Understanding Sequential Decision-Making by Platform Participants: A Structural Analyses of Crowdsourcing and Cryptocurrencies

Abstract

In this dissertation, I study the decision-making process of participants on online platforms. I specifically focus my attention on the sequential aspects of such processes, i.e. my interest is in understanding the endogenous determinants of the decision-making processes in the contexts crowdsourcing and cryptocurrencies. First, I study the user participation in an open-innovation platform, specializing in solving socio-economic problems via crowdsourcing ideas and solutions. I hypothesize that platform participants are creatives who, with the right mix of knowledge and application, can find solutions to these problems requiring inter-disciplinary skills. I model their participation decisions as dependent on their entire history of knowledge acquisitions and applications i.e. every focal interaction, with the platform, is dependent on all the past interactions and all the future interactions are affected by the focal interactions. This allows me to uncover the nuanced usage of existing knowledge by platform participants in generating creative ideas that has implications for managers of knowledge-based crowdsourcing platforms. We also highlight the importance of feedback mechanisms that adds to platform participants' knowledge over time. Further, we also find a significant impact of learning-by-doing in improving users' creative ideation abilities. In my second study, I study the cryptocurrency mining industry, specifically focusing on the competitive equilibria in Bitcoin and Bitcoin-Cash around the time of Bitcoin fork (which led to the birth of Bitcoin-Cash). I also investigate the impact of emergency difficulty adjustment algorithm on the Bitcoin-Cash's equilibria. I model these mining pools (or miners) as forward-looking profit-maximizing firms whose strategic interactions lead to dynamic changes in the underlying system's protocols. Our results indicate that competition among Bitcoin miners has a positive externality, improving the expected payoffs for every miner. Further, our investigation reveals the dynamic impact of this externality for the Bitcoin-Cash ecosystem owing to emergency difficulty adjustment algorithm. We also estimate the varying impact of the within-miner competition, from BTC mining, on the BCH mining rate. Our counterfactual simulations provide further insights into possible configurations for the emergency difficulty adjustment algorithm.