Where in the World Is Your Blockchain? (And Why It Might Matter More Than You Think)

Table of Links

Abstract and 1. Introduction

2 Methodology

3 Hardware

4 Software

5 Network

6 Consensus

7 Cryptocurrency Economics

8 Client API

9 Governance

10 Geography

11 Case Studies

12 Discussion and References

A. Decentralization and Policymaking

B. Software Testing

C. Brief Evaluations per Layer

D. Measuring decentralization

E. Fault Tolerance and Decentralization

10 Geography

Geographic decentralization is a key point of interest [122,161,156], and it touches upon all layers covered in the previous sections. Accordingly, it involves all resources described so far, e.g., hashing power or tokens. Nonetheless, it constitutes a dimension on its own, as parts of a system may be well distributed w.r.t. one dimension but geographically concentrated.[15]

The tendency to centralize in certain areas arises due to economical, technological, or sociopolitical factors. For example, miners often set up their operations in countries with low electricity costs, hardware companies operate in countries with small production costs, nodes are hosted in areas with high internet speed, and tokens are accumulated by residents of countries with low taxes and where many exchanges operate. Geographical centralization poses two main threats to the properties of a ledger: i) physical hazards and ii) legal impediments.

Physical safety. Physical hazards could threaten a system’s infrastructure. If part of a system is located in a small area, connectivity failures or outages could destroy or split the ledger’s network.[16] This concern is particularly relevant in PoW, where equipment is hard to relocate. All resources examined above can be impacted (e.g., via drops in hashing power or token loss when mining equipment or cold storage is damaged), while the relevant parties are the regions of resource concentration. Single points of failure may arise when geographically-concentrated nodes act as central hubs, harming either safety or liveness (cf. Section 5) and, indirectly, stability (e.g., due to increased market volatility).

Legal compliance. Failures can also possibly occur due to legal pressures. If some layer is concentrated in a specific jurisdiction, authorities can possibly restrict or subvert it. Again, this touches upon all resources examined so far, as all are influenced by the law, with the relevant parties being legal jurisdictions. Depending on the occasion, different properties of the system are impacted. For example, if a country bans Bitcoin mining, the power drop could decrease the threshold for controlling a majority (safety hazard), while blocks are produced at a slower pace until the PoW difficulty is recalculated (liveness hazard). Stability could also be hurt, if some part of the system, e.g., mining, software access, or asset ownership is restricted. Additionally, exchanges can be legally bound to follow KYC procedures to comply with AML regulations [127], linking the users’ identities to their activity something that may lead to compromising their privacy. Arguably, a system is more likely to uphold its properties by falling under many jurisdictions, such that violating the properties requires the coordinated efforts of multiple authorities.

[15] For example, independent actors may participate in mining within a single country.

[16] Although exogenous hazards, like natural disasters, are outside the scope of this work, concentration in a small area can enable weaker adversaries to disrupt a ledger’s execution. For example, an adversary could isolate a building or a single computing center from the rest of the network, although assuming an adversarial disruption of the grid of an entire country, or continent, can be considered unrealistic.