Working Papers

Reputation, Media Capture and Bias [Code]
This paper discusses the media bias induced by media firm's inability of getting accurate information, firm's reputation concern and the existence of media capture. The study finds that in low quality firms, there appears to be higher incentive to distort information and to follow prior beliefs. Reputation concern tends to drive the media firms to increase the precision of their reports, in some cases, firm's reputation rises when anti-prior belief reporting is chosen. Thus, media firms face the balance between anti-prior and pro-prior belief reporting. As for media capture, by modeling electorate's behavior in political election, the study suggests that if the update of beliefs based on media report is significant, media capture would arise. In particular, candidate of non-preferred type has stronger incentive to capture media. Media capture also shows external effects as non-captured firms becomes more likely to report the candidate as preferred type as well. For welfare, reputation concern appears to hurt individual welfare since it distorts the precision of reporting, whilst in media capture, individual welfare is determined by the probability of the media being captured.

Redistribution Effect and Pension Choice: Theory and Evidence
This paper mainly focuses on two issues, the factors influencing pension choice and the redistribution effect of the pension system in China. Our model studies the trade-offs of relative financial benefits and risks provided by various plans, as well as the accessibility to specific pension plans and accessibility to information on pensions. The features examined include individual features such as hukou, gender and education, family features like marital status and work features like job types. By analyzing these features, we find that features of individual, family and work have significant impacts on the choice of pensions. Consistent behavior on pension choice also appears to exist within a family. Moreover, the results indicate that education and physical health of family members have a positive impact on each other's pension behavior. The above results appear to be consistent with the prediction of our model. Finally, we take the Newly Rural Social Pension (NRSP) as an example to study the existence of redistribution effect. The effect is found between people with different income level and gender, and the direction of the redistribution is from people with higher income to those with lower income, from male to female.

Work in Progress
  1. Rui Sun and Hulai Zhang (June 2017), Early Career Life of CEOs and Firms' Innovation
  2. Rui Sun and Hulai Zhang (April 2017), Gender Discrimination within Family: Evidence from Woman's Education Attainment