Optimal Execution in Cryptocurrency Markets

Abstract

The purpose of this paper is to study the Almgren and Chriss model on the optimal execution of large block orders both on the NYSE and in cryptocurrency exchanges. Their model minimizes execution costs, which include linear temporary and permanent price impacts. We focus on how the stock market microstructure differs from a cryptocurrency exchange microstructure and what that means for how the model functions. Once the model and microstructures are explained, we examine how the Almgren-Chriss model functions with stocks from the NYSE, looking at specifically selling a large number of shares. We then investigate how a large "wholesale" exchange like Binance has very different execution costs when compared to a "retail" exchange like Coinbase's consumer exchange, and what this means for a trader trying to make a large block order optimally. We examine how traders should place large orders on Bitcoin exchanges, and quantify the importance of choosing the right exchange.