We study the impact of history curricula on national identity in Taiwan. The high school curriculum reform of September 2006 separates the history of Taiwan from Chinese chronology and increases Taiwan-oriented content to transmit Taiwanese identity. We document an unintended "backlash" that individuals studying the new curriculum are more likely to hold both greater Taiwanese and Chinese identities. Our analysis suggests endogenous changes in information demand as a prominent mechanism: treated high schoolers show greater identity awareness and acquire more information related to both identities. We further observe consistent attitudinal changes, with milder political views and an increase in votes for median candidates or abstention.

Throughout history, political leaders have not been immune to their subjective beliefs — ranging from superstitions to denialism. This paper documents for the first time the substantial impact of politicians’ non-factual beliefs, leveraging enduring traditional Chinese superstitions about spaces that allow us to link quasi-random, leader-specific variation to regional development in their cities. We find that municipal zones perceived as unfavorable to mayors have an average 2 percent lower GDP compared to other zones. Exploiting mayoral reports and administrative micro-level data, we show  greater disruption in local public investment as a key driver. Downstream changes in firms and households further amplify the loss, with a 6% decrease in firm entry, a 4% reduction in the productivity of remaining firms, and a small population decline. While better institutional environments can mitigate these effects, campaign-style ideological training appears less effective. Our back-of-the-envelope calculation suggests such non-factual beliefs of Chinese mayors are associated with at least a 0.1% annual GDP loss over the past two decades. Overall, our findings highlight subjective beliefs as an important determinant of leader performance that matters for economic development. 

Merit is not directly observed but measured, and the conditions of measurement can shape how ability is revealed. This paper provides the first causal evidence that where assessments take place can systematically bias measured performance and, in turn, the allocation of opportunity. Using administrative data from China’s National College Entrance Examination and its random assignment of test centers, we find that students assigned to an off-site test center score 0.14 standard deviations lower than classmates testing at their home school, translating into a 3.8-percentage-point lower likelihood of college admission. Mechanism analysis suggests unfamiliar environments as a primary driver, while longer travel distances are also associated with reduced performance. The performance penalty is disproportionately larger among low-achieving and disadvantaged students. Because test centers are predominantly located in high-performing schools, these ostensibly neutral institutional arrangements may unintentionally widen achievement gaps between privileged and less privileged groups. Together, our results reveal assessment location (``where'') as an overlooked dimension of institutional design that matters for both the accuracy and the equity of human-capital evaluation. 

Presentations: SJTU (2025), Peking  U NSD (2025), NEUDC (2025)

I study the impact of corruption crackdown on human capital supply for the state, exploiting China's staggered anti-corruption inspections. Using unique applicant data from state organizations, I find that anti-corruption induces positive selection for integrity and prosociality into the state sector, without significantly affecting overall ability. These shifts in supply are associated with enhanced work performance. Changes in occupational preferences corroborate static talent allocation as a salient mechanism, in which treated honest types show higher preferences for state jobs -- even when conditioned on ability and family background. I further document dynamic effects wherein households increase investment in human capital and the integrity of the next generation. Together, these findings highlight reward structures as an important determinant of the state sector's human capacity. 

Presentations/Awards: Stanford DevPEC (2022), NEUDC (2022), ASSA (2023); Rosenstein-Rodan Prize  (2024)

We study favoritism in the selection of elite scientific talent, by examining the relationship between host institution affiliation and performance in the Chinese Science Olympiad, where a gold medal guarantees a student's admission to top universities. Using hand-collected participant-level data (2003 - 2021), we find that host-affiliated students have a significantly higher winning probability, and the effect is more pronounced in host provinces where corruption norms are more prevalent. We present evidence suggestive of cheating behavior using a portion of the contest vulnerable to information leakage, as well as the centralized post-Olympiad selection outside the control of host provinces. Together, our findings shed light on the crucial role of organizational structure in designing equitable assessment systems for talent. 

Many societies hold traditional culture tied to psychological factors that are socioeconomically relevant.  This paper shows its substantial impact on public leadership and governance, leveraging widely-held "zodiac year" beliefs about risk-avoidance in rural China, which follow an exogenous 12-year cycle tied to a person's birth year. Using a representative village panel, I find that village heads in their zodiac years follow governance processes more and enhance villagers' perception about responsiveness. I also observe consistent expenditure changes, with higher public good spending and a comparable decline in administration spending that is prone to misuse. However, treated leaders are also less likely to promote policy innovation. These results can be most easily reconciled with a shift in village heads' risk-taking, which can yield a potential trade-off between accountability and public entrepreneurship.

We study the impact of China’s Southbound Bureaucrat Program (1948-1950), the world’s largest state workforce migration — which sent around 100,000 front-line bureaucrats to newly liberated South China. Digitizing historical records, we characterize positive selection on bureaucrats’ socialist ideology. We find southbound bureaucrats foster a more pro-socialist development trajectory over half a century. In the short run, they better promote communist policies. In the longer run, more affected regions have lower inequality and greater welfare provision, likely attributed to stronger state interventionism and human capacity. Further analysis suggests bureaucrats’ socialist ideology plays a pivotal role. Their influence persists via local personnel and cultural spillovers. Collectively, our findings highlight  that bureaucrats can substantially shape short and longer-run development trajectories. 

Funding: BU GDP Field Fellowship, Manual A. Abdala Fund, RUC Scientific Research Fund