Research

Selected publications

“Snowball Sampling and Sample Selection in a Social Network.” Advances in Econometrics: Volume 42 (Forthcoming).

“The ‘Value’ of Propaganda: Predicting Political Crackdown in China Based on Text Data” (with Weifeng Zhong), Yale Journal of International Affairs, Spring 2020.

“Marriage-Induced Homeownership as a Driver of Housing Booms: Evidence from Hong Kong” (with William Ka Shing Cheung and Paavo Monkkonen), Housing Studies, August 6, 2019.

“All the General Secretary’s Men” (with Weifeng Zhong), AEIdeas, October 2017.

Working Papers

Reading China: Predicting Policy Change with Machine Learning (Paper)

(Joint with Weifeng Zhong)

For the first time in the literature, we develop a quantitative indicator of the Chinese government’s policy priorities over a long period of time, which we call the Policy Change Index (PCI) of China. The PCI is a leading indicator of policy changes that covers the period from 1951 to the third quarter of 2018, and it can be updated in the future. It is designed with two building blocks: the full text of the People’s Daily — the official newspaper of the Communist Party of China — as input data and a set of machine learning techniques to detect changes in how this newspaper prioritizes policy issues. Due to the unique role of the People’s Daily in China’s propaganda system, detecting changes in this newspaper allows us to predict changes in China’s policies. The construction of the PCI does not require the researcher’s understanding of the Chinese context, which suggests a wide range of applications in other settings, such as predicting changes in other (ex-)Communist regimes’ policies, measuring decentralization in central-local government relations, quantifying media bias in democratic countries, and predicting changes in lawmakers’ voting behavior and in judges’ ideological leaning.

Additive Model with Endogenous Network Formation (Slides)

Birds of a feather flock together: peer selection is prevalent in social interactions. This paper proposes a novel identification strategy for social interactions in an additive model with endogenously formed social network. The network endogeneity arises from the correlation between the links of the network and the unobservables that determine the outcome of interest. I show that the eigenvectors of the adjacency matrix that define the social network are control variables for network endogeneity without imposing any parametric assumption. This identification strategy can be applied to a wide range of networks such as spatial networks, networks of organizations or networks of flow data. The eigenvectors can be interpreted as the latent variables that drive the network endogeneity. Under a restriction on the dimension of the latent variables in the network formation, a subset of eigenvectors captures all the network endogeneity. I propose an information criterion to select the number of eigenvectors to be included as control variables. I use this method to estimate peer effects of friends and studymates on academic achievement using a social network data collected in Chan and Lam (2014). I find that the peer effect of studymates is reduced by 50% (about one standard deviation) after controlling for the selection of friends and studymates, this suggests substantial biases in peer effects estimates that do not account for endogenous peer selection.

Type of Peers Matters: A Study of Peer Effects of Friends, Studymates and Seatmates on Academic Performance (joint with Chungsang Tom Lam)

This paper studies the peer effects of friends, studymates, and seatmates on academic performance. We obtain the information of social networks, personality traits, and cognitive ability measures from a unique data set based on a survey we conducted in three schools in Hong Kong. We estimate a social interaction model which accounts for endogenous network formation and correlation between multiple networks. Our results show that the cognitive ability of studymates and the conscientiousness of friends positively affect a student’s mathematics exam score, while the conscientiousness of studymates and the cognitive ability of friends do not produce such an effect. We find that students with elder siblings are less affected by the cognitive ability of studymates, but the effect of conscientious friends is not related to that. By contrast, we find no such effect from seatmates. These results are consistent with the idea that studymates influence each other through discussing and teaching where cognitive ability is essential. On the other hand, friends influence each other by creating an atmosphere of studying, so being conscientious is valued. Our study hint at the way different types of peer work and which particular qualities are important for each peer type.

Work in Progress

Endogenous Organizational Structure and Internal Allocation of Capital in Banking (Coming Soon) (joint with Vladimir Yankov)

We document that all large US bank holding companies have organizational structures composed of a large number of individual subsidiaries with different geographic or functional specialization. Moreover, and much less known, the organizational structures across different holding companies vary in their network characteristics. We combine information from the organizational network structure of bank holding companies with data from the balance sheets and income statements of the holding company bank and non-bank subsidiaries to construct a unique dataset of the internal allocation of funds within a conglomerate. We document that these internal capital flows are sizeable and increase with the cost of external funding. Further, banks within a conglomerate with higher centrality in the organizational network receive disproportionally higher funding controlling for profitability and investment opportunities. We also show that previous research on the internal allocation of capital in banking, which did not control for the network structure and the endogeneity of organizational structure of bank holding companies, significantly underestimates the role of the internal capital markets.