“Media Control and Medical Expenses: Responsiveness to Water Pollution News Reports in China” (with Zhao Chen)

Abstract: This study is among the first to evaluate the impact of government media control on households’ welfare losses, as measured by medical expenses in the case of water pollution. We construct a unique data set of 377 Chinese local newspaper reports of 33 severe water pollution events occurring between 2004 and 2009. It shows that, in polluted areas, one water pollution report from local party newspapers was negatively associated with USD $2 household yearly medical expenses, while pollution itself was positively associated with USD $4 yearly household medical expenses. One potential explanation is that households who read pollution reports are prepared to protect themselves from health risks, while one report from party newspapers is positively associated with USD $0.20 yearly household bottled water consumption. Correlations with reports from non-party newspapers are less significant. The findings indicated that even though media control policies in China limit party newspapers’ coverages on critical news such as pollution, households still benefited from their news reports rather than non-party newspapers. This could be because news reports from party newspapers are more precise, so consumers trust them more. At least USD $30 million in household medical expenses were saved in urban China related with party-newspapers' water pollution reports between 2004 and 2009, based on our estimations.

“Local Labor Market Effects of Exports and Internal Migration : The Case of China” (Job Market Paper)

Abstract: To understand the impact of international trade on global labor markets, this paper examines the local labor market effects of rising exports in China by exploiting cross-regional variations in export exposure. The results show that, between 1990 and 2005, rising export exposure has increased Chinese manufacturing employment by around 19 million jobs. Given the same amount of changes in trade exposure, this positive impact on Chinese labor markets triples the negative impact on American manufacturing employment, as estimated by Autor, Dorn, and Hanson (2013). One key reason for this substantial positive impact on Chinese labor markets is the extensive interregional mobility in China. I find that the increasing export exposure gave rise to approximately 39 million internal migrants during the same period, one-third of whom worked in the manufacturing sectors. To correct the endogeneity of local export growth, I instrument changes in Chinese exports using the contemporaneous export growth of low- and middle- income South Asian countries (LMSAs). Internal migration keeps trade-sector wages from increasing dramatically and contributes to lower wage inequality across regions as well as a structural shift from agriculture to manufacturing and services.  

“News Reports when Pollution Occurs: the Role of Media Control in China” (with Zhao Chen, Jingbing Feng)

Abstract: We construct a unique data set of 377 Chinese local newspaper reports of 50 severe water pollution events between 2004 and 2013 to document pollution news reporting under government media control and the subsequent impact on self-protection behavior. We focus on the differences between party and non-party newspapers to illuminate the impact of government media control. We find that compared with non-party newspapers, party newspapers are 30 percentage points less likely to report local water pollution, but 50 percentage point more likely to report upstream pollution, due to local governments' media control.  The difference is robust when water quality is controlled. To the contrary, there is no significant difference in reporting flood events between party and non-party newspapers. It also shows that party-newspapers take longer time and use fewer words to report pollution. The differences between party and non-party newspapers can partly be explained by market forces, such as circulation and advertising revenues. We also find that in polluted areas, each water pollution report by party newspapers increases household bottled water consumption by around $0.2, while reports by non-party newspapers have smaller and less significant impact. It suggests though pollution reports by party newspaper have been limited in China households still trust and respond to their reports. 

Papers in Progress

“Closure of High-Polluted Firms and Productivity of Neighbors: Evidences from Chinese Firms and NASA Data”

Summary: Many high-polluted firms in coastal China were forced to close due to environmental regulation in recent years. Using this policy shock as a qusai-natural experiment, this research studies the impact of air pollution on labor productivity in firms that have been neighbors of those high-polluted firms by comparing differences before and after the shock. The Aerosol Optical Depth (AOD) data from NASA is employed in this study to estimate air quality. Preliminary results show that the closure of heavily polluted firms significantly increases local air quality and increases productivity in nearby firms. I found no evidence that the closure of these firms significantly affects local labor markets.

“Air Pollution and Labor Productivity in a Chinese Light Bulb Factory”(with Tao Chen and Richard Freeman)

Summary: Using daily production from workers’ worksheets and daily air quality measures from a nearby monitoring station, we study the impact of air quality on workers’ daily productivity in a light bulb factory in Changzhou, China. We apply a difference-in-differences approach, exploiting the difference between workers operating in open workshops and those in closed workshops on high-pollution days and low-pollution days. The detailed yearly physical exams of workers make it possible for us to test health as a potential mechanism.

Projects in Progress

 China’s Air Filtration Projects, Harvard Center for Green Buildings and Cities, with Richard Freeman

Summary: Air pollution has large social and economic costs. This project studies whether and how technology intervention can do in short and long run to reduce the harm of air pollution, as to examine the impact of air purification in the built environment, such as schools, factories, and residential houses, on individuals’ performance. The goal of projects is to conduct field experiments and statistical and epidemiological studies to estimate the impact of air filtration on health, education and productivity, and develop a rigorous benefit-cost analysis of the filtration solution.  The experiments are held in elementary schools in Hangzhou, China; a light bulb factory in Changzhou, China; and residential houses in Hebei, China.

Exports and Air Pollution in Chinese Cities: The Green Inputs Effect

Summary: Using a list of environmental goods put together by the Asia-Pacific Economic Cooperation (APEC), this study tests whether export production benefits air quality in China through adoption of green inputs. Using data from Chinese cities during years 2003-2009, I show that most of the positive correlation between exports and air quality in China can be explained by the usage of imports of green inputs in the production for exports. We also find that the relationship between exports and air quality was stronger during the Financial Crisis than in previous years, suggesting that when the change in exports is more likely to be driven by changes in foreign demand, as opposed to domestic supply, production for export is associated with better air quality.