Centralized tax administration with top-down monitoring is regarded as a key element of the modern fiscal state. However, centralized administration can also introduce new agency problems for the monitors and create distortions across levels of the bureaucratic hierarchy. This paper studies how a fiscal administration reform aimed at monitoring local taxation can have unintended consequences on state capacity in late 18th-century China, using a novel dataset on tax collection constructed from administrative archives. We exploit predetermined variation across counties in exposure to the reform, which imposed formal oversight on county taxation by provincial treasuries. We find that the reform had a negative effect on tax collection. The decline in tax collection can be explained by a top-down encroachment mechanism: the reform created greater scope for monitors to withhold and misappropriate county funds. Such intergovernmental fiscal encroachment undermined local tax enforcement capacity and distorted county bureaucrats’ incentives to underreport and embezzle revenue. We provide a range of evidence consistent with these causal chains.
This paper investigates whether merchant organizations shaped the wartime redistribution of political resources in late imperial China. During the Taiping Rebellion (1851–1864), the Qing government turned to merchant networks for war finance and, in exchange, expanded examination quotas for the shengyuan degree. Using crosssectional data on 262 prefectures, I show that pre-war merchant guild presence strongly predicts wartime quota increases, concentrated in battle-affected areas. Instrumenting guild presence with Ming-dynasty trade routes supports a causal interpretation. The lijin commercial tax is a plausible channel: prefectures collecting more lijin revenue received larger quota allocations. A difference-in-differences analysis confirms that quota expansions translated into higher jinshi production after 1864. The findings provide quantified evidence that fiscal contributions can purchase political selection in a premodern autocracy.
Defections during war are extreme changes in loyalty. What motivates military officers to betray their motherland and serve the invaders? Using a novel dataset of career paths for over 2,800 high-ranking (colonels and generals) Nationalist (KMT) military officers during the Second Sino-Japanese War (as part of World War II), we examine defection cases to Japanese puppet regimes. Three findings emerge. First, high-ranking KMT officers who advanced more slowly in their careers were more likely to defect; suggesting that internal organizational recognition matters. Second, officers who were underpromoted compared to their schoolmates and townsmen were more likely to defect, suggesting that peer comparison matters. Third, officers were more likely to defect when their defected peers had better career prospects in the enemy’s camp, suggesting that external recognition matters.
This paper revisits the old thesis of the contrasting paths of modernization between Japan and China. It develops a new analytical framework regarding the role of knowledge acquisition (propositional vs. prescriptive) and political centralization as the key drivers behind these contrasting paths. Our model and historical data highlight how the introduction of these elements contributed to Meiji Japan’s decisive turn towards the West and Qing China’s lethargic response to Western imperialism. Our analytical framework, developed from a comparative historical narrative and quantitative data, sheds new insights onto the importance of knowledge acquisition for enabling developing countries to reach the world’s economic frontier.
This paper examines how mass media influences language assimilation by studying the introduction of radio broadcasting in Japan between 1926 and 1950. Leveraging Nippon Hoso Kyokai (NHK)’s monopoly in radio broadcasting and its exclusive use of Tokyo dialect, I analyze how exposure to NHK radio affected local linguistic patterns. Using comprehensive linguistic atlases and radio signal data, I find that exposure to NHK radio significantly reduced language distance to Tokyo dialect. I further analyze how economic conditions, human capital, and institutional context moderated this assimilation effect.
Having credible information is crucial for improving politicians’ decision-making. Existing research suggests that expanding the range of information and access to firsthand sources can enhance decision quality. However, quantitative evidence regarding the effect of information control, particularly on bureaucracies, remains scarce. Our study investigates an information monitoring reform within the Qing China bureaucracy in 1722. This reform enabled local officials in designated positions to report directly and confidentially to the emperor, thereby streamlining information flow and strengthening monitoring capabilities. Using a generalized difference-in-differences design, we find that the introduction of the information monitoring system surprisingly undermined the efficiency of governmental disaster relief allocation. Specifically, prefectures with more direct reporting positions were more likely to receive disaster relief programs, even in the absence of severe disasters. This bias arose from information overload caused by an expanded but uneven information supply, making the emperor more susceptible to being swayed by dominant information flows due to limited attention. Additionally, we identify a corrective mechanism for the emperor’s misjudgment. The emperor placed greater weight on the reports of trusted officials who shared his ethnicity or were members of the ruling class, thereby mitigating the negative effects of information overload. This study contributes to improving the understanding of the unintended consequences of bureaucratic information control.
This paper examines the importance of local knowledge for state building. Drawing upon the quantity and quality of local gazetteers, the regional encyclopedias of imperial China, we measure the accumulation of local knowledge in 267 prefectures over a millennium. We find that the gazetteers facilitated the local penetration of state power and infrastructure between 1000 and 1820, and continued to contribute to the modernization of state infrastructure during the period of reform between 1860 and 1911. Analyzing the content of gazetteers reveals that the gazetteer effect arises from the richness, concretization, and uniqueness of records of local characteristics. These findings indicate that the state is not merely built on unified and standardized measures but also on context-based local knowledge that increases the legibility of subjects.
How can governments use redistribution to mobilize support for war? This paper shows that redistributive policies can increase political participation by altering the cost-benefit calculus of participation. I study a unique case of wartime land reform conducted by the Chinese Communist Party during the Chinese Civil War. Using newly digitized death records of more than 566,000 soldiers, I show that land redistribution increased death tolls on the extensive margin at the county-month level, but reduced fighting effort on the intensive margin when transfers were large—consistent with greater free-riding. However, both effects were driven by regions near enemy lines, where peasants faced higher risks of landlord reprisals. The findings highlight how strategic redistribution and violent class struggle interacted to reshape incentives for participation in collective violence. More broadly, the paper contributes to understanding how states engineer compliance during wartime through economic and coercive tools.
States often deploy public education to instill official ideology and advance state-building. Yet schooling can also enlighten and empower; when it reaches disadvantaged groups, it may nurture challengers to the state. This paper examines this paradox in the context of the Nationalist government’s 1930s expansion of tuition-free secondary teachers’ schools. Using a panel dataset covering 1,697 counties, we show that counties with more teachers’ schools were more likely to establish Chinese Communist Party (CCP) county committees from March 1938 onward, when the CCP resumed mass recruitment and readmitted intellectuals. Two mechanisms account for this pattern. First, tuition-free teachers’ schools created a pool of educated youth from non-elite backgrounds who, given their origins, were especially receptive to CCP appeals. Second, following the 1927 KMT–CCP split, many CCP intellectuals fled to the countryside and taught covertly, giving students at rural teachers’ schools greater exposure to communism.
