Explaining Sustained Blockchain Decentralization with Quasi-Experiments: The Resource Flexibility of Consensus Mechanisms

Harang Ju, Madhav Kumar, Ehsan Valavi, Sinan Aral, Johns Hopkins Carey Business School, [email protected], Harvard Business School, [email protected], MIT Sloan School of Management, {evalavi, sinan}@mit.edu · 2025

Read Explaining Sustained Blockchain Decentralization with Quasi-Experiments: The Resource Flexibility of Consensus Mechanisms by Harang Ju, Madhav Kumar, Ehsan Valavi, Sinan Aral, Johns Hopkins Carey Business School, [email protected], Harvard Business School, [email protected], MIT Sloan School of Management, {evalavi, sinan}@mit.edu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-24663.txt.

Introduction

Blockchains promise that no single party runs the system, yet some chains drift toward control by a few. Harang Ju and co-authors ask what keeps a chain spread out in this 2025 paper. Their guess centers on how flexible the chain's agreement method is about the resources it needs. A method that can swap one resource for another may survive shocks that would shrink a rigid one.

The team used three natural shocks as tests: a policy shift, an infrastructure event, and a technical change. Each shock hit the resources used to reach agreement on some chains. Comparing chains before and after gave a near-experiment. The results gave strong hints that flexible methods hold their spread better. The work speaks to designers who want a chain to stay open over time.

Catalog record

CREATORHarang Ju, Madhav Kumar, Ehsan Valavi, Sinan Aral, Johns Hopkins Carey Business School, [email protected], Harvard Business School, [email protected], MIT Sloan School of Management, {evalavi, sinan}@mit.edu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE125 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-24663

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Explaining Sustained Blockchain Decentralization with Quasi-Experiments: The Resource Flexibility of Consensus Mechanisms Harang Ju, Madhav Kumar, Ehsan Valavi, Sinan Aral, Johns Hopkins Carey Business School, [email protected], Harvard Business School, [email protected], MIT Sloan School of Management, {evalavi, sinan}@mit.edu Abstract Decentralization is a fundamental design element of the Web3 economy. Blockchains and distributed consensus mechanisms are touted as fault-tolerant, attack-resistant, and collusion-proof because they are decentralized. Recent analyses, however, find some blockchains are decentralized, others are centralized, and that there are trends towards both centralization and decentralization in the blockchain economy. Despite the importance and variability of decentralization across blockchains, we still know little about what enables or constrains blockchain decentralization. We hypothesize that the resource flexibility of consensus mechanisms is a key enabler of the sustained decentralization of blockchain networks. We test this hypothesis using three quasi-experimental shocks (policy-related, infrastructure-related, and technical) to resources used in consensus. We find strong suggestive evidence that the resource flexibility of consensus mechanisms enables sustained blockchain decentralization and discuss the implications for the design, regulation, and implementation of blockchains. 1 Introduction Blockchains offer an alternative to centralized systems through a model of trust built on distributed consensus ( Ammous, 2018 ) . While they were originally conceived as a time-stamping mechanism to prevent back- or forward-dating digital documents ( Haber and Stornetta, 1991 ) , in their modern incarnation, blockchain implementations are designed to remove centralized institutions from the process of tracking electronic cash ( Nakamoto, 2008 ) and to execute generalized software programs without intermediaries ( Buterin, 2014 ) . Decentralization is fundamental to the ability of blockchains to operate without centralized intermediaries while maintaining resistance to attacks, faults, and collusion ( Buterin, 2017 ) . Blockchain decentralization is multidimensional, and different dimensions may benefit from different levels of centralization ( Halaburda and Mueller-Bloch, 2020 ; Hsieh and Vergne, 2023 ) : while centralization can improve coordination in platform governance and system development ( Chen et al., 2021 ) , decentralization in the consensus layer is critical for security and trust. Indeed, failure to sustain decentralization in blockchain consensus mechanisms can cause catastrophic failures that lead to the loss of billions of dollars of economic value and data integrity regarding asset ownership. For example, the Ronin blockchain was hacked in 2022 for $625 million because of centralized points of failure ( Thurman, 2022 ) and 51% attacks occur frequently and have significant and lasting negative effects on token prices ( Shanaev et al., 2019 ) . Since 2009, with blockchains growing rapidly and cryptocurrencies surpassing a 3.7 trillion-dollar market capitalization, 1 1 1 CoinMarketCap Accessed February 18, 2025. and with institutions, including governments, increasingly tokenizing hundreds of billions of dollars of bonds and other assets on blockchains, understanding and sustaining decentralization is becoming increasingly critical ( Bank, 2021 ; Securitize, 2024 ) . While many tout the decentralized nature of blockchain technology, it is still unclear whether blockchains are actually becoming more centralized or decentralized over time and, perhaps more importantly, what drives blockchain centralization or decentralization. Indeed, decentralization faces countervailing economic forces, including economies of scale and coordination costs ( Hui and Tucker, 2023 ) . It also faces economic trade-offs with blockchain’s other goals, such as net neutrality and permissionlessness ( Halaburda and Obermeier, 2024 ; Bakos et al., 2021 ) . Recent empirical work also estimates that several key subsystems of blockchain ecosystems are becoming more centralized ( Ju et al., 2025 ) . In short, decentralization cannot be taken for granted; it must be actively sustained against countervailing economic and technical forces. In this paper, we explore potential enablers of sustained blockchain decentralization in consensus mechanisms, a key element of this technology critical to establishing trust and fault tolerance without centralization. While prior work has identified economies of scale ( Capponi et al., 2023 ) , governance structures ( Chen et al., 2021 ) , and technological innovations ( Cong et al., 2023 ) as factors influencing decentralization, the role of the resources required for consensus participation, and specifically their flexibility, has not been empirically examined. We hypothesize that the flexibility of resources required for consensus mechanisms is an important enabler of sustained blockchain decentralization. Resource flexibility refers to the ease with which participants can acquire, deploy, and redeploy the resources needed for consensus, from portable hardware like GPUs to transferable staked tokens (see Section 2 for a full development of this concept). Many blockchains require the use of costly resources to verify transactions for proof-based consensus mechanisms, including hardware and electricity for Proof-of-Work (PoW) and native blockchain tokens for Proof-of-Stake (PoS). Empirical examination of the implications of technical resource flexibility on sustained blockchain decentralization can therefore inform design decisions in the pursuit of decentralization at the heart of blockchain technologies. 2 2 2 The Proof-of-Authority consensus mechanism, which relies on a small number of validators selected by a central authority, is an exception to this requirement, though its inclusion in the validator set, determined by centralized authorities, is also a costly resource. We leverage quasi-experimental variation across six blockchains ( i.e. , Bitcoin, Ethereum, Solana, Gnosis, BNB, Ronin) to investigate how resource flexibility, proxied by differences in blockchain designs, influences blockchain decentralization in the context of three independent shocks: a policy-related shock (China’s crypto mining ban on May 15, 2021), an infrastructure-related shock (Hetzner’s shutdown of Solana validators on November 2, 2022), and a technical shock (the Ethereum Merge on September 15, 2022). China’s ban shows the impact of resource flexibility ( i.e. , ASICs vs. GPUs) in response to a policy shock; Hetzner’s shutdown contrasts with China’s ban by revealing the impact of an even more flexible resource ( i.e. , tokens in PoS vs. hardware in PoW) in response to an infrastructure shock; and the Ethereum Merge assesses the effect of increased resource flexibility as Ethereum upgraded from PoW to PoS. To measure these effects, we employ event studies, difference-in-difference (DiD) estimation, and synthetic DiD (SDiD) methods and use five different measures of decentralization: Shannon entropy, the number of validating nodes, the Gini coefficient, the Nakamoto coefficient, and the Herfindahl–Hirschman Index. Our analysis provides strong evidence suggesting that the resource flexibility of distributed consensus mechanisms enables sustained blockchain decentralization. We find that China’s ban reduced Bitcoin’s decentralization in comparison to Ethereum’s by around 0.21 bits of Shannon entropy from around 3.7 bits, that Hetzner’s shutdown reduced Solana’s decentralization by around 0.38 bits and that the dynamics of the recovery from the initial shocks varied widely across the two shocks. Bitcoin’s decentralization took over 6 months to recover from China’s ban, while Solana’s decentralization fully recovered from the Hetzner shutdown in approximately 49 days. Our analysis also decomposes the entropy into the number of nodes in the network and the distribution of block production across those nodes. We discuss why the difference in recovery rates between Bitcoin and Ethereum in response to the China ban and Solana’s rapid recovery from the Hetzner shutdown together support our hypothesis. We also find that the Ethereum Merge, which transitioned Ethereum from a Proof-of-Work (PoW) consensus mechanism (which is less flexible) to a Proof-of-Stake (PoS) mechanism (which is more flexible), provides a natural proxy for increased resource flexibility, thereby increasing Ethereum’s decentralization

Read the complete text