Measuring Ethereum-Based ERC20 Token Networks

The blockchain and cryptocurrency space has experienced tremendous growth in the past few years. Covered by popular media, the phenomenon of startups launching Initial Coin Offerings (ICOs) to raise funds led to hundreds of virtual tokens being distributed and traded on blockchains and exchanges. The trade of tokens among participants of the network yields token networks, whose structure provides valuable insights into the current state and usage of blockchain-based decentralized trading systems. In this paper, we present a descriptive measurement study to quantitatively characterize those networks. Based on the first 6.3 million blocks of the Ethereum blockchain, we provide an overview on more than 64,000 ERC20 token networks and analyze the top 1,000 from a graph perspective. Our results show that even though the entire network of token transfers has been claimed to follow a power-law in its degree distribution, many individual token networks do not: they are frequently dominated by a single hub and spoke pattern. Furthermore, we generally observe very small clustering coefficients and mostly disassortative networks. When considering initial token recipients and path distances to exchanges, we see that a large part of the activity is directed towards these central instances, but many owners never transfer their tokens at all. In conclusion, we believe that our findings about the structure of token distributions on the Ethereum platform may benefit the design of future decentralized asset trade systems and can support and influence regulatory measures.