Abstract
In response to the sudden outbreak of COVID-19 in 2020, the Chinese government implemented a stringent "Zero-COVID Policy." Although these measures initially appeared to rapidly control the virus's spread, two years later, they caused significant restrictions on residents' personal freedoms, leading to widespread controversy and strong distrust of local governments among many residents. In the past, scholars have used social capital and political culture to study political trust. This paper uses data collected from the CGSS2021 database during the pandemic to demonstrate that social capital and political culture significantly influenced Chinese residents' political trust in local governments during the pandemic. It further proposes the theoretical basis behind this influence mechanism.
Hypothesis
Hypothesis 1: During the pandemic, social capital continues to significantly influence residents' political trust in local government.
Hypothesis 2: The pandemic has led to a decrease in social frequency, thereby causing a decline in social capital and resulting in weakened political trust in local government among residents.
Hypothesis 3a: During the pandemic, the shallow level of obedience to the government significantly affects residents' political trust in local government.
Hypothesis 3b: During the pandemic, the deep level of obedience to the government significantly affects residents' political trust in local government.
Research Design
The study treats political trust (a binary variable) as the dependent variable. The key independent variables are grouped into two categories. The first group focuses on social capital, including "social activity frequency," "change in social activity frequency" (which denotes the change in social activity frequency during the pandemic), and "social trust." The second group examines citizens' agreement with the government’s pandemic policies. Additionally, the study accounts for two sets of control variables. The first set includes basic demographic information, such as gender, age, annual income, and length of residence in the locality since the age of 14. The second set consists of variables related to political participation attitudes and overall political trust.
Â
Thesis Conclusion: Logistic Regression
Through logistic regression analysis, this study confirms that during the pandemic, social capital significantly influenced residents' political trust in local government in authoritarian China. It addresses the debate on the role of social capital in political trust outside democratic contexts. The findings highlight that decreased social interaction during the pandemic further challenged residents' trust in the local government.
This research also contributes to the literature on political culture and trust in China by demonstrating that residents' attitudes toward political participation and obedience to government policies directly impacted their political trust. Notably, this study is the first to measure residents' compliance with government pandemic policies, revealing a strong positive correlation between compliance and political trust. However, the study also cautions that the high levels of obedience seen in the CGSS 2021 data may not accurately reflect residents' true sentiments due to China's unique political environment. As citizens’ obedience decreases, their trust in the government significantly wanes, potentially leading to governance failures.
Real-world evidence indicates that despite government attempts to suppress dissent, dissatisfaction has surged, undermining trust in local government. The pandemic’s policies have led to deeper resentment, further challenging local governance in China.
Based on these findings, three policy recommendations are proposed: 1) In economically underdeveloped regions, governments should focus on rebuilding social trust by enhancing public social interactions and restoring disrupted social networks. 2) The government should adjust policies related to major infectious disease outbreaks, as excessive restrictions undermine trust and create a vicious cycle of discontent. 3) Local governments facing a decline in trust should implement reasonable policies aimed at restoring trust between residents and government.
Â
Â
Interaction analysis, estimating and robustness check
The author conducted an interaction analysis to explore potential interaction effects between different independent variables; however, none of these interactions were statistically significant. Additionally, the author used the model to estimate the average levels of political trust across various provinces and cities and check the robustness of the model by replacing the model. These results are presented in the Appendix (currently not available online).
Further Exploration: Random Forest
The random forest model identifies several key predictors of political trust. Among them, provinces, social trust, income, and political interest emerge as the most influential variables. The province of residence demonstrates the highest contribution to classification accuracy, suggesting substantial regional variation in political trust. Social trust also shows a strong positive association, aligning with existing theories that interpersonal trust facilitates institutional trust. Income exhibits high importance in terms of Gini reduction, indicating that economic status plays a notable role in shaping trust in political institutions. Political interest and frequency of social activities further contribute to model performance, reflecting the relevance of civic engagement and political attentiveness. In contrast, variables such as gender and the first measure of political obedience display limited predictive power. Overall, the results underscore the multifaceted social and regional determinants of political trust.
Â
