Autocratic rulers often need to maintain a viable political coalition and share power with their supporters. However, autocratic regimes vary greatly in their capacity to implement power-sharing. To explain variations in power-sharing institutions, most scholarship relies on commitment theory: the effectiveness with which autocratic rulers build power-sharing coalitions is constrained by their ability to credibly commit to positions, rents, and policy concessions. Institutions that constrain executive power—such as legislatures, political parties, or military bodies—are viewed as necessary to facilitate power-sharing. Yet, while rulers face moral hazard in potentially eliminating elites, elites also confront moral hazard in withdrawing their support. Co-optation of outsiders can introduce heterogeneity within the ruling coalition and may lead to further instability. Following this logic, the conventional wisdom posits a trade-off between loyalty and competence, suggesting that political selection in autocracies often results in the aggregation of mediocrity due to commitment problems.
This paper advances a revised perspective on autocratic power-sharing. Departing from conventional accounts, we shift focus to the role of institutionalized bureaucracy. Contrary to the loyalty-competence trade-off, we argue that both loyalty-based and performance-based selection can enhance the political resilience of the regime. We illustrate this argument using late Qing China as a case. A defining feature of the Chinese bureaucracy was its high degree of institutionalization. Lacking constitutional authority and coercive means to challenge the ruler, bureaucrats were unlikely to pose an immediate threat. This ensured credible power-sharing motives on the part of the ruler and incentivized elites to defend the established social and political order. However, bureaucratic selection in the Qing was complicated by ethnic considerations. For Qing rulers, ethnic identity served as a heuristic for the prior political loyalty of bureaucrats. To preserve Manchu dominance at the political center, rulers adopted a system of political patronage that disproportionately promoted Manchu bureaucrats to high-ranking positions.
Yet such favoritism was bounded: the Manchu ruling bloc could not maintain social stability by excluding Han Chinese. A relatively open system of bureaucratic recruitment was necessary to attract Han cooperation. Han elites gained administrative competence at all levels of local governance through competitive examinations. As a result, Han bureaucrats were relatively more skilled at managing complex local conditions and more effective at suppressing rebellions. At the same time, Han officials might be perceived as less loyal. This paper argues that Qing rulers adopted a fine-tuned system combining both meritocratic and patronage elements to implement political power-sharing.
This argument is supported by an empirical analysis of the interaction between the promotion, demotion, appointment, and dismissal of governors-general and governors (provincial-level officials) and internal military conflict at the provincial level. We treat ruler selection as a revealed preference for bureaucratic power-sharing. Internal conflict provides rulers with a tangible measure of performance. The paper focuses on the following empirical questions: How does internal conflict influence political turnover and the selection of provincial leaders? Do Manchu and Han officials face disparities in opportunities for promotion and sanction? If ethnic favoritism existed, how did it interact with the threat of large-scale rebellion?
About the author:
Xi Tianyang, Tenured Associate Professor of National School of Development, Peking University; Ph.D. in Political Science, New York University; B.A. and M.A. in Economics, School of Economics, Fudan University. His Research interests include new political economy, development economics, historical political economy, and comparative institutional analysis. His research covers the political economy of bureaucratic selection and government governance in contemporary China, bureaucratic institutions, great-power decentralization, and state capacity from the perspective of historical institutional analysis, as well as institutional bargaining and leader behavior in comparative politics. He has published extensively in leading peer-reviewed journals in both Chinese and English, including Journal of Politics, British Journal of Political Science, Journal of Theoretical Politics, Comparative Political Studies, and Journal of Development Economics.
