pith. sign in

arxiv: 2606.13555 · v1 · pith:FLRQ67SOnew · submitted 2026-06-11 · 💰 econ.EM · cs.GT

Price Elasticity of Gas Demand on L1 and L2: Evidence from Ethereum and Arbitrum

Pranay Anchuri , Akaki Mamageishvili This is my paper

Pith reviewed 2026-06-27 05:01 UTC · model grok-4.3

classification 💰 econ.EM cs.GT
keywords price elasticity of demandgas feesEthereumArbitruminstrumental variablesblockchaindemand estimationfee mechanisms
0
0 comments X

The pith

Instrumented with lagged base fees, gas demand elasticities are -0.006 on Ethereum and -0.036 on Arbitrum

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The authors estimate the causal price elasticity of gas demand on Ethereum mainnet and Arbitrum using a two-way fixed effects panel regression. They instrument with each wallet's lagged base fee to address endogeneity from congestion that biases naive estimates. Results indicate near-inelastic demand overall, with a 10% fee increase reducing demand by only 0.06% on Ethereum and 0.36% on Arbitrum. L2 demand is more responsive than L1, and decompositions show variation by resource type and wallet behavior clusters. These estimates support better calibration of fee mechanism simulations and evaluation of pricing reforms.

Core claim

Using two-way fixed effects panel regression instrumented by each wallet's own lagged base fee removes the congestion-driven endogeneity that causes naive regressions to substantially underestimate demand sensitivity. On Ethereum mainnet over the full year 2025, the pooled IV elasticity is -0.006, near-inelastic. On Arbitrum One from October 2025 to April 2026, it is -0.036. Both chains exhibit inelastic aggregate demand, though L2 is measurably more responsive, with further heterogeneity across resources and behavioral clusters.

What carries the argument

Two-way fixed effects panel regression instrumented by each wallet's own lagged base fee, which removes endogeneity from current congestion.

Load-bearing premise

The lagged base fee is a valid instrument, relevant for current base fee but uncorrelated with current demand shocks after accounting for wallet and time fixed effects.

What would settle it

Observing a significant correlation between the lagged base fee and current demand shocks after including the two-way fixed effects, or an exogenous fee change producing demand response far from the estimated elasticity values.

read the original abstract

We estimate the causal price elasticity of gas demand on Ethereum mainnet (L1) and Arbitrum One (L2), a quantity necessary for calibrating fee mechanism simulations, evaluating resource pricing reforms, and explaining observed usage patterns. A two-way fixed effects panel regression instrumented by each wallet's own lagged base fee removes the congestion-driven endogeneity that causes naive regressions to substantially underestimate demand sensitivity. On Ethereum mainnet (full year 2025), the pooled IV elasticity is -0.006***, near-inelastic: a 10% fee increase reduces total gas demand by approximately 0.06%. On Arbitrum One (October 2025--April 2026), the pooled IV elasticity is -0.036**. Both chains are inelastic in the aggregate, with L2 measurably more responsive than L1. A per-resource decomposition of L2 demand reveals elasticities ranging from modestly elastic computation (-0.027*) to -0.27*** for refunds, with storage growth (-0.15***) and calldata (-0.06*) in between. Behavioral clustering identifies always-on protocol wallets as near-inelastic and high-volume operators as substantially more responsive, with cluster-level elasticities up to roughly 6x the pooled estimate. These results establish an empirical foundation for downstream simulations and for evaluating fee mechanism designs.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper estimates the causal price elasticity of gas demand on Ethereum mainnet (L1) and Arbitrum One (L2) via two-way fixed effects panel IV regressions. Each wallet's own lagged base fee serves as the instrument for current base fee to address congestion-driven endogeneity. Pooled IV estimates are -0.006*** on Ethereum (full year 2025) and -0.036** on Arbitrum (Oct 2025–Apr 2026), indicating aggregate inelasticity, with L2 more responsive; further per-resource and behavioral-cluster decompositions are reported.

Significance. If the causal claims hold, the estimates supply an empirical foundation for calibrating fee-mechanism simulations and evaluating resource-pricing reforms on L1/L2 chains. The wallet-level panel design and decomposition by resource type and behavioral clusters are strengths that allow heterogeneity analysis beyond aggregate OLS.

major comments (1)
  1. [IV strategy and identification discussion] The central identification assumption—that each wallet's lagged base fee is uncorrelated with current-period demand shocks conditional on wallet and time fixed effects—is load-bearing for the causal interpretation but receives no supporting diagnostics in the manuscript. Because base fees evolve deterministically from prior aggregate gas usage (EIP-1559) and many wallets exhibit serial correlation in activity, the exclusion restriction may fail, biasing the IV elasticity toward zero in the direction of the reported near-inelastic results. First-stage F-statistics, tests for residual autocorrelation, or robustness to alternative lag structures or placebo instruments would be required to substantiate the claim.
minor comments (2)
  1. [Results] The abstract and results section would benefit from explicit reporting of first-stage coefficients, F-statistics, and the exact lag length used for the instrument.
  2. [L2 decomposition] The per-resource elasticities on Arbitrum (computation, storage, calldata, refunds) are presented with varying significance levels; a table or figure showing economic magnitudes (e.g., implied gas-use change for a 10% fee increase) would improve interpretability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful and constructive comments, which highlight an important aspect of our identification strategy. We address the concern below and will revise the manuscript to incorporate the suggested diagnostics.

read point-by-point responses

  1. Referee: [IV strategy and identification discussion] The central identification assumption—that each wallet's own lagged base fee is uncorrelated with current-period demand shocks conditional on wallet and time fixed effects—is load-bearing for the causal interpretation but receives no supporting diagnostics in the manuscript. Because base fees evolve deterministically from prior aggregate gas usage (EIP-1559) and many wallets exhibit serial correlation in activity, the exclusion restriction may fail, biasing the IV elasticity toward zero in the direction of the reported near-inelastic results. First-stage F-statistics, tests for residual autocorrelation, or robustness to alternative lag structures or placebo instruments would be required to substantiate the claim.

    Authors: We agree that the manuscript as currently written does not report first-stage F-statistics, residual autocorrelation tests, or robustness checks with alternative instruments or lags. The two-way fixed effects are designed to absorb wallet-specific heterogeneity and common time-varying shocks, while the wallet-specific lagged base fee is intended to isolate variation in fees driven by prior aggregate congestion rather than contemporaneous demand. Nevertheless, we recognize that serial correlation in wallet-level activity could in principle create correlation between the instrument and current demand shocks through the deterministic EIP-1559 update rule. In the revised version we will add the requested diagnostics, including first-stage F-statistics, tests for residual autocorrelation, and checks using alternative lag lengths or placebo instruments constructed from unrelated network variables. These additions will allow a direct assessment of instrument strength and exclusion-restriction plausibility. revision: yes

Circularity Check

0 steps flagged

No circularity: standard IV estimation on observed data

full rationale

The derivation consists of a two-way fixed effects IV regression of gas demand on base fee, instrumented by each wallet's lagged base fee. The resulting elasticities (-0.006 on L1, -0.036 on L2) are direct outputs of this regression applied to transaction data; they are not obtained by fitting a parameter to a subset and relabeling it a prediction, nor by any self-definitional loop, self-citation chain, or imported uniqueness theorem. The identification assumption (lagged fee ⊥ demand shock | FE) is an external empirical claim, not a reduction to the paper's own equations. No steps match the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard econometric identification assumptions for instrumental variables in panel data; no free parameters or invented entities are introduced beyond the regression specification.

axioms (1)
  • domain assumption The lagged base fee is a valid instrument for the current base fee after including two-way fixed effects, satisfying relevance and exclusion restrictions.
    Invoked in the abstract to remove congestion-driven endogeneity in the demand estimation.

pith-pipeline@v0.9.1-grok · 5808 in / 1336 out tokens · 42077 ms · 2026-06-27T05:01:21.131413+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

40 extracted references · 7 canonical work pages

  1. [1]

    Cryptoeconomic Systems , year =

    Tim Roughgarden , title =. Cryptoeconomic Systems , year =

  2. [2]

    Proceedings of the 22nd ACM Conference on Economics and Computation (EC 2021) , year =

    Tim Roughgarden , title =. Proceedings of the 22nd ACM Conference on Economics and Computation (EC 2021) , year =

    2021

  3. [3]

    Ferreira and Daniel J

    Matheus V.X. Ferreira and Daniel J. Moroz and David C. Parkes and Mitchell Stern , title =. Proceedings of the 3rd ACM Conference on Advances in Financial Technologies , year =

  4. [4]

    Proceedings of the 2023 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA 2023) , year =

    Hao Chung and Elaine Shi , title =. Proceedings of the 2023 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA 2023) , year =

    2023

  5. [5]

    CoRR , volume =

    Stefanos Leonardos and Barnab. CoRR , volume =. 2021 , url =

    2021

  6. [6]

    2021 IEEE International Conference on Blockchain (Blockchain) , year =

    Danijar Reijsbergen and Stefanos Leonardos and Barnab. 2021 IEEE International Conference on Blockchain (Blockchain) , year =

    2021

  7. [7]

    Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security , year =

    Yulin Liu and Yuxuan Lu and Kartik Nayak and Fan Zhang and Luyao Zhang and Yinhong Zhao , title =. Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security , year =

    2022

  8. [8]

    2026 , url =

    Maria Silva , title =. 2026 , url =

    2026

  9. [9]

    Economic Inquiry , volume =

    Anil Donmez and Alexander Karaivanov , title =. Economic Inquiry , volume =. 2022 , doi =

    2022

  10. [10]

    2nd International Workshop on Emerging Trends in Software Engineering for Blockchain (WETSEB) , pages =

    Giuseppe Antonio Pierro and Henrique Rocha , title =. 2nd International Workshop on Emerging Trends in Software Engineering for Blockchain (WETSEB) , pages =. 2019 , doi =

    2019

  11. [11]

    2022 , url =

    Barnab. 2022 , url =

    2022

  12. [12]

    2023 , url =

    Davide Crapis , title =. 2023 , url =

    2023

  13. [13]

    27th USENIX Security Symposium (USENIX Security 18) , year =

    Harry Kalodner and Steven Goldfeder and Xiaoqi Chen and Matt Weinberg and Edward Felten , title =. 27th USENIX Security Symposium (USENIX Security 18) , year =

  14. [14]

    Felten , title =

    Akaki Mamageishvili and Edward W. Felten , title =. Financial Cryptography and Data Security (FC) 2023 Workshops , series =

    2023

  15. [15]

    Felten and Akaki Mamageishvili , title =

    Davide Crapis and Edward W. Felten and Akaki Mamageishvili , title =. Financial Cryptography and Data Security (FC) 2024 Workshops , series =. 2024 , url =

    2024

  16. [16]

    CoRR , volume =

    Seongwan Park and Bosul Mun and Seungyun Lee and Woojin Jeong and Jaewook Lee and Hyeonsang Eom and Huisu Jang , title =. CoRR , volume =. 2024 , url =

    2024

  17. [17]

    Financial Cryptography and Data Security (FC) 2025 , series =

    Lioba Heimbach and Jason Milionis , title =. Financial Cryptography and Data Security (FC) 2025 , series =. 2025 , url =

    2025

  18. [18]

    5th Conference on Advances in Financial Technologies (AFT 2023) , series =

    Theo Diamandis and Alex Evans and Tarun Chitra and Guillermo Angeris , title =. 5th Conference on Advances in Financial Technologies (AFT 2023) , series =. 2023 , url =

    2023

  19. [19]

    Moallemi , title =

    Guillermo Angeris and Theo Diamandis and Ciamac C. Moallemi , title =. 7th Conference on Advances in Financial Technologies (AFT 2025) , series =. 2025 , url =

    2025

  20. [20]

    Financial Cryptography and Data Security (FC) 2020 , series =

    Friedhelm Victor , title =. Financial Cryptography and Data Security (FC) 2020 , series =

    2020

  21. [21]

    IEEE International Conference on Decentralized Applications and Infrastructures (DAPPS) , pages =

    Ferenc B. IEEE International Conference on Decentralized Applications and Infrastructures (DAPPS) , pages =. 2021 , url =

    2021

  22. [22]

    CoRR , volume =

    Hanzhi Hu and Tom Beer and Daniel Perez , title =. CoRR , volume =. 2024 , url =

    2024

  23. [23]

    The Energy Journal , volume =

    Severin Borenstein , title =. The Energy Journal , volume =

  24. [24]

    AEA Papers and Proceedings , volume =

    Natalia Fabra and David Rapson and Mar Reguant and Jingyuan Wang , title =. AEA Papers and Proceedings , volume =. 2021 , doi =

    2021

  25. [25]

    Energy Economics , year =

    Silvana Tiedemann and Raffaele Sgarlato and Lion Hirth , title =. Energy Economics , year =

  26. [26]

    Econometrica , volume =

    Partial Time Regressions as Compared with Individual Trends , author =. Econometrica , volume =. 1933 , doi =

    1933

  27. [27]

    Journal of the American Statistical Association , volume =

    Seasonal Adjustment of Economic Time Series and Multiple Regression Analysis , author =. Journal of the American Statistical Association , volume =. 1963 , doi =

    1963

  28. [28]

    Angrist and J

    Joshua D. Angrist and J

  29. [29]

    Wooldridge , title =

    Jeffrey M. Wooldridge , title =

  30. [30]

    Angrist and Alan B

    Joshua D. Angrist and Alan B. Krueger , title =. The Quarterly Journal of Economics , volume =. 1991 , doi =

    1991

  31. [31]

    Stock , title =

    Douglas Staiger and James H. Stock , title =. Econometrica , volume =. 1997 , doi =

    1997

  32. [32]

    2019 , url =

    Vitalik Buterin and Eric Conner and Rick Dudley and Matthew Slipper and Ian Norden and Abdelhamid Bakhta , title =. 2019 , url =

    2019

  33. [33]

    Myerson , title =

    Roger B. Myerson , title =. Mathematics of Operations Research , volume =

  34. [34]

    Pai and Max Resnick , title =

    Nir Lavee and Noam Nisan and Mallesh M. Pai and Max Resnick , title =. CoRR , volume =. 2025 , url =. doi:10.48550/ARXIV.2504.15438 , eprinttype =. 2504.15438 , timestamp =

    work page doi:10.48550/arxiv.2504.15438 2025

  35. [35]

    7th Conference on Advances in Financial Technologies (AFT 2025) , pages =

    Ozan Solmaz and Lioba Heimbach and Yann Vonlanthen and Roger Wattenhofer , editor =. Optimistic. 7th Conference on Advances in Financial Technologies,. 2025 , url =. doi:10.4230/LIPICS.AFT.2025.28 , timestamp =

    work page doi:10.4230/lipics.aft.2025.28 2025

  36. [36]

    CoRR , volume =

    Wenhao Wang and Aditya Saraf and Lioba Heimbach and Kushal Babel and Fan Zhang , title =. CoRR , volume =. 2026 , url =. doi:10.48550/ARXIV.2604.00234 , eprinttype =. 2604.00234 , timestamp =

    work page doi:10.48550/arxiv.2604.00234 2026

  37. [37]

    CoRR , volume =

    Bruno Mazorra and Christoph Schlegel and Akaki Mamageishvili , title =. CoRR , volume =. 2026 , url =. doi:10.48550/ARXIV.2602.22032 , eprinttype =. 2602.22032 , timestamp =

    work page doi:10.48550/arxiv.2602.22032 2026

  38. [38]

    CoRR , volume =

    Aggelos Kiayias and Elias Koutsoupias and Giorgos Panagiotakos and Kyriaki Zioga , title =. CoRR , volume =. 2025 , url =. doi:10.48550/ARXIV.2506.13271 , eprinttype =. 2506.13271 , timestamp =

    work page doi:10.48550/arxiv.2506.13271 2025

  39. [39]

    Moallemi and Shouqiao Wang , editor =

    Davide Crapis and Ciamac C. Moallemi and Shouqiao Wang , editor =. Optimal Dynamic Fees for Blockchain Resources , booktitle =. 2024 , url =. doi:10.1007/978-3-031-78676-1\_15 , timestamp =

    work page doi:10.1007/978-3-031-78676-1 2024

  40. [40]

    Moallemi , editor =

    Shouqiao Wang and Davide Crapis and Ciamac C. Moallemi , editor =. A Framework for Combined Transaction Posting and Pricing for Layer 2 Blockchains , booktitle =. 2025 , url =. doi:10.1007/978-3-032-07024-1\_4 , timestamp =

    work page doi:10.1007/978-3-032-07024-1 2025