Political scientists have examined which characteristics of electoral candidates are preferred by voters, commonly using conjoint survey designs where candidates' characteristics are randomized. In these experimental studies, survey respondents tend to prefer young and female hypothetical candidates. However, young and female candidates are substantially less likely to hold political offices. In this study, we focus on the case of Japan, where gender and age disparities among politicians are large, to disentangle the reasons for the discrepancy between experimental evidence and real-world elections. Employing a conjoint survey experiment conducted immediately after the July 2022 Upper House elections, we compare respondents' choices in the experiment with their self-reported voting behavior in the actual election. Our analysis reveals that while voting in the actual election aligns with preferences on most candidate characteristics shown in the experiment, real-world incumbent candidates have a much greater advantage than the conjoint estimate suggests. This implies that gender and age disparities among politicians persist because voters have a strong preference for incumbents, even though they are predominantly male and older.
Existing studies on propaganda in autocracies predominantly focus on how it highlights the instrumental, pragmatic benefits delivered by the regime, like economic development. However, the effectiveness of such instrumental propaganda is inherently bounded by the actual instrumental benefits received by the mass public. How does propaganda maintain the appeal of the autocratic regime, especially during periods of weakened instrumental appeals? To answer this question, this study explores the role of non-instrumental propaganda—propaganda that features the sentimental and moral appeals of the regime. Specifically, it investigates whether and how such non-instrumental propaganda varies with the actual instrumental benefits delivered by the regime, as is manifested in short-term economic performance. Examining nearly one million social media posts published by major state media accounts in China from 2013 to late 2020, this study finds that during short-term economic underperformance, these state media are more likely to feature a quintessential non-instrumental propaganda: Patriotism. Further insights from topic modeling reveals the potential mechanism of how patriotism propaganda can offset the decline in instrumental appeals: It highlights the sentimental and moral appeals of the country, seamlessly intertwining them with the appeals of the regime. In contrast to conventional understandings, it does not seek to invoke nationalism--the antagonistic, chauvinistic discourses of foreign actors. These findings shed light on the strategic use of non-instrumental propaganda by autocratic regimes, especially when confronted with limitations in demonstrating its legitimacy solely on instrumental grounds.
Chair: Xingchen Lan
This paper reviews the successes and failures of both the United States and China in shaping preferences through aid in Indonesia during the 2010s. The analyses in the study draw on Indonesian survey data from the Asian Barometer, Gallup World Poll, and Google Trends. Results indicate that evaluations of aid often depended on dynamics in the Indonesian political landscape. When Indonesians near Chinese aid projects expressed negative impressions of Chinese influence, these trends were mediated by higher nationalist sentiment. When Indonesians near U.S. aid projects expressed negative impressions of U.S. influence, these trends were mediated by higher concerns about corruption, likely due to elite capture. On average, over the period between 2011 and 2017, exposure to Chinese aid projects increased negative evaluations of Chinese leadership, while exposure to U.S. aid projects increased positive evaluations of U.S. leadership. Increases in positive perceptions of U.S. leaders were mediated through greater interest in the United States, measured by volume of Google searches related to America.
In government bureaucracies, achieving policy objectives typically involves multiple agents. Conventional wisdom suggests that increasing the number of agents leads to more shirking and less collusion due to cooperation challenges. I argue, however, that when agents possess distinct, non-overlapping expertise, a greater number of agents can promote collusion and mitigate shirking. I illustrate this idea through a modified version of the classic public goods game. The theoretical prediction is supported by the empirical analysis of China’s automatic ambient air quality network expansion between 2012 and 2022. I build an original dataset of the mission statements from all functional agencies at the prefectural level in China. Using this granular data, I develop empirical metrics to assess how executive responsibilities in environmental governance are distributed across functional agencies in each prefecture. I find that prefectures with more unique functional units dedicated to environmental governance build a surplus of monitoring, indicating a decrease in shirking behaviors. Interestingly, the same group of prefectures are also more likely to place their stations in cleaner sites post-automation, raising concerns about the potential for collusion. The results underscore the critical influence of organizational structure on the efficacy and integrity of policy implementation.
Chair: Bokai Qi
When do foreign policy bureaucracies in authoritarian regimes provide expert information and when do they withhold it? We argue that bureaucratic protection from arbitrary punishment shapes information provision. Bureaucrats are more likely to provide substantive and original information pertaining to their functional position when leaders use power sharing institutions to tie their hands from punishing bureaucrats if the leader arbitrarily dislikes their information. We test our theory by applying a range of supervised and automated text analysis methods to an original set of over 6,000 archival documents from the People's Republic of China during the Cold War, including nearly 2,000 diplomatic cables.
Established research on authoritarian information control has extensively examined the top-down dissemination of political propaganda and its impact on public attitudes and behaviors. This paper introduces a novel theory, performative propaganda engagement, which focuses on individuals who engage with state propaganda in a performative manner to benefit an individual or a group they align with, rather than diligently endorsing or promoting propaganda. Through mixed methods research approaches, this research empirically investigates performative propaganda engagement within the realm of Chinese online celebrity fandom, a rising cultural force on Chinese social media. The findings reveal that celebrity fans in China actively incorporate the promotion of state propaganda into their daily activities, aiming to enhance the visibility and reputation of their celebrities. Fans primarily engage with celebrity-signaling propaganda, and their engagement is instrumental. By exploring the manifestations of performative propaganda engagement, this research contributes to a deeper understanding of authoritarian information control and its downstream effects, contemporary online fandom culture in China, and the metrics-driven nature of social media ecosystem in the digital age.
Chair: Sophie Xiangqian Yi
Political selection is often studied as a conscious and calculated decision based on candidates’ competence and connections. This article highlights an important, yet overlooked, tacit domain that involves subconscious (or even unconscious) processing of certain perceptual traits. Using an original dataset of over 18,879 photos for 5,124 Chinese officials and a novel supervised machine-learning method to rate officials' photos, we examine the impact of officials' facial appearance characteristics on their career advancement and political survival. Our results suggest that perceptions of attractiveness, intelligence, trustworthiness, and non-aggressiveness are all strong and systematically associated with better promotion prospects, and trustworthiness is associated with a low likelihood of purge. Moreover, we find that promotions to the top level posts place greater emphasis on perceived trustworthiness and non-aggressiveness, whereas those to mid-level posts prioritize attractiveness and intelligence. We reaffirm these selection patterns with a survey experiment wherein respondents evaluated the rank of simulated official profile pictures. These findings challenge the rationalist assumption in political selection research by highlighting the diverse criteria and the complex cognitive process involved.
How might face-based judgement of female candidates influence election outcomes? Visual information plays a crucial role in non-verbal communication. The first impressions formed based on visual information in the initial stages of interaction can continuously influence subsequent communication. Especially in elections where voters do not have sufficient information about the candidates, the first impression of the candidates encountered through posters can significantly impact voters’ choices. This study analyzes the impact of face-based judgement of female candidates on their perceived competence and election viability, using photos of actual candidates who ran in local elections in Korea. Focusing on candidates from major parties, photos from election posters submitted by candidates who ran in the four local elections held since 2010 were collected, after removing textual information, including party affiliations, from the posters. About 2000 photos were selected and included in a survey targeting respondents selected considering gender, age, and region. Each respondent randomly evaluated 10 candidates (5 men and 5 women) on gender typicality, competence, trustworthiness, and election viability. The survey also included questions to determine respondents’ stance on sexism or feminism, aiming to understand the heterogeneity in the impact of candidates’ appearance evaluations based on respondents’ characteristics.
Chair: Cheng Cheng
This study explores whether the patronage between village leaders and an authoritarian regime continues to influence voting outcomes in later elections under authoritarian and democratic settings. We claim that authoritarian patronage remains effective over time, even after democratization, only when the clients can maintain strong trust in patronage. Using original data on village-leader trainees under the New Village Movement, a nationwide campaign-style rural development program in 1970s South Korea, we find that townships with more leaders trained by the authoritarian government were more likely to vote for the incumbent party in the subsequent legislative election. This tendency persisted in the following authoritarian elections in the 1980s under a new dictator. After South Korea transitioned to democracy in 1987, this effect initially waned but reemerged strongly as the authoritarian successor party rebranded through a strategic party merger. However, the effect of patronage eventually disappeared, as democracy consolidated and the agricultural sector became marginalized with neoliberal globalization. Our findings highlight that the legacy of authoritarian patronage is not simply an outcome of a strong dictatorship, but significantly shaped by the political and economic conditions of a new democracy.
In this paper we explore how extraordinary natural occurrences influenced executive decisions across different reigns in history. We utilize the Veritable Records of the Joseon Dynasty of Korea, the official daily accounts of the kingdom’s various affairs recorded between 1413 and 1865 CE. Covering the longest continual period of any single dynasty in the world, the annals provide an invaluable historical case study of daily decisions that were made by rulers and their council. Under the Confucian tradition the annals also record in detail various uncommon natural phenomena, interpreting anomalies and disasters (천변지이 [天變地異]) as heavenly signs of the ruler’s legitimacy. We find that these phenomena led to meaningful changes in how decisions were made by the governing body. Subsequent to the events the royal court took on a greater executive role and witnessed more cohesion between the king and his council, i.e. more decisions were made under the king’s direct orders and with less disagreement between the ruler and the council. These changes reflect concerted effort by the ruling body to strengthen their position in society and address legitimacy concerns.
Chair: Sangyong Son
How rebel groups interact with civilians has significant implications for the dynamics of civil war, yet most research has focused on the origins of rebel governance rather than its consequences. We argue that coercive rebel actions to control the population can negatively affect civilian compliance with rebel rule, undermining the military resources available to the rebels. We provide evidence for this claim with archival data from the Chinese Civil War (1946-1949), when the Chinese Communist Party carried out violent land reform in occupied areas that killed hundreds of thousands of people. We digitize a new dataset of 260,000 Communist soldiers to estimate how land reform affected the Communist Party's soldier recruitment during the Civil War. Comparing county-level soldier recruitment before and after land reform with a difference-in-differences estimation, we document a substantively large and negative effect of land reform on the number of new soldiers joining the Communist army. Our estimation shows land reform led to a 15% decrease in new soldiers. This effect was driven primarily by the violence in land reform, which outweighed any potential positive effects associated with land redistribution. These findings highlight the military cost of coercive rebel governance in civil war.
Poster Presenters