Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
Free 30-min Web3 Consultation
Book Consultation
Smart Contract Security Audits
View Audit Services
Custom DeFi Protocol Development
Explore DeFi
Full-Stack Web3 dApp Development
View App Services
crypto-marketing-and-narrative-economics
Blog

Why L2 Tokenomics Are Fundamentally Flawed

An analysis of how leading Layer 2 networks like Arbitrum, Optimism, and Base have created tokens with no protocol-enforced utility, relying on governance and speculative fee-sharing promises that fail to create sustainable value.

introduction
THE PREMISE

Introduction

Current L2 tokenomics are a misaligned patchwork of incentives that fail to secure the underlying technology.

Sequencer revenue is ephemeral. Fees are paid in ETH, not the L2's native token, creating a fundamental misalignment where the token's value accrual is decoupled from the network's core economic activity.

Governance tokens lack utility. Holding OP or ARB confers no protocol rights beyond voting on treasury funds, a model that mirrors traditional corporate equity rather than crypto-native staking.

The security model is circular. L2s rely on Ethereum's consensus for finality, making their native tokens redundant for security and relegating them to subsidized liquidity mining programs.

Evidence: Over 90% of Arbitrum's DAO treasury is denominated in stablecoins and ETH, not ARB, highlighting the token's exclusion from the core fee revenue stream.

thesis-statement
THE INCENTIVE MISMATCH

The Core Flaw: Governance is Not Utility

Layer 2 tokens conflate protocol governance with network security, creating a fundamental misalignment that fails to capture value from the underlying economic activity.

Governance is not a revenue stream. Token holders vote on treasury allocations or technical upgrades, but this does not entitle them to fees generated by the sequencer. This creates a fee abstraction layer where value accrues to centralized operators or the L1, not the token.

The security model is broken. Unlike Proof-of-Stake L1s like Ethereum, where staking secures the chain, L2 staking often secures only a bridging multisig or fraud proof system. This is a narrow, low-demand security service that does not justify a multi-billion dollar valuation.

Compare Arbitrum vs. Optimism. Arbitrum's ARB token is purely for governance, while Optimism's OP partially funds public goods. Neither directly captures sequencer profits, proving the utility abstraction is a design choice, not a technical constraint.

Evidence: The sequencer capture rate is near 100%. Users pay fees in ETH, which the sequencer (often the founding team) keeps. The governance token is a spectator to this primary revenue flow, making its valuation purely speculative.

L2 TOKENOMICS

The Fee-Sharing Mirage: A Comparative Look

Comparing the economic reality of major L2 fee-sharing models against the promise of sustainable protocol revenue.

Key Metric / MechanismArbitrum (ARB)Optimism (OP)Base (No Token)Starknet (STRK)

Primary Revenue Source

Sequencer Fees

Sequencer Fees

Sequencer Fees

Sequencer Fees

% of Fees Shared with Token

~12% (via DAO grants)

~20% (via RetroPGF)

0%

~8% (proposed, not live)

Token Utility for Fee Discount

Annualized Fee Revenue (Est.)

$150M - $200M

$80M - $120M

$60M - $90M

$10M - $20M

Annualized Token Emissions (Est.)

$300M+

$250M+

N/A

$100M+

Inflation > Revenue?

N/A

Direct User Rebate Model

Sovereign Treasury (e.g., EigenLayer)

deep-dive
THE INCENTIVE MISMATCH

The Airdrop-to-Speculation Pipeline

Layer 2 tokenomics create a misaligned system where token value accrual is decoupled from core protocol utility.

Airdrops are exit liquidity. They distribute tokens to users who immediately sell, creating a permanent sell-wall that suppresses price discovery and rewards speculation over real usage.

Governance tokens lack utility. Tokens like $ARB and $OP provide voting rights on treasury funds, not protocol fee capture, making them a governance derivative rather than a productive asset.

The speculation loop dominates. Protocols like Blast and zkSync prioritize points programs for future airdrops, which incentivizes capital parking over genuine transaction volume, creating a mercenary capital problem.

Evidence: Over 60% of airdropped $ARB was sold within one month. This capital flight demonstrates that token value accrual fails when the token's only utility is governance of a non-revenue-generating DAO.

counter-argument
THE VALUE ACCRUAL FALLACY

The Bull Case (And Why It's Weak)

The dominant narrative that L2 tokens accrue value from transaction fees is structurally unsound.

Sequencer revenue is negligible. The primary source of L2 revenue is sequencer fees, but these are a fraction of the gas paid by users. On Arbitrum, sequencer revenue is less than 10% of total fees; the rest is paid to Ethereum for data and settlement. This creates a massive revenue-to-fee disparity.

Token utility is artificial. Governance rights over a decentralized sequencer set are the core proposed utility. However, protocols like Optimism and Arbitrum have centralized sequencers today. This makes the token a governance placeholder with no current cashflow rights to the underlying business.

Fee abstraction breaks the model. The rise of intent-based architectures and account abstraction (ERC-4337) means users pay fees in any token via paymasters. This severs the direct link between network usage and demand for the native L2 token, a flaw already visible in UniswapX and 1inch Fusion.

Evidence: Arbitrum's annualized sequencer revenue is ~$50M. Its fully diluted valuation is ~$15B. This implies a 300x P/S ratio, a multiple that assumes perfect, unproven value capture from future, non-existent revenue streams.

protocol-spotlight
WHY L2 TOKENOMICS ARE FUNDAMENTALLY FLAWED

Glimmers of an Alternative Future

Current L2 token models prioritize speculation over utility, creating misaligned incentives and unsustainable security. Here are the emerging alternatives.

01

The Problem: The Security Subsidy

L2s use their native token to pay for L1 data posting (e.g., Ethereum calldata). This creates a massive, perpetual subsidy where token inflation or treasury spend funds security, not user fees.

  • Economic Drag: Token value must constantly outpace L1 gas costs.
  • Misaligned Incentive: Security cost is divorced from chain usage and revenue.
  • Unsustainable: A $10B+ market cap is required to secure a fraction of that in TVL.
>90%
Subsidy Rate
$10B+
Cap Required
02

The Solution: Fee-Based Security (EIP-4844 & DA)

Shift security funding directly to users via fees, using data availability layers like EigenDA, Celestia, or Ethereum's blobs. The L2 becomes a pure sequencing engine.

  • Direct Cost Pass-Through: Users pay for the DA they consume; no token middleman.
  • Sustainable Economics: Protocol revenue is profit, not a cost center.
  • Modular Future: Enables viable sovereign rollups and high-throughput app-chains.
~100x
Cheaper Data
0%
Token Inflation
03

The Problem: Speculative Governance Tokens

L2 governance tokens like OP and ARB confer minimal real power (sequencing, upgrades are centralized) while being marketed as equity. This is a regulatory liability and value accrual mirage.

  • Empty Governance: Core protocol upgrades are executed by a multisig, not token votes.
  • No Cash Flows: Tokens don't capture protocol revenue (fees are in ETH).
  • SEC Target: The "useless utility token" model is a clear regulatory grey area.
7/8
Multisig Keys
$0
Fee Accrual
04

The Solution: Stake-for-Service & Shared Sequencing

Replace governance staking with stake-for-service models. Validators/stakers perform real work (e.g., sequencing, proving) and earn fees. Shared sequencers like Astria or Espresso commoditize this layer.

  • Real Yield: Stakers earn from transaction ordering MEV and fees, not inflation.
  • Decentralized Core: Breaks the developer multisig stranglehold on production.
  • Interoperability: Shared sequencers enable atomic cross-rollup composability.
Real
Yield Source
Atomic
Cross-Rollup
05

The Problem: The Vampire Attack Dead-End

L2s use token incentives (OP Stacks, ARB Grants) to bootstrap TVL and users. This attracts mercenary capital that leaves post-airdrop, creating no sustainable moat. It's a Ponzi-style customer acquisition cost.

  • Zero Stickiness: >60% of airdrop recipients sell immediately.
  • Cannibalistic: Incentives compete for the same DeFi TVL, driving up costs.
  • No Innovation: Capital is directed to farming, not building novel applications.
>60%
Sell-Off Rate
$B+
Wasted Incentives
06

The Solution: Native Yield & Intrinsic Utility

Build tokens with indispensable utility: as a canonical gas token, a staking asset for restaking (EigenLayer), or a unified liquidity layer. Celestia's TIA for DA or EigenLayer's AVS staking are early examples.

  • Structural Demand: Token is consumed by the protocol's core operation.
  • Value Accrual: Captures fees from a broader ecosystem, not just one chain.
  • Sustainable Growth: Bootstrapping via utility, not bribery.
Native
Gas Asset
Ecosystem
Fee Capture
future-outlook
THE INCENTIVE MISMATCH

The Path Forward: Enforced Scarcity or Irrelevance

Current L2 token models subsidize usage instead of securing the network, creating a fundamental economic flaw.

Sequencer revenue is insufficient. L2s like Arbitrum and Optimism generate fees from users, but this revenue is a fraction of the cost to secure the underlying Ethereum L1. This creates a structural subsidy where token incentives must perpetually fill the gap.

Tokens are not a fee asset. Users pay fees in ETH, not the native token. This divorces the token's utility from the network's core economic activity, relegating it to a governance token with airdrop farming as its primary use case.

The path is enforced scarcity. The only viable model is to make the token a mandatory staking asset for sequencer rights or data availability, as seen in Polygon's AggLayer or emerging shared sequencer projects like Espresso and Astria.

Evidence: Arbitrum's annualized sequencer revenue is ~$120M, while the cost to attack Ethereum via a 51% attack is orders of magnitude higher. The token must bridge this security budget gap.

takeaways
THE L2 VALUE EXTRACTION PROBLEM

TL;DR for Protocol Architects

Current L2 tokenomics are a leaky bucket, extracting value from users and sequencers while failing to secure the underlying chain.

01

The Sequencer Subsidy Trap

L2s use their native token to pay sequencers, creating a circular economy that fails to capture real value. The token's utility is artificial, propped up by inflationary rewards.\n- Key Problem: Sequencers are paid in a token they must immediately dump to cover ETH gas costs.\n- Result: Constant sell pressure and misaligned incentives, with ~90%+ of token emissions flowing straight to the L1.

>90%
Value Leak
Artificial
Demand
02

The Shared Security Fallacy

Claiming "security from Ethereum" while running a centralized sequencer is marketing, not architecture. The L2 token does not meaningfully contribute to the system's cryptographic security.\n- Key Problem: The only real security is the L1 bridge, which is secured by ETH, not the L2 token.\n- Result: Token accrues fees but provides zero marginal security, a fundamental mispricing of its value.

Zero
Marginal Security
Centralized
Sequencer
03

The Fee Market Illusion (See: Arbitrum, Optimism)

Proposals to use tokens for fee payment or burning are a weak attempt to create demand. Users will always choose the cheapest gas option, which is ETH or stablecoins.\n- Key Problem: Introducing a volatile, non-native asset as a unit of account adds friction and complexity.\n- Result: Fee mechanisms become a governance token voting game, not a genuine market-driven utility. Models like EIP-4844 blob fees further cement ETH as the base currency.

High Friction
For Users
ETH-Denominated
Real Costs
04

The Validium/Alt-DA Endgame

The only viable path for L2 token value is to secure an alternative Data Availability (DA) layer, like Celestia or EigenDA. This moves the token from a subsidy instrument to a staking asset for a real service.\n- Solution: Token stakers act as DA guarantors, with slashing for data withholding.\n- Result: Value accrual is tied to a provable, external service with real cost savings (~100x cheaper than calldata).

100x
Cost Save
Real Utility
Staking for DA
05

The Shared Sequencer Opportunity (Espresso, Astria)

Decentralizing the sequencer role creates a new market for block production that an L2 token can capture. This turns the sequencer subsidy into a competitive service market.\n- Solution: Token used for staking in a Proof-of-Stake sequencer set or for governance of the shared sequencer network.\n- Result: Token secures liveness and censorship resistance for the L2, a tangible service beyond fee abstraction.

Decentralized
Liveness
Service Market
For Blockspace
06

The Hyperliquid Thesis: Burn All Fees

The cleanest model: treat the L2 as a pure cashflow engine. All fees are used to buy and burn the native token from the open market, creating direct, deflationary value accrual.\n- Solution: No complex utility forcing. Protocol earns ETH/stables, uses revenue for open-market buybacks.\n- Result: Token becomes a pure equity claim on the L2's fee revenue, a simple and compelling value model.

Pure Equity
Value Accrual
Deflationary
Buyback Pressure
ENQUIRY

Get In Touch
today.

Our experts will offer a free quote and a 30min call to discuss your project.

NDA Protected
24h Response
Directly to Engineering Team
10+
Protocols Shipped
$20M+
TVL Overall
NDA Protected Directly to Engineering Team
Why L2 Tokenomics Are Fundamentally Flawed | ChainScore Blog