Topics in crypto asset and blockchain finance

This thesis contributes to the crypto asset and blockchain empirical finance literature in three key areas: (i) market risk modelling, by developing simple volatility models which exhibit equal forecasting ability in terms of crypto asset tail risk measure and volatility forecasts, when compared against complex models; (ii) market manipulation, by extending a methodology derived from securities fraud litigation studies to identify blockchain transactions with a manipulative effect on crypto asset prices; (iii) crowdfunding via token offerings, by identifying factors of fundraising success using regression models, and exploring how these factors vary across time. Each of the above contributions is developed in a separate chapter. Firstly, the market risk modelling chapter provides extensive backtests of hourly and daily Value-at-Risk and Expected Shortfall forecasts regarded as best practice in the industry and used for regulatory approval. Results demonstrate that simpler models in the EWMA class are just as accurate as GARCH models for VaR and ES forecasting, and similarly when using average scores generated from proper univariate and multivariate scoring rules. Secondly, the market manipulation chapter examines large blockchain transactions of the tether stablecoin and assesses whether they produce positive abnormal returns for bitcoin prices. The methodology is adapted from single-firm event studies used in securities fraud litigation, using regression factor models. The chapter’s findings can be useful in determining materiality and estimating damages in legal cases of crypto asset market manipulation. Finally, the tokenomics of crowdfunding chapter examines the fundraising success of token offerings for the 2017 – early 2022 period, constituting one of the most comprehensive studies in this topic. We proxy fundraising success with the amount of funding raised and also by minimum funding target exceedance. Success factors are derived from the venture, token and offering characteristics, as well as additional common factors such as the price level of ether and the launchpad platforms used. The findings of this chapter provide insights as to the evolution of token offering success factors, with the choice of launchpad platform emerging as a new and significant factor and to some extent overshadowing the determinants previously documented in the relevant literature.