Ethereum Bets on Data Availability to Drive Future Demand

Ethereum is currently facing a common critique within the crypto industry: "too many chains, not enough apps." This issue mirrors the early days of the internet, particularly the dot-com boom of the late 90s and early 2000s, when skeptics criticized the overinvestment in infrastructure, such as fiber optics. Telecom companies spent billions laying undersea cables and long-haul fiber networks, believing that internet traffic would grow exponentially. Even internet pioneer Bob Metcalfe criticized this perceived overinvestment, arguing that there wouldn't be enough demand. However, with hindsight, we now know that Metcalfe was wrong. The surplus of bandwidth enabled the development of products like YouTube, cloud computing, and Netflix. Netflix, for instance, transitioned from a mail-based DVD rental business to a streaming service in 2007, thanks to the existing bandwidth infrastructure.

Ethereum is making a similar bet today. According to its rollup-centric roadmap, Ethereum's Layer 1 (L1) is giving up execution fees to Layer 2s (L2s). However, Ethereum plans to eventually profit from data availability (DA) fees through upcoming upgrades that will scale the supply of DA. The idea is that Ethereum's DA will become so abundant that it will induce user demand. Ethereum researcher Justin Drake believes that "if we build it, demand will come." Drake's optimistic math suggests that at 10 million transactions per second (TPS), with each transaction paying $0.001 in DA fees to Ethereum, the network could generate approximately $1 billion per day in revenue. This calculation assumes that L2s will use Ethereum L1 for DA, rather than alternative DA layers like Celestia or EigenLayer. Drake believes that the benefits of synchronous composability will make this the preferred choice.

This is a bold thesis, considering the current DA consumption on Ethereum. In March, Ethereum's blob gas fees amounted to about $272,000. While Ethereum bulls may believe that Web3 will follow a similar trajectory to Web2, Ethereum faces stiff competition from other chains. Additionally, there are still few clues about the types of on-chain applications that will dominate blockchain usage. Will the world need that much data availability? It's too early to tell, as it was in the early days of the internet. The claim that there are "too many chains" may eventually be true, but it's still uncertain. Markets are serving to curb further wasteful spending, as evidenced by the rapidly compressing L1 premium. The days of $300-$500 million raises for an L1 are over. L1/L2 blockchain raises in 2024 ranged from as low as $14 million to as high as $225 million. The high costs of running an L1 and a lack of value capture send a feedback signal to markets to stop funding these projects. In comparison, L2s are much cheaper to launch, estimated to be around $5-$15 million. This may explain the recent transition of the Celo L1 into an L2, or what Kydo, EigenLayer’s head of special projects, calls Ethereum’s “inevitable gravity toward efficiency.”