Thoughts on ETH re-staking

Article author: Larry Sukernik, Myles O'Neil Article translation: Block unicorn

"If you could see yourself through my eyes" - Dermot Kennedy "Lost"

We have spent a lot of time at Reverie researching re-pledge protocols, an area that is attractive to us for investment because it is full of uncertainty (opportunities often exist in fuzzy markets) and development is very active (dozens of projects will be launched in the re-pledge space in the next 12 months).

In the process of our work on re-pledge, we have made some observations about the future development of the re-pledge market. Many situations are emerging, so today's truth may not apply tomorrow. Nevertheless, we still want to share some of our initial observations on the business dynamics of the re-pledge market.

LRT as Leverage

Today, LRTs (liquidity re-staking tokens) like Etherfi/Renzo hold a powerful position in the re-staking supply chain: due to their proximity to both the supply side (stackers) and the demand side (AVS, referring to active validation services), they are in a favorable position on both sides of the transaction. As a result, LRTs (liquidity re-staking tokens) have the ability to (i) determine their rake rate, and (ii) influence the rake rate of the underlying markets (such as EigenLayer, Symbiotic). Given their powerful position, we can expect the re-staking market to launch first-party LRTs to control the power of third-party LRTs.

AVS/Re-staking as Leverage

The best markets in the world have two characteristics: a decentralized supply side and a decentralized demand side. To understand this intuitively, it helps to reflect on the opposite case of concentration on the supply side or the demand side (or both) (AVS refers to active validators).

Imagine a Simple Market for Trading Apples

Suppose there is a simple market for trading apples, where the largest seller of apples controls more than 50% of the supply of apples. In this case, if the market operator decides to increase the market fee from 5% to 10%, the large apple sellers may threaten to take their business elsewhere.

Similarly, on the demand side, if the largest buyer of apples controls more than 50% of the demand for apples, she can also threaten to use other markets (or buy directly from apple suppliers) if the market operator increases the market fee.

Back to the re-pledge market

If the final market structure of the re-pledge market is concentrated on the AVS side (the top 10% of AVSs take more than 50% of revenue) or the re-pledger side (the top 10% of re-pledgers take more than 50% of deposits), then the natural result is that the market's ability to extract its own revenue will be weakened (and therefore the market's valuation should also be reduced).

While there is not enough data to rigorously verify this at present, our intuition is that the power law distribution will also apply here: large AVSs will take most of the total payment volume, giving them bargaining power over the market's revenue.

Fighting for exclusive AVS

From the perspective of each re-pledge market, anything that can do something that competitors can't is worth trying. The simplest way to differentiate is to provide re-stakers with access to exclusive AVS - either first-party AVS like EigenDA, or third-party AVS acquired through exclusive partnerships. This is similar in concept to Sony developing exclusive games for the PS5 to drive hardware sales.

Given these dynamics, we expect to see some action in the re-staking market, such as the launch of more first-party AVS, or exclusive agreements with third-party AVS. In short, we will see a battle for AVS in the coming months.

AVS Subsidies

AVSs need to pay operators/re-stakers a fee for the services provided, which effectively means that AVSs need to be prepared to pay out in their native tokens, ETH/USDC, or perhaps points/future airdrops. That being said, since most AVSs are currently early-stage startups without tokens, large balance sheets, or well-designed credit programs/airdrops, signing up operators/re-stakeholders can prove to be a cumbersome process (most EigenLayer partnerships are custom contracts negotiated privately).

In short, this is a situation where a customer wants to buy a service, is likely well-funded, but does not currently have sufficient funds.

To facilitate business, the re-staking market will likely make an initial payment to the operator/re-stakeholder upfront, which can be paid in their native tokens, balance sheet assets, or by issuing “cloud credits” for the AVS to spend at the operator/re-stakeholder. In return for the upfront funds, it is expected that the AVS will commit to airdropping tokens or allocating tokens to the re-staking market. Alternatively, the re-staking market can pay this money upfront to the AVS to convince it to choose itself over a competing re-staking market.

In short, we expect to see intense competition in the restaking market over the next 12-24 months by subsidizing the spending of AVS. Similar to the market dynamics of Uber and Lyft, it may be that the restaking market with the most funds/tokens will win in the end.

Personalized Guidance

Going from "I want to start an AVS" to actually launching it in production is not as simple as it seems, especially for small teams without much R&D resources. Teams need to solve problems such as how much security to buy, how long to buy security, how much fees to pay operators/restakers, what should be used to punish, and how much?

Best practices will eventually emerge, but until then, the restaking market will need to hand-hold AVS teams through these issues (it is worth noting that EigenLayer currently has no payment or punishment mechanism).

To this end, we expect the winning re-staking market to resemble an enterprise sales business, providing personalized guided integration/service assistance to customers so that they can smoothly onboard to the product.

The Evolution of AVS

The most successful AVS may gradually move away from the re-staking market and begin to use their own tokens or revenue to purchase security. Today, the incentives of the re-staking market are most important for smaller projects that do not have the time, money, brand or connections to recruit a validator set. But as projects scale, these projects may turn to recruiting validators themselves and using their own more valuable tokens to secure the network. This situation is similar to the staged development of the market, with the most successful customers gradually developing independently, and market operators need to be prepared for this.

One-Stop Shop for Crypto SaaS

To illustrate this observation, let’s first look at some software history: Cloud service providers like AWS give developers easy access to everything they need to launch an application or web service (e.g., hosting, storage, and compute). By drastically reducing the cost and time required to develop software, a more specialized class of web services emerged. Cloud service providers combine first-party cloud services with a multitude of “microservices” provided within their platforms, making them a one-stop shop for everything you need beyond your core business logic.

Restaking marketplaces like EigenLayer aim to enable a similar set of microservices for Web3. For example, before EigenLayer, crypto microservices could either fully centralize their off-chain components (and pass this risk on to their customers), or bear the cost of bootstrapping a group of operators and economic shares to purchase security.

Restaking markets could break this microservices tradeoff - if it works as intended, you’ll be able to prioritize security without sacrificing cost and speed to market.

Let’s say you’re developing a cheap, high-performance zk-rollup. If you go to a restaking market like EigenLayer, you’ll have multiple options for core services like DA and bridges to easily plug in. Through this process, you’ll see dozens of other AVS microservices that you can integrate with.

The more microservices a restaking market offers, the better the experience will be for customers - instead of having to evaluate the features and security of dozens of independent vendors, applications will be able to purchase all the services they need from one restaking market. As a result, users may come for service X, but stay for services Y and Z.

Certain AVS will have network effects (e.g., preconfs/pre-configurations)

To date, restaking use cases have been primarily focused on exporting validators and economic shares of Ethereum. But there is another class of “inwardly” focused restaking use cases that could add functionality to Ethereum consensus without requiring protocol changes.

The idea is pretty simple — allow validators to choose to make additional commitments on proposed blocks in exchange for payments, and hold them accountable through penalties if they don’t honor those commitments. We suspect that only a few of the commitment types will have enough demand to attract high levels of participation, but the amount of value flowing through these commitments could be huge.

Unlike the “external” restaking use case, the effectiveness of this class of use case is directly tied to validator participation. That is, even if you are willing to pay to be included in a block, it won’t be of much use if only 1 in 10 validators choose to honor that commitment.

However, if every validator opted into a given commitment, the guarantee behind it would be equivalent to the guarantee provided by the Ethereum protocol itself (i.e., valid blocks). Following this logic, we can expect this category to have strong network effects, as users of AVSs will benefit from every marginal validator who opts into the commitment market.

While it is still early days for this category of AVSs, a logical distribution channel to facilitate these use cases could be through Ethereum client helper tools and plugins (e.g., Reth). Similar to the proposer-builder split, it seems likely that proposers will outsource this work to specialized participants in exchange for a share of the revenue.

What is less clear is what shape these AVSs will take. While it is possible that one entity could create a general purpose market for any type of commitment, we suspect it is more likely to see the emergence of a few participants focused on sources of demand (e.g., L2 vs. L1 DeFi-driven demand for interoperability).