateral redemption value. This also contributed to community disillusionment due to creditors being trapped in their positions.

• To cover operational expenses and maintain native tokens and reserve funds, CDP-backed protocols introduced a fee on earned yield, further diminishing incentives for liquidity provision.

• Coupled with high Collateral Ratio figures, the situation unfolded as follows: the protocol advertised users could earn 8–10% yield on staked LST, even just for holding the stablecoin, which seemed enticing. However, there was a catch: users deposited, say, $2000, at a CR of 150%, they received $1333 in protocol stablecoins, and at a CR of 200%, they received $1000. Essentially, the issue was that lenders deposited, say, $2000, for which the LSD protocol provided a 3.6% yield, while the LST protocol provided stablecoins worth $1000, on which the yield was derived from $2000, i.e., 7.2%. However, in reality, when looking at the actual yield in LST-backed CDPs and LSD protocols, CDPs yielded even less profit in quantitative terms on invested funds, but they attempted to compensate for this through various mechanics involving native tokens.

• Ultimately, if we look at the current state of these protocols, we can conclude that Raft, Gravita, and Lybra haven't gained significant traction. Even Prisma Finance, with broad support and intricate liquidity pool mechanics, as well as incentives from Curve and Yearn (and according to publications and tweets, Prisma is associated with the independent research group Llama Risk), shows less than stellar Total Value Locked (TVL) figures, partly due to security issues:

• Also, it's worth considering potential token pressure factors, similar to those investigated in the article "Analysis of suspicious Lybra Finance (LBR) transactions: fraud or intended use of funds?" which could also contribute to a negative sentiment.

Differences in the restaking narrative compared to the LSDFi narrative

When EigenLayer first emerged and EigenDA wasn't yet introduced as an example of practical AVS application, it was associated with LSDFi, as it was initially seen as just another way to utilize LST. However, a few months later, it became evident that this was a push towards the evolution of the crypto industry. The key lies in the delegation of crypto-economic security, allowing for a reevaluation of project launches, where there's no longer a need for hefty bootstrap costs or maintaining incentives within the network for infrastructure decentralization.

In essence, even at a basic level, the restaking narrative offers more sustainable use cases for LST utilization compared to LSDFi, primarily represented by LST-backed CDPs. This creates demand from multiple angles:

• From the perspective of projects aiming to utilize AVS, at least 25 projects are already being launched based on EigenLayer.

• From node providers, who can potentially earn higher yields through restaking and offer more favorable conditions to their users.

• From stakers and validators, who can both delegate and directly run AVS nodes, having their own nodes.

Thus, what was lacking in the LSDFi narrative and why its decline shouldn't be projected onto restaking:

• Lack of support from major players. This sector attracted weak interest from Tier-1 VCs, unlike EigenLayer, LRTFi, and the restaking theme in general.

• In the last months of 2023 and the first months of 2024, a significant amount of liquidity in the form of LST flowed into EigenLayer and LRT, as they offered better returns and retrodrops for their native tokens.

• The LSDFi sector was primarily represented only in the EVM, and in other ecosystems, it was in its infancy.

• Despite peaking at over $1 billion in capitalization, LSDFi's capacity is relatively small — in Lybra, a significant portion of the stake was held by Sifu, and the lack of TVL fluctuations in Gravita may indicate that there are a few large depositors who continue to hold its TVL.

At the same time, analogous to the emergence of LSD and LST, infrastructure around restaking began to take shape, termed LRTFi. These protocols include Renzo, Puffer, Ether.fi, Kelp, Swell, Restake, and others. Their functionality is similar to LST built on top of ETH staking, but in this case, there's layering — they offer derivatives on liquid staking derivatives, which can also be Reward-bearing. They emerged much faster than LST and in significantly larger numbers because the action algorithms and architectures were already tested.

However, it's worth noting that at the moment, restaking faces some issues, which can currently be seen in the case of EigenLayer, but will dissipate as the restaking primitive and its ecosystem develop:

• Uncertainty about profitability due to rapid TVL growth, and potential liquidity outflow from LRT protocols by farmers after airdrops.

• Unlaunched core features such as fee systems and slashing, which cast doubt on obtaining adequate returns at present. However, this is a temporary issue that will be resolved as protocols are further launched and the ecosystem is deployed.

• A large number of LRT derivative protocols on LST, carrying their separate risks, such as peg risks of LRT with LSD and ETH. Thus, the risk of LSD peg breakdown adds a double risk with the accumulation of LRT on top. However, it's worth noting that such cases have been quite rare — 2 cases with stETH Lido due to different reasons and the associated time-wise peg breakdown of rETH Rocket Pool. The key to preventing such incidents is maintaining deep liquidity to avoid price fluctuations.

Additionally, restaking protocols in other networks have an excellent framework for building their development — precisely EigenLayer, which works very quickly in terms of attracting integrations and various collaborations. This allows for continuous expansion, attracting new users, developers, and liquidity. However, a pause or stagnation is likely possible while projects in AVS fully deploy and test the first implementations of tokenomics according to new standards.

A Brief Overview of the Current Restaking Protocol Landscape

Just a few months ago, Eigen Layer was the only player in this field, along with projects that talked about restaking but essentially offered the use of LST in liquidity pools, stakes, etc., instead of "physical" restaking. However, as of mid-May 2024, we see rapid development in various ecosystems beyond EL in Ethereum:

• Cambrian — a restaking protocol in Solana, inspired by Eigen Layer and closely working in that direction, taking into account the specifics of SVM and the Solana ecosystem. According to the whitepaper and documentation, Cambrian is immediately focused on wide customizability and flexibility, even up to using different consensus mechanisms to cover as many use cases as possible, positioning itself as decentralized web services like Amazon in the restaking world.

• Babylon Chain — offers BTC restaking, the protocol is built on Cosmos, and it introduced some interesting mechanics during its creation, such as timestamping, coordinated in Cosmos and Bitcoin, which is an interesting implementation due to the significant difference between the technologies.

• Restaking Cloud — a restaking protocol aimed at infrastructure around ETH, compatible with DVT, solo stakers, and using its denomination for LRT — Restaked Staking Tokens. With just a few clicks, anyone can create RST tokens backed by ETH staking, which have a wide range of use cases, including testing, node networks, and DeFi. Solo stakers and node operators receive income in ETH and protocol tokens.

• Persistence One — building Cosmos infrastructure for restaking on top of the Terra Alliance module, allowing users to stake Liquid Staked tokens (ATOM, TIA, DYDX, etc.) using pSTAKE, Stride, Quicksilver, and Milkyway on Persistence One. This allows them to re-stake their assets and protect additional chains, starting with the Persistence Core-1 chain, while receiving additional rewards.

• BounceBit — BTC restaking, the PoS BounceBit mechanism presents a unique dual-token staking system, using built-in BTC security fully compatible with EVM.

• Exocore — a restaking protocol based on Layer Zero technology, assuming aggregation of LST across different chains and using this delegated liquidity to launch AVS.

• Solayer — another Solana restaking protocol, which instead of AVS, uses the term SVN (Shared Validator Network).

• Additionally, in the Solana ecosystem, the restaking direction is being developed by the major staking provider Jito.

• Picasso — provides LST Solana restaking based on Cosmos technologies.

• Karak Network — here, AVS is called Distributed Secure Service (DSS), Karak positions itself as a restaking layer that makes it easy to provide crypto-economic security for any asset, but currently, it works with Ethereum.

In addition to this, numerous projects similar to LSD protocols have emerged around Eigen Layer: they offer LRT for restaking LST through them in Eigen Layer (Renzo, Puffer, Ether.fi, Kelp, Swell, Restake, and others). And some projects from LSDFi now work with LRTs based on Eigen Layer (for example, Davos and PrismaLRT). Overall, Eigen Layer itself is an excellent framework for developing restaking protocols in other chains.

Aspects of Ecosystem Development Around Curve Finance That Can Be Applicable to Restaking and LRTFi

As mentioned earlier, I also conducted research on the development of the ecosystem around Curve to understand what projects might emerge around LSDFi in an optimistic scenario of capturing a large TVL and deploying wars with liquidity and community incentives. However, the situation turned out such that the LSDFi sector gave way to the restaking narrative. And despite the entire infrastructure around Curve being built on governance and liquidity pool management with gauges and bribes, this approach in a remote sense can be applicable to LRTFi as well.

1. Firstly, it's worth noting why Curve Wars emerged: various protocols and projects fought for liquidity pool depth and attracting LPs, for which they needed Curve governance tokens and its liquid version — Convex (in this context somewhat akin to LSDFi and LRTFi wrapped tokens). Projects incentivized holders of these tokens with bribes — increased yields and native token airdrops.

2. Secondly, it might be appropriate to draw an analogy between liquidity pools and AVS and LRT protocols. In the case of LSDFi, protocols assumed that LSD protocols would be the ones competing for token holder attention and would also buy LSDFi protocol tokens. But if you're essentially a resource provider used by end-users, why would you compete for intermediaries' attention, who should themselves offer the best conditions to all sides? Thus, it's likely that the Wars scenario in LSDFi didn't develop due to misaligned priorities.

3. Thirdly, the question arises: what's different in restaking and LRTFi? The answer to this question was slightly above: in the case of restaking, there are many more stakeholders interested, all of whom have applied interest. Accordingly, there are prerequisites for competing for user liquidity from the side of LRT protocols and, in the future, from the side of AVS, which can compete to offer the best conditions. In this case, bribes can be replaced simply by token yields from specific projects launched on AVS.

Some of the important aspects that can influence the landscape of the ecosystem around restaking and LRT are the mention that one restaker can delegate to different protocols. And this suggests the creation in the future of the following protocols based on the ecosystem development trajectory around Curve with the exclusion that the original goal of Curve Wars was to promote the CRV token, whereas in restaking, LSTs of different LSD protocols can play a major role:

• Aggregators, similar to Llama Airforce — for users who delegate to multiple AVSs and who don't want to manually collect all rewards and possibly automatically reinvest further into certain AVSs (pools in the case of Llama Airforce).

• Proxy protocols, either automatically distributing delegation to different AVSs in a passive mode or offering users aggregated information to choose appropriate AVSs — here, analogies can be drawn with Votium, which essentially was the next iteration of Bribe.crv. A remote example of this could be restaked.app for EigenLayer.

• With restaking, the LRT factor comes into play — in Curve, everything was built around CRV and its wrappers. And in restaking, different LRTFi protocols can offer different conditions, including improved ones, conditioned by partnerships and so forth. Therefore, protocols similar to Llama Airforce and Votium may emerge, but for LRT, not AVS.

• Protocols similar to CLever from Aladdin DAO — for the ability to obtain loans against both current assets and their future yields. This can also include the possibility of leverage protocols operating with mechanisms similar to EigenLayer farming points — such as Gearbox and looping strategies (except with real tokens instead of points).

• Considering the development of cross-chain technologies and Account Abstraction, it's possible to see the emergence of both aggregator protocols, allowing for the distribution of LST across different chains with optimal routing and cost minimization, and automated protocols that allow this with minimal user involvement. The first example from this category could be Layerless, which allows restaking in EigenLayer through an omni-chain protocol based on LayerZero and provides users with LRT based on omni-chain token technology.

• Due to the multitude of use cases for which AVSs can be used and the emergence of restaking in different chains, there may be something akin to gaming guilds, but specializing in different technology directions based on restaking. Among such directions, ZK, General decentralized networks, MEV, financial protocols, cloud computing, DePin can be highlighted.

Thus, even with a cursory examination of possible points for developing infrastructure solutions based on retrospective analysis, we already see several potential directions. And as the entire infrastructure unfolds, other use cases may emerge that cannot be predicted until bottlenecks or, conversely, new opportunities arise.

Moreover, one of the most obvious ways to use restaking is for launching DVT, Distributed Validator Technology, which allows running a single validator on multiple servers to increase its fault tolerance, and we are likely to see projects in this direction soon. At the moment, DVT protocols have weak tokenization incentives and are mainly used as client-paid services — for example, P2P recently merged with SSV and EigenLayer to launch a product offering DVT for decentralized restaking. ClayStack is also working in this direction. But perhaps the DVT technology will spread to other networks faster with restaking, and one of the closest candidates may be Solana.

Conclusion

Unlike LSDFi in its form as it was in 2023, where profitability was essentially generated by reshuffling the difference between invested liquidity and received assets, restaking offers the opportunity to earn "real" returns from protocols that will leverage AVS. However, whether projects launching on AVS will offer genuinely good returns and adopt the philosophy of delegating economic security is another question that will take time to answer. There's unlikely to be immediate high returns for restakers: first, time is needed for the distribution of value within EL itself, then time will be needed for the launch of projects with functioning tokenomics on AVS. Only after achieving a critical mass of activity in AVS will restakers be able to earn a steady, stable profit on top of the base staking returns. Similarly, protocols for restaking will develop in parallel in other networks.

As was evident in the quick overview of the landscape, there is room in the development of the ecosystem around restaking protocols for the emergence of yield optimizers, bribes, and so on, similar to what occurred in the ecosystem around Curve. Unlike LSDFi, there is room for Wars in the direction of restaking, both between protocols on top of restaking protocol infrastructures, which may compete for liquidity, and between projects launched on AWS clusters, which may compete for operators and locked LST volume. Moreover, teams working on restaking protocols can be compared to L1 developers in 2017–2020, especially considering that restaking is not about DeFi mechanics as it was with LSDFi. This is an area that requires truly deep research in applied development, understanding, and inventing new slashing mechanisms, incentives, customization possibilities for Middleware on AVS, and so on.

Therefore, with successful development, the direction of restaking could catalyze the emergence of even more projects not only by facilitating initial launches and avoiding huge expenditures on economic security for various projects but also by the emergence of supporting infrastructure around AVS and restakers with point optimizations similar to the ecosystem around Curve.

It is also worth noting that there are now many more developers in web3, a huge number of different strategies have been tested, and there are many open-box solutions. Therefore, the speed of implementation and development has significantly increased compared to 2019–2021. A vivid example of this is the emergence of a large number of LRTs on top of EigenLayer and the rapid influx of liquidity into them compared to the slow development of LSDFi. So, it's unlikely that the ecosystem and infrastructure around restaking will develop as slowly as around Curve, but at the same time, it will be a more concrete process than the trajectory of LSDFi.

Note: This material represents the author's thoughts, subjective opinions, and conclusions based on observations and does not reflect the position of any company on this issue.

Linkedin | Twitter