Research

Publications

从“中国人必须走”到“中国病毒” ——美国社会对华态度的历史根源（合作者：刘丛，薄诗雨）《经济学（季刊）》已接收
[From "the Chinese Must Go" to "Chinese Virus": The Historical Origins of Sinophobia in the United States (with Cong Liu and Shiyu Bo). Accepted by China Economic Quarterly

Abstract: This paper studies the long-term effect of historical anti-Chinese incidents on current public attitudes towards China in the United States. Using an unexpected diplomatic event—President Trump publicly referred the coronavirus as “Chinese virus” on Twitter—as an exogenous shock, we combine historical archives with daily Google search data to conduct a difference-in-differences analysis. The results show that residents in regions that witnessed more historical anti-Chinese incidents are more likely to search “Chinese virus” on Google after Trump's twitter post. The heterogeneity analysis by regional characteristics shows that the persistence of Sinophobia is stronger in regions with higher exposure to Chinese import competition and lower Asian population share.

How Restricted is the Job Mobility of Skilled Temporary Work Visa Holders? (with Jennifer Hunt). Journal of Policy Analysis and Management 2019 Vol.38(1)

Abstract: Using the National Survey of College Graduates, we investigate the job mobility of skilled workers holding U.S. temporary visas. Such workers either have legal restrictions on their ability to change employers or may be reluctant to leave an employer who has sponsored them for permanent residence. We find that the voluntary job-changing rate is similar for temporary work visa holders and natives with similar characteristics, but that it spikes when temporary work visa holders obtain permanent residence. The spike magnitude implies mobility is reduced during the application period by about 20 percent, alleviating concerns that employers exercise strong monopsony power.

Working Papers

Institutional Discrimination and Assimilation: Evidence from the Chinese Exclusion Act of 1882 (with Shuo Chen). Under Review
[Latest version] [Online Appendix] [IZA Discussion Paper]

Abstract: The Chinese Exclusion Act of 1882 banned Chinese immigration and institutionalized discrimination against Chinese immigrants in U.S. society. This study examines the impact of institutional discrimination on the assimilation of Chinese immigrants in the US. Using linked individual samples and birth cohort samples from historical US censuses and digitized archives, We exploit the passage of the Act and construct the state-level measure of the intensity of discrimination using the vote result of the Act and the frequency of anti-Chinese incidents to conduct difference-in-differences and triple-difference estimations. The difference-in-differences estimates show that discrimination severely impeded the labor market assimilation of Chinese immigrants in the Exclusion Era (1882--1943). In response to discrimination, Chinese immigrants made significantly more efforts to assimilate: post-1882 born Chinese cohorts were more likely to attend school, become literate, speak and write English, and adopt Americanized names. The triple difference estimates show that these responses were significantly stronger in states with a higher share of approval votes on the Act or a higher frequency of anti-Chinese incidents.

Overseas Chinese Networks and Processing Trade in China (with Sen Ma and Yan Zhou).

Abstract: Processing trade plays a key role in China's trade liberalization since the economic reform in the 1970s. Using a newly collected dataset of county-country paired overseas Chinese network matched with firm-country paired trade data, this study empirically show that overseas Chinese networks significantly increase the processing exports of firms in China using fixed-effects models. We show that overseas Chinese networks promote processing trade by attracting foreign investments and stimulating outsourcing of production from high-income and middle-income countries in Europe and Asia to China. We also show that overseas Chinese networks reduce information frictions to promote processing trade. Finally, we find that overseas Chinese networks are effective in generating large initial trade volumes but play a limited role in expanding the size of existing trade relationship over time.

The Aggregate and Distributional Effects of Immigration Restrictions: Immigration Quotas and the Great Black Migration. Under Review
[Latest version] [Online Appendix]
[Old version: The Effects of Immigration Quotas on Wages, the Great Black Migration, and Industrial Development]

Abstract: This study examines the labor market effect of the negative immigration shock caused by the US immigration quota system between 1920 and 1930 and identifies the causal impact of immigration restrictions on the Great Black Migration. Using county-level panel data, I find that immigration restrictions did not affect average wages and lowered the wages of US-born whites and immigrants at the aggregate level. The analysis based on linked individual samples reveals substantial internal migration across counties and distributional effect caused by the immigration shock: in-migrants to counties more affected by the shock had greater wage gains while non-movers suffered a sharper decline in wages. In particular, the negative immigration shock causally and substantially increased the migration of black southerners to northern counties. Black migrants who moved to more affected counties earned higher wages, and they were more likely to become literate and work in urban areas as low-skilled manufacturing workers.

Can High-Skilled Immigrants Transfer Their Human Capital to the US?. Revise and Resubmit at Southern Economic Journal

Abstract: Using the National Survey of College Graduates, I estimate the return to human capital of high-skilled immigrants in the US to study the transferability of their pre-migration human capital. I also examine the difference in the transferability of human capital between STEM and non-STEM immigrants as well as across immigrant groups who initially migrated with different types of visa. I find the following: (1) high-skilled immigrants transfer partial pre-migration education and virtually no work experience to the US; (2) STEM immigrants transfer more pre-migration work experience than non-STEM immigrants; (3) English skills increase the transfer of immigrants’ work experience to the US and the effect is stronger among non-STEM immigrants; (4) immigrants who initially entered with temporary work visas can transfer more pre-migration work experience than other immigrants.

The Fruit of Egalitarianism: People's Communes and Population Growth in Rural China (with Shuo Chen).

Abstract: People’s Communes in rural China implemented equal income distribution that strongly subsidized the fertility behaviors of commune members. This study exploits the variation in the pace of the agricultural reform across counties in the late 1970s to examine the causal effect of the abolishment of the People’s Communes on rural fertility. The difference-in-differences estimates show that the reform reduced the annual number of rural births by 4.5%. On the other hand, the reform had no substantial effect on fertility in urban area or mortality. We further find that rural fertility declined more sharply after the reform in counties where incomes had been distributed more equally in the commune period, which suggests that the abolishment of equal income distribution was the driving force behind the decline in rural fertility in post-reform China.

Instruction Time, Shadow Education, and Academic Performance. Under Review

Abstract: Based on a panel survey of secondary school students in China, I use fixed effects model to examine the effect of instruction time in school on students’ shadow education and academic performance and how this effect varies by family socioeconomic status. The results show a negative effect of school instruction time on private tutoring attendance, which suggests the substitution between instruction time and shadow education. I also find that school instruction time and shadow education positively affect students’ test scores. When examining the heterogeneous effect of instruction time by family background, I find that the elasticity of substitution between instruction time and private tutoring is in fact higher among low-income and low-education families, meaning that their demand for private tutoring increases more when instruction time decreases. Eventually, I find that instruction time has a smaller positive effect on the test scores of students from low-income and low-education families. These findings suggest that the reduction in instruction time is unlikely to widen the gaps in educational resources and academic performance of students across families with different SES.

Equal Education, Unequal Outcomes? (with Jinseong Park).

Abstract: A nationwide educational reform in South Korea, known as the Equalization Policy, switched the high school admission system from sorting to mixing by assigning students into schools based on their residential location. Exploiting the regional and temporal variation in the implementation of the policy in the 2000s, we examine its impact on college admission and the earnings of college graduates. We document the following findings: first, the policy substantially redistributed high-performing students from a few selective high schools to other schools, thus equalizing the peer composition across schools; second, the policy reduced the probability of entering college; third, the policy lowered the earnings of college graduates on average but the negative effect on the earnings was concentrated on students from disadvantaged families, which likely lowered the intergenerational income mobility.