- This event has passed.
“Preferences for Government Concessions amid Protests: A Conjoint Experiment with Causal Interactions in Hong Kong” by Hans H. Tung and Ming-Jen Lin
Event co-organized with the Cambridge University Press Taiwan Studies Series
Abstract: The paper empirically tests at a micro-level Acemoglu and Robinson’s (2006) commitment problem thesis by conducting a conjoint experiment with causal interactions on protesters in Hong Kong’s recent anti-extradition movements. According to the thesis, the dictator’s (redistributive) concessions to rebels won’t be credible ex ante without democratization, which leads to a change in the median voter’s identity. While the bulk of the literature has mostly investigated the thesis cross-nationally, our survey experiment provides an empirical test for a micro-level prediction derived from it that a concession without any franchise extensions offered by the government should be rejected vis-a-vis that with them. Our experimental design has the following critical features. First of all, our conjoint design captures the multidimensionality of future potential concession packages (including political reform, economic reforms, social assimilation, law enforcement, and government personnel) and allowed us to estimate the independent effect of each dimension. Moreover, to understand the effect of democratization from the Acemoglu-Robinson thesis, we also conducted a conjoint analysis featuring interaction effects (Egami and Imai, 2019). Finally, as de la Cuesta et al. (2019) contend, the external validity of a conjoint experiment very much depends on knowing the population distributions of attributes. As a result, we also adopted a data science approach to scrap all the related news reports from online media outlets to capture the population distributions of all the reform proposals listed above.