Overview of Cryptoeconomics
The science of cryptography has been in existence for over millennia but in a formal and systematized form for just a couple of decades — can be simply defined as the study of communication in an adversarial environment (Rabah, 2004). Similarly, we can define cryptoeconomics as a concept that goes one step further, i.e., the study of economic interaction in an adversarial environment (Davidson, De Filippi & Potts, 2016; Ernst, 2016). To distinguish itself from the traditional economics, which certainly studies both economic interaction and adversaries, cryptoeconomics generally focuses on interactions that take place over network protocols. Particular domains of cryptoeconomics, include: Online trust and reputation systems;
Cryptographic tokens / cryptocurrencies, and more generally digital assets; Self-executing “smart” contracts; Consensus algorithms; Anti-spam and anti-sybil attack algorithms; Incentivized marketplaces for computational resources; Decentralized systems for social welfare / mutual aid / basic income; Decentralized governance (for both for-profit and non-profit entities).
In the last couples of years we have witnessed an increased in prominence of cryptoeconomics, which to a large extent as a result of the growth of cryptocurrencies and digital tokens, and which brings a new, and interesting dimensions to cryptography (Potts, Davidson & De Filippi, 2016). While before, cryptography was, by and large, a purely computational and information theoretic science, with strong guarantees built on security assumptions that there are close to absolute; once money enters the picture the perfect world of mathematics must interact with a much more messy reality of human social structure, economics incentives, partial guarantees and known vulnerabilities that can only be mitigated, and not outright removed. While a cryptographer is used to assumptions of the form “this algorithm is guaranteed to be unbreakable provided that the underlying math problems remain hard”, the world of cryptoeconomics must content with fuzzy empirical factors such the difficulty of collision attack, the relative quantity o altruistic, profit-seeking and anti-altruistic parties, the level of concentration of different kinds of resources, and in some even sociocultural circumstances (Ernst, 2016; Davidson, De Filippi & Potts, 2016).
In contrast, the traditional applied cryptography, security assumptions tend to look something like this: i) No one can do more than 279 computational steps; ii) Factoring is hard (i.e., superpolynomial) (Rabah, 2005, 1); iii) Taking nth roots modulo composites is hard; iv) The elliptic curve discrete logarithm problem cannot be solved faster than
In cryptoeconomics, on the other hand, the basic security assumptions that we depend on are alongside those of cryptographic assumptions, roughly as follows (Ernst, 2016):
- No set of individuals that control more than 25% of all computational resources is capable of colluding
- No set of individuals that control more than 25% of all money is capable of colluding
- The amount of computation of a certain proof of work function that can be accomplished with a given amount of money is not superlinear beyond a point which is reasonably low
- There exist a non-negligible number of altruists and a non-negligible number of crazies or political opponents of the system, and the majority of users can be reasonably modeled as being close to economically rational
- The number of users of a system is large, and users can appear or disappear at any time, although at least some users are persistent
- Censorship is impossible, and any two nodes can send messages to each other relatively quickly
- It is trivial to generate a very large number of IP addresses, and one can purchase an unlimited amount of network bandwidth
- Many users are anonymous, so negative reputations and debts are close to unenforceable
In this respect, it’s important to note that there are additional security assumptions that are specific to certain problems. Thus, quite often it will not even be possible to definitively say that a certain problem has been solved. Rather, it will be necessary to create solutions that are optimized for particular empirical and social realities, and continue further optimizing them over time (Ernst, 2016).