The Schedule

See you next year!

Speaker: Lei Li

Title: Skill-Biased Imports, Human Capital Accumulation, and the Allocation of Talent,

Abstract: This paper proposes that imported capital goods, which embody skill-complementary technologies, can lead to an increase in the supply of skill in developing countries like China. By exploiting the cross-prefecture variation in imported capital goods, I show that the surge in imported capital goods encourages human capital accumulation and migration in China. To tackle causality, I instrument a prefecture's import growth of capital goods with that in other regions. There are three main findings. Firstly, the regional difference in imported capital goods can explain 27 percent of the regional difference in college share between 2000 and 2010. A prefecture with a $100 increase in imported capital goods per capita had a 1.4 percentage points increase in college share. Secondly, this paper quantifies the importance of the three channels, namely skill acquisition of local stayers, immigration of skilled workers, and emigration of skilled workers, through which imported capital goods increase college share. I find that the first channel is the most important. Thirdly, I trace out the responses of skill supply to the demand shift. I find that imported capital goods increase college wage premium and the effect attenuates over time with the increase in skill supply

Speaker: Jingshu Wen

Title: Incentivization or expropriation? All ESOPs are not created equal

Abstract: Firms often claim that they adopt Employee Stock Ownership Plans (ESOPs) to incentivize employees, but we find Chinese listed firms could also tunnel through ESOPs during 2014 to 2018. Chinese listed firms with more independent boards and higher salary level are more likely to adopt ESOPs to incentivize employees, while those with higher leverage, more previous tunneling, and higher separation of ownership and control are more likely to adopt ESOPs to expropriate minority investors. Controlling shareholders with tunneling motives tend to use earnings management, leveraged ESOPs, and ESOPs with high participation rates to inflate the stock prices and then cash out soon after ESOP adoption announcements. Their tunneling results in a decrease in productivity, operating cash flow, and stock price, as well as an increase in delisting risk after ESOP adoption. Controlling owners in firms with poor outlook choose this myopic form of tunneling, as they can divert more from new investors than from existing corporate assets. Our research, together with previous studies, shows that controlling shareholders may tunnel in more covert ways after regulatory tightening, as long as their return from tunneling still far exceeds their return from investing in positive-NPV projects.

Speaker: Tengfei Zhang

Title: Manager Uncertainty and Cross-Sectional Stock Returns

Abstract: This paper evidences the explanatory power of managers’ uncertainty for cross-sectional stock returns. I introduce a novel measure of the degree of managers’ uncertain beliefs about future states: manager uncertainty (MU), defined as the count of the word “uncertainty” over the sum of the count of the word “uncertainty” and the count of the word “risk” in filings and conference calls. I find that manager’s level of uncertainty reveals valuation information about real options and thereby has significantly negative explanatory power for cross-sectional stock returns. Beyond existing market-based uncertainty measures, the manager uncertainty measure has incremental pricing power by capturing information frictions between managers’ reported uncertainty and investors’ perception of uncertainty. Moreover, a short-long portfolio sorted by manager uncertainty has a significantly positive premium and cannot be spanned by existing factor models. An application on COVID-19 uncertainty shows consistent results.

Speaker 1: Baptiste Souillard

Title: Import Competition and Corporate Tax Avoidance: Evidence from the China Shock,

Abstract: This paper examines the effect of import competition on corporate tax avoidance. I exploit the rapid surge of China’s exports as a competition shock and balance sheets and income statements to measure tax avoidance of US-headquartered publicly listed manufacturing firms. The baseline results reveal that a 1 percentage point in- crease in the penetration ratio of US imports from China entails, on average, a 0.20 percentage point decrease in the effective tax rate. They are supported by a series of sensitivity tests and robust to using the US conferral of the Permanent Normal Trade Relations status on China in late 2000 as a quasi-natural experiment. Furthermore, the results are entirely driven by multinational firms. In response to the China shock, these firms invested in intangible assets, and these intangibles allowed them to shift more profits towards low-tax countries. These findings shed light on the determinants of corporate tax avoidance. More generally, they help understand the decline in the average effective tax rate of US publicly listed firms and the recent backlash against large firms and globalization.

Speaker 2: Lu Feng

Title: Non-tariff Measures and Firm’s Export Quality - Evidence from China,

Abstract: With the success of the GATT/WTO in reducing conventional tariff barriers, much of the recent focus of both regional and multilateral trade agreements has switched to non-tariff measures (NTMs). However, the effect of NTMs is still unexplored, especially on firms. In this paper, we focus on an important aspect of export firm performance---product quality. We use information from Chinese government's official documents to construct a novel dataset on China's NTMs by 2012, and classify these NTMs according to whether it's related to product standard, and which government department issued it. Using this dataset, we analyze how the NTMs on firms' importing input affect their export product quality. We find that NTMs related to product standards or issued by non-trade authorities (mainly for health, environment or other public interest) are positively correlated with export quality, while the impact of other NTMs or issued by trade authorities (may have protection purpose) is not significant.

Speaker: Zhiting Wu

Title: Firm Heterogeneity in Production-Based Asset Pricing: The Role of Habit Sensitivity and Lumpy Investment,

Abstract: I study the linkage between risk premiums and lumpy investment in both time-series and cross-sectional levels. To this end, I introduce a logarithm surplus-consumption ratio into habit preference that produces 100% equity volatility by volatile SDFs and robustly matches investment dynamics under non-convex costs. Second, my model reproduces almost 100% equity premiums because low logarithm surplus-consumption ratios and adjustment probabilities assign additional weights on precautionary savings and constrained firms, respectively. My model also generates considerable size premiums since small and constrained firms are riskier. Finally, my model reasonably matches crucial moments of macro-dynamics and the cross-sectional investment rate.

Speaker: Shiqi Guo

Title: Chinese Aid and Local Employment in Africa,

Joint with Haicheng Jiang (Queen's University Belfast)

Abstract: China has played an increasingly important role in global aid. Literature has found that Chinese aid boosts economic growth and could also have deeper social impacts in the recipient countries. However, how the local residents are involved in Chinese projects and are thus affected in such a profound way has not been examined clearly. This paper sheds light on the first-order question regarding the impact of Chinese infrastructure aid projects on local employment. We exploit the spatial and temporal variations in Chinese infrastructure aid projects and different waves of household surveys in ten African countries during 2000–14, and identify the change in individual employment after the local Chinese projects become active, based on a difference-in-differences type of estimation. We find that the impacts exist mostly in the short term during the construction of projects. Local employment increases by two percentage points in the first year and by four percentage points in the second year after Chinese projects started. The effect diminishes after the third year. The Chinese aid projects are likely to create job opportunities for local residents both directly and indirectly through relevant sectors, as more employment is observed for manual labor and also in the service sector. More year-round and cash-earning jobs are provided. Urban, female, and youth employment is more benefited. Among the different infrastructures, the impact of roads and railways exists mostly in the short run, while the impact of schools, hospitals, and water and electricity facilities is large and significant both in short term and over longer timeframes. To isolate the impact of Chinese aid projects, we have also addressed the potential confounds from other local developmental resources in various ways.

Speaker 1: Zhesheng Zheng

Title: Structural Regularization,

Joint with Jiaming Mao (WISE, Xiamen University)

Abstract: We propose a novel method for modeling data by using structural models based on economic theory as regularizers for statistical models. We show that even if a structural model is misspecified, as long as it is informative about the data-generating mechanism, our method can outperform both the (misspecified) structural model and un-structural-regularized statistical models. Our method permits a Bayesian interpretation of theory as prior knowledge and can be used both for statistical prediction and causal inference. It contributes to transfer learning by showing how incorporating theory into statistical modeling can significantly improve out-of-domain predictions and offers a way to synthesize reduced-form and structural approaches for causal effect estimation.

Speaker 2: Jingzhi Xu

Title: Ensemble Learning with Statistical and Structural Models,

Joint with Jiaming Mao (WISE, Xiamen University)

Abstract: Statistical and structural modeling represent two distinct approaches to data analysis. In this paper, we propose a set of novel methods for combining statistical and structural models for improved prediction and causal inference. Our first proposed estimator has the doubly robustness property in that it only requires the correct specification of either the statistical or the structural model. Our second proposed estimator is a weighted ensemble that has the ability to outperform both models when they are both misspecified. Experiments demonstrate the potential of our estimators in various settings, including fist-price auctions, dynamic models of entry and exit, and demand estimation with instrumental variables.

Speaker 1: Juncong Guo (Antai College of Economics and Management, Shanghai Jiao Tong University)

Title: The educational anxiety of Chinese parents: competition in shadow education

Abstract: We examine the competition in shadow education investments among students’ parents under a social interaction framework by using a representative survey of middle school students in China. Based on the data structure and the identification issues, we apply spatial Tobit models and substantiate that the competition does exist but will be overestimated if we ignore the reflection problem. Moreover, the competition effect is higher of the more observable outcome for agents in the same reference group. Heterogeneity analyses indicate that the competition reaction is stronger for parents with 9th-grade children and for parents whose children may have competitive disadvantages in the future. The competition is more intense among parents who have similar attributes and during the beginning of an academic year. As for the consequences, the competition is inefficient for parents in the short term but benefits shadow education companies.

Speaker 2: Ji Luo (School of Public Policy & Management, Tsinghua University )

Title: Transport Infrastructure Connectivity and Conflict Resolution: The Comparisons of Developed and Developing Countries

Abstract: Transport infrastructure connectivity has strong endogeneity issues, which makes it difficult to directly assess its impact on local conflict resolution. Based on global conflict data from 2010 to 2017, this paper adopts a series of machine learning and natural experiment methods to investigate the impact of transport infrastructure connectivity on global conflict resolution. The empirical results show that (1) transport infrastructure connectivity can significantly improve countries’ global ranking for conflict resolution (an average of 6-13 ranking positions), and, in particular, the marginal benefit of developed countries is greater than the one of developing countries; (2) the mechanism behind this effect is the promotion of trade facilitations, a more balanced employment ratio across genders, and improved income levels through transport infrastructure connectivity, which further enhances the conflict governance capacity of countries. This study uses new methods and a new perspective to reveal the important contribution of infrastructure connectivity to global conflict governance.