close× Call Us
close×
Title Nationalist Repression in China and Japan
Post date 06/19/2018
C1 Background and Explanation of Rationale Nationalist sentiment is hypothesized to constrain both authoritarian and democratic regimes. Democracies are unable to repress nationalist actions as the ruling party will be punished at the polls. In autocracies, repressed actors or repressive agents can morph into political rivals. We investigate whether two governments – one authoritarian (China) and one democratic (Japan) – can effectively repress popular nationalism, and attempt to isolate the regime- related constraints they face. We use a survey experiment deployed in both countries to examine differences in levels of public support for government repression in both nationalist and non-nationalist, neutral contexts. By leveraging a politically sensitive territorial dispute common to both countries – the Diaoyu/Senkaku Islands dispute – we also examine whether public support for government repression of actions by citizens differs between China and Japan. See full pre-analysis plan for more detail.
C2 What are the hypotheses to be tested? One may expect nationalist issues such as the Diaoyu/Senkaku Islands dispute to face particularly high costs of repression due the visible and nationally charged nature of the dispute. This suggests that we may find larger treatment effect estimates for our repression treatment in the charged Diaoyu/Senkaku scenario, but smaller effects for our non-nationalist, labor dispute scenario. Treatment effects may also be larger in Japan due to the country’s democratic political system, which should be less receptive to the use of repression. But it is also possible that the government increases its support due to the repression under some conditions (see Heterogenous Treatment Effects section), and larger treatment effects can be observed in China than in Japan. Therefore, depending on the direction of differential treatment effects (if they exist), our experiment may reveal: (1) costless repression for both a democracy and an autocracy (or even “popular” repression), (2) repression that is costly only in a democracy, (3) costly repression only in an autocracy, or (4) costly repression in both an autocracy and democracy. If repression of protests in China increases support for the ruling Chinese Communist Party (CCP) or has a null effect, this would run against the predominant theoretical arguments that repression is costly in authoritarian regimes. In Japan, if even some subgroups support repression (e.g. supporters of the ruling Liberal Democratic Party), this would add nuance to current theories of the high costs of repression in democracies. Our experiment is also unique in the sense that we consider both nationalist and non- nationalist scenarios. For both scenarios, we consider all the possible outcomes described above: (1) costless repression for both a democracy and an autocracy (or even “popular” repression), (2) repression that is costly only in a democracy, (3) costly repression only in an autocracy, or (4) costly repression in both an autocracy and democracy. Although we may observe similar repression costs for both scenarios, it is possible that some citizens in both or either countries may find it easier to support or oppose the government’s repression in either scenario. This leads to twelve possible combinations in terms of treatment effects, summarized in Table 1. See full pre-analysis plan for more detail.
C3 How will these hypotheses be tested? * Our research design takes the form of a survey experiment administered in the form of a vignette to 3000 respondents in both China and Japan (6,000 respondents total). The experiment considers whether there are indeed costs to each type of repression, whether the costs of repression differ by type (nationalist and economic), whether the costs of repression differ by regime of interest (China and Japan), and whether the costs of repression differ amongst different groups of respondents. We test for costs of repression by calculating difference-in-means in our treatment group minus our control group. Differential costs of repression by type are estimated by comparing the magnitude of our treatment effect estimates in each of our treatment arms – nationalist and economic. Differential costs of repression by regime are estimated by comparing the magnitude of our treatment effects estimates in each of our surveys – China and Japan. Finally, we look for the existence of heterogenous treatment effects across a battery of pre- treatment covariates by regressing our outcome variable on treatments conditional upon the data representing the covariate of interest, as well as using Bayesian Additive Regression Trees (BART). See full pre-analysis plan for more detail.
C4 Country Japan, China
C5 Scale (# of Units) 3000 – Japan; 3000 – China; 6000 - Total
C6 Was a power analysis conducted prior to data collection? Yes
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number 2000023209
C9 Date of IRB Approval 30 April 2018
C10 Will the intervention be implemented by the researcher or a third party? Nikkei Research (Japan) and Survey Sampling International (Mainland China)
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? No
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? No
C13 JEL Classification(s) not provided by authors