Geographical Centralization Resilience in Ethereum's Block-Building Paradigms

TL;DR

The paper tackles the problem of geographical centralization in Ethereum by introducing an agent-based simulation that jointly models validator locations, information-source geography, and two block-building paradigms—local and external. It shows that block-building design, information-source placement, and supplier geography create location-dependent payoffs that drive migration and hub formation, challenging the assumption of geographic neutrality. The study demonstrates that consensus parameters, such as attestation thresholds and slot times, modulate latency sensitivity and can either amplify or dampen centralization tendencies. The results have practical implications for protocol design, suggesting mitigation strategies like decentralized builders and reduced latency sensitivity to preserve decentralization, liveness, and fairness in Ethereum.

Abstract

Decentralization has an important geographic dimension that conventional metrics, such as stake distribution, often overlook. Where validators operate affects resilience to regional shocks (e.g., outages, natural disasters, or government intervention) as well as fairness in reward access. Yet in permissionless systems, validator locations cannot be prescribed by protocol rules; instead, they emerge endogenously from economic incentives. When certain locations offer systematic advantages, validators may strategically co-locate to maximize expected rewards, as observed in Ethereum, where validators cluster along the Atlantic corridor, which exhibits structurally favorable latency. In this paper, we design and implement an agent-based simulation framework to study how Ethereum's protocol design, particularly its block-building paradigms of local and external block building, interacts with validator and information-source distributions to shape geographical positioning incentives. Our simulations show that Ethereum's block-building architecture is not geographically neutral: both paradigms induce location-dependent payoffs and migration incentives, with asymmetric access to information sources amplifying geographical centralization. We further demonstrate that consensus parameters, such as attestation thresholds and slot times, modulate latency sensitivity and can amplify these effects, acting as protocol-level levers. Finally, we discuss the implications of our findings for protocol design and outline potential mitigation directions informed by our analysis.

Figures (14)

• Figure 1: Validator distribution and inter-region Internet latencies illustrate the geographic concentration of Ethereum validators and the latency landscape across regions. Validator location data covers $\sim$16000 Ethereum validators, provided by Chainbound chainbound_geovalidators. Latency values are macro-regional medians of round-trip times between Google Cloud regions. Darker shading indicates a larger number of validators in a given country, showing that validators are primarily concentrated in the United States and Europe. Darker inter-regional links indicate lower average round-trip latency, highlighting the particularly low latency between North America and Europe relative to other regions.
• Figure 2: Block production and dissemination under the Local and External block-building paradigms.
• Figure 3: Baseline configuration (homogeneous validators and information sources). Evolution of centralization metrics under Local and External block-building paradigms.
• Figure 4: Information-source placement effect (homogeneous validators). Evolution of centralization metrics under Local and External block-building paradigms for latency-aligned and latency-misaligned information-source placements.
• Figure 5: Validator distribution effect (homogeneous information sources). Evolution of centralization metrics under Local and External block-building paradigms with a heterogeneous validator distribution.

Theorems & Definitions (1)

• Remark