Characterization of the adopters of the Bitcoin

FRANCESCO PARINO; MARIANO G. BEIRO; LAETITIA GAUVIN

Zaragoza

Conferencia; BIFI International Conference 2018; 2018

Institute for Biocomputation and Physics of Complex Systems

Bitcoin is a cryptocurrency which became very popular during the recent years. The bitcoin transactions are recorded in a public ledger, the bitcoin blockchain, and can be collected since its creation.The bitcoin blockhain works without a central authority but is based on a peer-to-peer system. The transactions take place between users directly and not necessarily with a transaction fee. More exactly the bitcoins exchanged during a transaction are sent from an address to another, each address belonging to a user. One of the challenge to study such a system and understand the behaviour of users in this new transaction system is the possibility for any user to generate multiple addresses.Here we propose a preliminary analysis to get a sense of the users of the bitcoin despite the high degree of anonymity inherent to the system. We do not attempt to identify single users but to charaterize the spatial distribution of users at the country level. Even though it is not possible to link an IP address to the authors of a transaction, one can obtain the IP of the first user which relays the transaction. The current literature on the bitcoin blockchain would tend to support that this IP has some chance to be the IP of the source of the transactions. As a first attempt we thus assume that these IP addresses allow to estimate where the users come from. In order to test this assumption, we compare the countries of provenance of the IP addresses to the geolocalization of the user downloading wallets which are softwares to access the bitcoin technology. We observed a good agreement between the two. This confirms that the IP address that relay transactions can be used to some extent as a good proxy to identify the provenance of the users. However due in particular to the increasing broadcasting of transactions by services instead of single users, the IP addresses are only accessible for a given time interval. For this reason, in order to characterize the evolution of the adoption in different countries we use Google Trends data to quantify the collective attention for the bitcoin, this appears as good proxy when compared to the unique IP addresses. Looking at the evolution of the bitcoin search through Google trends for different countries we can have an hint on the early and new adopters. To complete the study we also build the network of users where a link is given by a transaction and look at the bilateral exchanges among countries using the IP addresses mentioned earlier as well as some heuristic developed in the literature to assign a country to each transaction. We finally compare this to a null model to extract what are the preferred relationships between countries for transactions. To summarize, we propose a way to measure the collective interest for the bitcoin at the level of the country and to understand how the transactions are distributed among the countries.