Research Fields
Labor Economics, Health Economics, Public Economics
Publications
The Intergenerational Effects of Housing Regulations: the Regression Discontinuity Estimates on Child Development (with Han Li, Jiangyi, Li and Yi Lu), Journal of Public Economics, Vol. 227 104995. 2023
This paper studies the intergenerational effects of housing policies on child development. Exploiting a policy-induced discontinuity in house size in China, we identify that favorable housing policies significantly reduce children’s cognitive skills, lessen their beliefs in an internal locus of control, and decrease their self-esteem but have little impact on physical health or depression symptoms. The heterogeneous analyses show that children at a critical stage of skill formation and girls respond more strongly to housing policies. Our mechanism decomposition shows that parenting skills play a large role in the effect of housing wealth on child development.
Pay for Performance Scheme and Labor Productivity: Evidence from a Kinked Payment in China (with Xiqian Cai, Hong Song, and Wei Jiang), Journal of Development Economics, Vol 156 102840. 2022
This paper examines workers’ response to a nonlinear payment system with a penalty and a reward design in a manufacturing firm in China. Using bunching methods, our estimates show that about 31% of workers make decisions with errors, and the structural elasticity of output response to the piece rate is 0.28. The focal nonlinear payment scheme generates 4% output gains compared with a linear payment scheme given the same total wage costs. Further decomposition analysis shows that the reward aspect of the nonlinear scheme mainly contributes to total output gains. Counterfactual analyses show that total output generated by the nonlinear payment system falls with the reduced worker heterogeneity. Our results illustrate the role of worker heterogeneity in designing compensation schemes and provide new insights into the increasing adoption of nonlinear payment systems in modern manufacturing production.
Education and Mental Health: Evidence and Mechanisms (with Wei Jiang, Yi Lu), Journal of Economic Behavior & Organization, Vol 180: 407-437. 2020
In this paper, This paper investigates how education affects mental health outcomes. The investigation exploits the compulsory schooling laws implemented in the mid-1980s in China. The laws generated dramatic kinks with small jumps in educational attainment across cohorts. Using a regression probability jump and kink design, we find a strong positive causal relationship between education and mental health outcomes in adulthood. Our mechanism analysis suggests that resources, cognition, and social integration play significant roles in the effects of education on mental health outcomes in the long run.
Housing Wealth and Labor Supply: Evidence from a Regression Discontinuity Design (with Han Li, Jiangyi Li and Yi Lu), Journal of Public Economics, Vol. 183 104139. 2020
This paper uses the discontinuity in house size generated by Chinese housing policies to identify the effect of housing wealth on labor supply. The analysis finds a substantial deterrent effect of housing wealth on labor supply. Much of the housing wealth effect comes from the labor participation decision, with the effect on employed workers being ambiguous. Females and elderly individuals are more responsive to gains in housing wealth.
The Average and Distributional Effects of Teenage Adversity on Long-Term Health (with Jie Gong, Yi Lu), Journal of Health Economics, Vol. 71 102288. 2020
A central question in human development is what causes health inequalities over the life cycle. This paper links adversity in the teen years to individuals' long-term health outcomes. We examine a mandatory rustication program, the"send-down" policy during China's Cultural Revolution, and employ Regression Discontinuity Design to estimate the impact on people's physical and mental outcomes 40 years later. Our results suggest that rusticated youths were more likely to develop mental disorders but not to have worse physical outcomes. Further assessing distributional effects through marginal treatment effect (MTE), we find strong heterogeneous treatment effects and selection on gains.
Long-Term Impact of Trade Liberalization on Human Capital Formation (with Jie Li, Yi Lu, and Hong Song), Journal of Comparative Economics, Vol. 47(4), pp. 946-961, 2019
While a growing line of research has assessed the effect of trade liberalization on human capital formation, most of these studies focus on its short-term effect on individual’s school attendance. Much less is known about its long-run effect, as well as the impact on other aspects of human capital formation such as labor market and noncognitive outcomes. This paper studies the impact of trade liberalization on individuals’ long-term human capital accumulation, including school attendance, cognitive abilities, labor market performance, and noncognitive outcomes. By constructing prefecture-year-level tariff barriers, our identification strategy exploits variations in different cohorts’ exposure to a trade shock at age 16 for individuals within the same prefecture. Empirical results suggest that trade liberalization leads to decreased completed years of schooling, cognitive abilities, wage, and noncognitive outcomes. We provide suggestive evidence that this observed pattern is explained by the expansion of job opportunities in relatively low-skilled and labor-intensive sectors.
Telecommunications Externality on Migration: Evidence from Chinese Villages (with Yi Lu, Colin Xu), China Economic Review, Vol. 39, pp.77-90, 2016.
We use a unique data set of Chinese villages to investigate whether access to telecommunications, in particular, landline phones, increases the likelihood of outmigration. By using regional and time variations in the installation of landline phones, our difference-in-differences (DID) estimation shows that the access to landline phones increases the ratio of out-migrant workers by 2 percentage points, or about 51 percent of the sample mean in China. The results remain robust to a battery of validity checks. Furthermore, landline phones affect outmigration through two channels: information access on job opportunities and timely contact with left-behind family members. Our findings underscore the positive migration externality of expanding telecommunications access in rural areas, especially in places where migration potential is large.
OLS bias and MSE in a Unit Root Model (with Albert K. Tsui), International Journal of Statistical Sciences, Vol 22(1), 79-98. 2022
We revisit the problem of deriving analytically tractable expressions for lower order moments of the ordinary least squares (OLS) estimator in autoregressive models with unit roots. Simple algebraic techniques are used to approximate the series sums of the first two moments derived by Tsui and Ali (1991). Compared to the exact moment values obtained by numerical methods, it is found that our approximate closed forms in simple functions are reasonably accurate for a wide range of sample sizes. We also validate the numerical accuracy of asymptotic mean and variance derived by Shenton and Vinod (1995).
Working Papers
Causal Inference under Kink Bunching (with Yi Lu and Jianguo Wang)
This paper develops a generalized framework for identifying causal impacts in a reduced-form manner under kinked settings when agents can manipulate their choices around the threshold. The causal estimation using a bunching framework was initially developed by Diamond and Persson (2017) under notched settings. Many empirical applications of bunching designs involve kinked settings. We first propose a model-free causal estimator in kinked settings with sharp bunching and then extend to the scenarios with diffuse bunching, misreporting, optimization frictions, and heterogeneity. The estimation method is mostly non-parametric and accounts for the interior response under kinked settings. Applying the proposed approach, we estimate how medical subsidies affect outpatient behaviors in China.
Early Childhood Conditions and Adolescent Mental Health (with Bilge Erten, Pinar Keskin, Rodrigo Pinto and Lianming Zhu)
We investigate how early life circumstances induced by trade liberalization affect adolescent mental health in China, exploiting variation in tariff uncertainty faced by prefecture economies pre-2001. Our model differs from the classic difference-in-differences design in that it considers a moderator variable determining the intensity with which the treatment affects the outcomes. Our findings show that children born in prefectures more exposed to an exogenous change in international trade policy experienced a significant decline in the incidence of severe depression during adolescence. We find that the estimated relationships are robust to controls for initial prefecture attributes and other policy changes. Improvements in parental income, early childhood investments, and care provision in formal early childhood education programs are likely operative channels of impact.
Test Scores, Noncognitive Outcomes and the Stereotyping of Nonlocal Students (with Qinyue Luo)
This study investigates the impact of teachers' stereotyping of nonlocal students in terms of both academic performance and noncognitive outcomes. Based on a random assignment of a representative sample of Chinese middle school students to teachers within schools and grades, we find that biased beliefs against nonlocal students, particularly among Chinese teachers, adversely affect the academic performance of nonlocal students and lead to increased behavioural problems. No significant impact, however, is observed among local students. The results of a mechanism analysis suggest that these negative outcomes result from teachers' reduced engagement with nonlocal parents, weaker integration of nonlocal students into the classroom, and a decrease in nonlocal students' self-confidence. These detrimental effects are predominantly experienced by nonlocal boys, who report a less favourable classroom climate, exhibit decreased self-confidence, and expend less effort on their studies. In contrast, nonlocal girls tend to exhibit resilience by increasing their academic efforts. The findings of this study highlight the significant influence of teachers' stereotyping on disparities in human capital development between local and nonlocal students.
Nutrition and Child Development: Evidence from the Student Nutrition Improvement Program (with Dan Su and Peng Zhang)
This paper analyzes the effects of a school nutrition program on child development. By exploiting the staggered implementation of this program across counties in China, we find that the school nutrition program significantly improved physical health, cognitive development, and noncognitive and behavioral outcomes for exposed children. The heterogeneous analyses reveal that the effects are more pronounced among marginalized groups, including girls and children with mothers possessing lower levels of education. The mechanisms driving this impact include enhanced parental income, improved nutrition intake, increased monetary investments in children, and the adoption of more engaged parenting styles and interactive family dynamics. Furthermore, our cost-benefit analysis shows that the additional benefit resulting from enhanced noncognitive and behavioral outcomes constitutes approximately 48.9\% of the program's overall benefit, suggesting that previous studies have largely underestimated the benefits of school nutrition programs.
Adolescent Environment and Noncognitive Skills (with Jie Gong, Yi Lu)
Does the adolescent environment affect individuals' noncognitive skills? Using regression discontinuity design induced from a large mandatory program in 1960s China, we find that individuals who experienced adversity during teen years have a less external locus of control, i.e., are less likely to believe that external circumstances, such as luck, fate, or powerful others, control their lives. This effect on locus of control can explain around 10 percent of the wage differences caused by the adolescent experience. We interpret our findings as a long-run effect of the adolescent experience of adapting to adversity and expending effort that leads to rewarding. We also examine the external validity of our findings using marginal treatment effect (MTE) methods.
Education and Noncognitive Skills (with Jie Chen, Yi Lu)
In this paper, we estimate the causal relationship between schooling and non-cognitive skills in later life. We exploit two exogenous changes in education policy in China and Indonesia and estimate the effect of schooling on non-cognitive skills as measured by the Locus of Control and the five-factor model. We employ Regression Probability Jump and Kink Design (RPJK) for identification. Our results indicate that schooling leads to a less Internal Locus of Control, and makes individuals more conscientious, open and extroverted. The effect of schooling on non-cognitive skills can explain significant variation in returns to schooling on the labor market.
Sibling Effects on Employment and Occupational Outcomes
This paper studies the effect of a sibling’s education on long-run employment and occupational outcomes. The sibling effect is identified by the regression discontinuity design, which uses the implementation of the compulsory education law in China. Individuals are around four percentage points more likely to be employed if their siblings’ education increased by one year. They are also more likely to engage in less prestigious, routine manual, and non-routine physical occupations. The mechanism analyses show that the positive sibling education effect on employment operates through the increased social capital from better-educated siblings, and the negative effect on occupation is caused by the decreased human capital from unequal intra-household resource allocation among children.