What would it take for ETH to succeed?

ETH’s success hinges on the resource of data availability, particularly how much it sells to L2s

article-image

Art by Crystal Le

share

This is a segment from the 0xResearch newsletter. To read full editions, subscribe.


ETH is trading at ~$1,600 today.

Can ETH pump again? Wen moon?

Getting to the heart of this simple question requires an inevitable detour into the boring, technical weeds of data availability (DA).

The ELI5 explanation of data availability is storage fees, or bandwidth resource. All chains – L1s and L2s – need them to operate, but their primary supplier is the L1.

Here’s the rough train of thought: 

  1. What is the fate of ETH as an asset?
  2. That depends on how well Ethereum can generate fees.
  3. Ethereum generates fees in three primary ways: execution, MEV, and data availability.
  4. Execution fees (aka congestion fees) have largely been given up to L2s, as per the intent of the rollup-centric roadmap.
  5. MEV fees are compressing due to better research. More and more wallets and apps are finding better ways to retain MEV and return them to users. The below chart of Ethereum’s REV (real economic value), excluding blob fees, is a rough proxy for how Ethereum’s MEV has been trending down over time.

Another way to look at the downtrend of MEV is the diminishing extracted value of sandwich attacks. This is largely due to an increasing amount of Ethereum private transaction orderflow that is not going through the public mempool.

Source: Libmev
  1. As more rollups launch, and since every chain needs DA, DA fees remain ETH’s last bastion of value accrual.

So we’ve established that data availability fees are essential to Ethereum’s success.

That is not just my opinion; it is the explicit position of leading Ethereum researchers like Justin Drake. As he puts it in the latest Ethereum Foundation Reddit AMA, DA is “the only sustainable source of flows for L1s,” not for execution or MEV fees.

(I realize this is a simplification. There are efforts to scale the L1 and to attract institutional capital to the L1, but put that aside for now.)

Now, let’s look at how much Ethereum makes from DA today.

Unfortunately, not much. Since the inception of EIP-4844 in March 2024, Ethereum has generated a total of $26m in fees.

Reaching that fee level is a function of both demand (rollups) and supply (Ethereum L1 DA capacity). 

Let’s assume Ethereum had sufficient user demand today. Even then, Ethereum lacks the supply. 

EIP-4844 introduced blobs to Ethereum’s DA capacity in March 2024. Blob capacity is about 384 KB per slot, or approximately 32 KB/s. 

For context, alternative-DA layers like Celestia have 8 MB blocks today that are about 182% larger. (Celestia is gearing up for 128 MB blocks.)

Yes, blob capacity is expanding, but it’s slow. DA capacity is set to expand from 3 blobs to 6 blobs in the Pectra hard fork next month. The eventual plan is to expand blob capacity to 48-72 in the subsequent Fusaka hard fork, but who knows when that’s going to happen.

Now, let’s look at the DA market.

More bad news. It’s a competitive market. There’s Celestia, EigenDA and Avail.

Celestia is the main alternative DA provider today, so let’s just consider them. Celestia’s DA is dirt cheap. Celestia’s co-founder, Mustafa Al-Bassam, states that this is not a competitive pricing strategy but rather a move to “achieve spam resistance” for the chain.

In any case, Celestia’s enormous price advantage means it dominates the DA market share compared to Ethereum (for now). Though it’s still early days for Celestia, it already attracts DA demand from L2s like Eclipse. In other words, less demand for Ethereum DA.

Celestia may even be competing with Ethereum on proof verification as per its “Accidental Computer” roadmap, as Delphi Digital analyst Facundo Indabera pointed out elsewhere. But that’s a whole other story.


Get the news in your inbox. Explore Blockworks newsletters:

Tags

Upcoming Events

Old Billingsgate

Mon - Wed, October 13 - 15, 2025

Blockworks’ Digital Asset Summit (DAS) will feature conversations between the builders, allocators, and legislators who will shape the trajectory of the digital asset ecosystem in the US and abroad.

Industry City | Brooklyn, NY

TUES - THURS, JUNE 24 - 26, 2025

Permissionless IV serves as the definitive gathering for crypto’s technical founders, developers, and builders to come together and create the future.If you’re ready to shape the future of crypto, Permissionless IV is where it happens.

Brooklyn, NY

SUN - MON, JUN. 22 - 23, 2025

Blockworks and Cracked Labs are teaming up for the third installment of the Permissionless Hackathon, happening June 22–23, 2025 in Brooklyn, NY. This is a 36-hour IRL builder sprint where developers, designers, and creatives ship real projects solving real problems across […]

recent research

Unlocked by Template (7).png

Research

Union’s improvements upon Tendermint consensus through CometBLS, coupled with ZK proving through Galois, allow for a broadly scalable, cost efficient, and low latency IBC implementation that is feasibly scalable across every existing blockchain, virtual machine and runtime. The implementation offers modular crosschain interoperability without the need for trusted intermediaries.  

article-image

35% of admitted teams are building AI apps, while 30% are using stablecoins

article-image

Those in the US who preregistered for the app got $150 worth of WLD

article-image

The L2 chain with opt-in privacy features was eight years in the making

article-image

Bitcoin stands on the shoulders of these Cypherpunk giants

article-image

Unto’s Will Yoo and Liam Heeger spoke to the Empire newsletter about their raise and how they plan to build Thru

article-image

Greater efficiency, William Jevons predicted, would lead to even greater consumption