主持人（上午）：马湘君，弗吉尼亚大学经济学博士，辽宁大学经济学部教授，研究方向为国际贸易、应用微观经济学、跨国公司与国际税收，所著论文发表于Journal of International Economics等一流SSCI期刊，主持国家自然科学基金、北京市社会科学基金等青年项目。曾任对外经济贸易大学副教授，北京大学国家发展研究院访问学者，德勤税务（美国硅谷办公室）、天职美国（加州尔湾办公室）转移定价经济学家/经理，并曾借调于商务部世界贸易组织司。

主讲人1：Lan Lan（Clemson University）

时间：2023年1月11日（周三）8:30-9:30AM（北京时间）

地点：腾讯会议：164-654-310

语言：英文

题目：Job Match Quality of College Graduates: The Effect of Major-Specific Market Size

摘要：Previous theoretical research suggests that larger cities improve job matching. Yet, empirical evidence on this relationship is mixed. This paper develops a novel measure of major-specific labor market size for college graduates and examines its effect on job match quality. I find that major-specific labor market size is predictive of higher job match quality. However, the overall labor market size has an inconsistent effect on job match quality. Moreover, I find the effects of major-specific labor market size tend to be more significant for men and younger workers.

主持人（下午）：鄢萍，经济学教授。2008年毕业于美国德克萨斯大学奥斯汀分校，获经济学博士学位。研究领域为宏观经济学、产业经济学。曾在经济学国内外顶级期刊Economic Journal, Rand Journal of Economics，管理世界，经济学（季刊）上发表多篇学术论文。主持四项国家自然科学基金。

主讲人1：Hongbing Wei（Tsinghua University）

时间：2023年1月11日（周三）2:00-3:00PM（北京时间）

地点：腾讯会议：164-654-310

语言：英文

题目：Super Factory Comes to Town: Identifying the Agglomeration Spillovers from Foxconn Factory in Henan

摘要：This paper investigates the spillover effects of super factories on local manufacturing sector through agglomeration in developing economies. We use evidence from the largest provider of electronic manufacturing service in the world: Foxconn. Using its unexpected investment in Henan province of China as a quasi-natural experiment, we adopt a geographic difference-in-difference approach that exploits the spatial decay of the agglomeration effects from Foxconn factories. We find that the increase in the labor productivity of the regions closer to Foxconn factories is larger than those of regions further away. This finding is robust to a battery of validity checks. Consistent with the agglomeration spillovers, firm entry and labor costs increase more in regions closer to Foxconn factories than further away.

主讲人2：Yongkun Yin （CEMFI）

时间：2023年1月11日（周三）3:10-4:10PM（北京时间）

地点：腾讯会议：164-654-310

语言：英文

题目：Missing Women: A Quantitative Analysis.

摘要：The sex ratio, males per female, is well above one in China, India, and other South and East Asian countries. Parents in these countries want more boys, exercise sex-selective abortions, and invest less in their daughters' education. Why do parents favor sons over daughters? What policies can be effective in normalizing the sex ratio? To answer these questions, I build an overlapping-generation model of fertility, sex selection, the quantity-quality trade-off, and marriage and estimate it for India. The quantitative analysis reveals that the main drivers of missing women are economic factors, i.e., old-age support by sons, dowry payment for daughters, and labor market discrimination against women. If the gender differences in these economic factors are removed, the sex ratio at birth (SRB） would reduce from 1.14 to 1.05. The fertility rate would drop from 3.1 to 2.4, and the share of women with secondary education would increase from 49% to 72%. The sons would also benefit from lower fertility, and the share of men with secondary education would rise from 65% to 79%. Once the economic factors become gender-neutral, eliminating intrinsic son preferences has a small additional effect. A subsidy for female births or female education, commonly-implemented policies in India and elsewhere, can both reduce the SRB. However, the former increases fertility and reduces children's education and women's labor supply, while the latter has the opposite effects. Finally, a pay-as-you-go pension system can lower the SRB to 1.09, but it also reduces children's educational attainment, as parents value them now less.

主持人（晚上）：游宇，教授，博士生导师，新兴经济体研究会理事。现任辽宁大学李安民经济研究院副院长。长期从事国际金融、经济增长、宏观金融和微观金融等领域的研究。现有SSCI和CSSCI期刊论文20余篇，发表刊物包括《经济研究》《金融研究》《Journal of Banking and Finance》等国内权威期刊和国际著名期刊。主持国家社科基金项目和教育部人文社科项目，并参与多个国家自科和教育部社科项目；主持辽宁省服务业发展“十四五”规划编制工作。入选2019年度辽宁省“兴辽英才计划”青年拔尖人才和2020年度辽宁省“百千万人才千层次”人才。

主讲人1：Qianqian Shang （CEMF） 

时间：2023年1月11日（周三）7:10-8:10PM（北京时间）

地点：腾讯会议：164-654-310

语言：英文

题目：Internal Migration and Education: The Role of Old-age Support

摘要：In most developing countries, pension systems are poorly developed and the elderly rely on their children for old-age support. However, with increasing trends in rural-to-urban migration, many elder parents are left alone with limited assistance. This paper investigates whether parents may strategically lower the education investment in children to reduce the probability of their children out-migrating. In particular, I examine whether the Hukou reforms, which greatly relaxed migration restrictions in rural China, had detrimental effects on educational investments. Exploiting the variation in the timing of exposure to Hukou reforms across prefectures and cohorts, I show that the probability of children being enrolled in high school decreases by 0.1 standard deviations if the migration restrictions at high-school entry age are relaxed by one standard deviation. The effect is larger for parents whose ideal living arrangement at old age is to live near their children. Furthermore, the negative effects only exist for parents without pension entitlement. The results suggest that in contexts with weak policies for old-age support, lifting migration restrictions may have detrimental effects on human capital accumulation.

主讲人2: Yiren Wang（Singapore Management University）

时间：2023年1月11日（周三）8:20-9:20PM（北京时间）

地点：腾讯会议：164-654-310

语言：英文

题目：Panel Data Models with Time-Varying Latent Group Structures

摘要： This paper considers the linear panel model with interactive fixed effects such that individual heterogeneity is captured by latent group structure and time heterogeneity is captured by an unknown structural break. We allow the model to have different number of groups and/or different group memberships before and after the break. With the preliminary estimates by nuclear norm regularization followed by row- and column-wise linear regression, we estimate the breaking point by the idea of binary segmentation and the latent group structure together with the number of groups before and after the break by sequential testing K-means algorithm simultaneously. Asymptotic theory shows that the breaking point and the number of groups as well as the group membership can be estimated correctly with probability approaching one. Finite sample performance of the estimation is illustrated via Monte Carlo simulation and a real dataset application.