Chinese Economy

October 9-10, 2015
Hanming Fang of the University of Pennsylvania and Shang-Jin Wei of Columbia University, Organizers

Cynthia Kinnan, Northwestern University and NBER; Shing-Yi Wang, University of Pennsylvania and NBER; and Yongxiang Wang, University of Southern California

Relaxing Migration Constraints for Rural Households (NBER Working Paper No. 21314)

This paper exploits two unique features of China's history to study the effects of access to internal migration: reforms to the household registration (hukou) system, and historical migration flows. Kinnan, Wang, and Wang show that temporary migration due to a government policy called the "sent-down youth" (SDY) program created lasting inter-province links, so that decades later, hukou reforms in cities which sent SDY increased migration in provinces where those SDY temporarily resided. Using this variation, the reseachers find that improved access to migration leads to higher consumption levels and lower consumption volatility for rural households. Furthermore, household production shifts into high-risk, high-return activities.


Koichiro Ito, University of Chicago and NBER, and Shuang Zhang, University of Colorado

Willingness to Pay for Clean Air: Evidence from Air Purifier Markets in China

This paper provides among the first revealed preference estimate of willingness to pay (WTP) for clean air in developing countries. Ito and Zhang's first approach exploits panel variation in air pollution in Chinese cities along with product-by-store level transaction data in air purifier markets. The researchers first estimate the nationwide average of marginal willingness-to-pay (MWTP) for removing 1 ug/m3 PM10 for a year, and WTP for removing the average level of PM10 (100 ug/m3 of PM10). Their second approach leverages the Huai River heating policy, which created discontinuous quasi-experimental variation in air pollution between the north and south of the river. Using a spatial regression discontinuity design, the researchers estimate the local average of MWTP for removing 1 ug/m3 PM10 that is generated by the Huai River policy. Combining their estimates on MWTP for clean air with estimates on the pollution-health relationship, the authors find that the lower bound of health valuation in China is substantially higher than previously understood for developing countries. Their findings provide important policy implications for optimal environmental regulation.


Wolfgang Keller and Carol Shiue, University of Colorado and NBER, and Xin Wang, University of Colorado

Capital Markets in China and Britain, 18th and 19th Century: Evidence from Grain Prices (NBER Working Paper No. 21349)

Capital markets allow surplus income to be invested into productivity-enhancing projects. Despite their prominence in general accounts of growth little is known on their role in the emergence of modern economic growth. In this paper Keller, Shiue, and Wang ask whether capital market performance might explain why Britain surged ahead of China in the 18th century. The researchers employ an asset-pricing model together with information on regional grain prices to derive interest rates, and then compare capital market development in large parts of Britain and China. The researchers first calibrate the method and show that it can replicate key features of the United States' early 19th century capital market, where more systematic data from bank interest rates is available. Using this approach they estimate interest rates for Britain that are at least 20% lower than those for China, for the years 1770 - 1860. Moreover, the regional integration of British capital markets, measured in terms of bilateral interest rate correlations, was far greater than it was in China. The Yangzi Delta correlations come close to the British average at distances below 200 kilometers, but at larger distances interest rate correlations in Britain are twice those of the Delta, and three or more times as high as elsewhere in China. The authors also find that Britain's advantage over China in terms of market integration existed already in the late 18th century. Backcasting on the 19th century trends suggests capital market divergence started by the year 1690. Overall, the results provide support for the hypothesis that divergence in capital market development occurred before income divergence, and may therefore be an important factor in explaining the Great Divergence.


Chen Lin and Xiaofeng Zhao, Chinese University of Hong Kong; Randall Morck, University of Alberta and NBER; and Bernard Yeung, National University of Singapore

Anti-Corruption Reforms and Shareholder Valuations: Evidence from China

Chinese share prices rose sharply on the Politburo's December 4th, 2012 announcement of its new Eight-point Regulation, outlining new party policy against corruption. The announcement came surprisingly soon after a change in leadership and was surprisingly detailed and concrete. The positive reaction is significantly larger in provinces with more advanced market reforms. The reaction is uniformly positive for state-owned enterprises, but heterogeneous across non-SOEs. Among the latter, the reaction is more positive in regions with more complete market reforms and for firms with lower reported entertainment and travel costs (ETC), higher prior productivity, greater external financing needs, and greater growth potential. Negative price reactions are evident for non-SOE firms with substantial entertainment and travel costs located in regions with less complete market reforms. Lin, Morck, Yeung, and Zhao posit that limiting corruption cuts the valuations of such non-SOEs by limiting their ability to "grease" bureaucratic gears. SOEs are well-connected in any case, and their ETC may be pure perks consumption or signs of self-dealing by the officials running them at the time. Reforms that limit this boost the SOEs' valuations and presumably increase state revenues from any subsequent sales of their shares. Overall, these results are consistent with investors believing the reforms to be meaningful, and to work to more the advantage of more productive firms with less past investment in connections located in more market-friendly regions.


Bei Qin, University of Hong Kong; David Stromberg, Stockholm University; and Yanhui Wu, University of Southern California

The Political Economy of Social Media in China

In this paper, Qin, Stromberg, and Wu study the role of Chinese social media in three areas: collective action, monitoring of politicians, and the gauging of public opinion. The researchers study is based on a data set of 13.2 billion blog posts from Sina Weibo — the most prominent Chinese microblogging platform — over the period of 2009-2013. In contrast to the previous studies that conclude that collective action events are censored in Chinese social media, the authors find millions of posts discussing protests, strikes, and demonstration. Moreover, they find that microblog posts 1) are highly informative in predicting collective action events and corruption charges; and 2) have a significant and positive effect on the incidence of strikes and protests, although their effects on the incidence of large-scale massive conflicts and government-sanctioned demonstrations (e.g., anti-Japan) are muted. Finally, they find that the number of government microblog accounts, based on machine learning estimation, is larger in areas with a higher level of internet censoring and where newspapers have a stronger pro-government bias. Overall, their findings suggest that the Chinese government regulate social media to balance threats against regime stability against the benefits of bottom-up information.

Yi Che, Shanghai Jiao Tong University; Anson Ly, National University of Singapore; Justin Pierce, Federal Reserve Board; Peter Schott, Yale University and NBER; and Zhigang Tao, University of Hong Kong

Do Chinese Imports Influence American Politics?

This paper examines the impact of U.S. imports from China on U.S. Congressional elections. Che, Lu, Pierce, Schott, and Tao find that counties with greater increases in imports from China during the 2000s exhibit increases in the share of votes cast for Democrats as well as the likelihood that a Democrat takes a seat from a Republican. These relationships are not present in the 1990s, before the surge in U.S. imports from China occurred, and they are consistent with tjhe researchers' finding that Democrats in office were more likely than Republicans to support legislation limiting import competition or offering economic assistance.


Andrew Ang, Columbia University and NBER; Jennie Bai, Georgetown University; and Hao Zhou, Tsinghua University

The Great Wall of Debt: Corruption, Real Estate, and Chinese Local Government Credit Spreads

Ang, Bai, and Zhou find that corruption in China has a significant effect on local government credit spreads, with one standard deviation increase in "Tigers" or "Flies" corresponding to 9 or 5 basis points in elevated bond yields. The researchers measure corruption using the graft cases conducted by the Central Commission of Discipline and Investigation (CCDI), with the rank-weighted average and total number of cases as proxies for the top level and low level corruptions — "Tigers" and "Flies", respectively. Real estate GDP is the most prominent determinant of local government credit spreads, and the effect of corruption works mainly through the channel of real estate — more corruption clearly depresses the real estate value. Since the anti-corruption campaign started in late 2012, the effect of corruption has become highly significant and non-redundant through the real estate channel.


James Choi, Yale University and NBER; Li Jin, Peking University; and Hongjun Yan, Yale University

Informed Trading and the Cost of Capital

Does information asymmetry among traders of a firm's securities increase its cost of capital? Choi, Jin, and Yan explore this question using representative portfolio holdings data from the Shanghai Stock Exchange. The researchers show that institutional investors have a strong information advantage, and that past aggressiveness of institutional trading in a stock positively predicts institutions' future information advantage in this stock. Sorting stocks on this predictor and controlling for other correlates of expected returns, they find that the top quintile's average annualized return in the next month is 10.8% higher than the bottom quintile's, indicating that information asymmetry raises the cost of capital.


Raul Santaeulàlia-Llopis, Washington University, St. Louis, and Yu Zheng, City University of Hong Kong

The Price of Growth: Consumption Insurance in China 1989-2009

Growth entails taking risks. This implies that the welfare gains of growth hinge on the ability of households to insure consumption against the risks associated with growth. Here, Santaeulàlia-Llopis and Zheng exploit a novel and unique opportunity to empirically study this question using as a laboratory an economy, China, that has witnessed enormous and sustained economic growth and for which the authors build a long panel of household-level consumption and income for the most recent two decades. They find that consumption insurance is disrupted along the growth process with a transmission of permanent income shocks to consumption that triples from 1989 to 2009. Their evidence suggests that the shortage of available options to store wealth limits the use of household savings for precautionary reasons. Their results have implications for the welfare assessment of growth across time and space. Across time, the loss of insurance implies that rural households would actually prefer to live in the growth-risk-insurance environment of the pre- rather than post-WTO years, despite higher rural growth in the 2000s. Across space, while the welfare gains of rural-to-urban migration drop by more than two-thirds due to consumption insurance losses in pre-WTO years, these gains remain high in the post-WTO years after a change in the composition of public transfers that substantially improves insurance in urban areas.


Tasso Adamopoulos, York University; Loren Brandt and Diego Restuccia, University of Toronto; and Jessica Leight, Williams College

Misallocation, Selection and Productivity: A Quantitative Analysis with Panel Data from China

Adamopoulos, Brandt, Leight, and Restuccia use micro panel data from China and a quantitative framework to document the extent of resource misallocation in agriculture within villages, across villages, and over time; and to assess the total factor productivity (TFP) gains from an efficient reallocation of resources. The researchers then develop and estimate a tractable heterogeneous-ability two-sector framework of agriculture and non-agriculture to assess the TFP implications of distortions with ability selection across sectors. An efficient reallocation of capital and land to existing farmers in China would increase output and productivity by 84 percent. Resource misallocation in agriculture implies substantial distortions in occupational choices. Eliminating misallocation across farms in agriculture generates substantial reallocation across sectors, increasing average ability in both sectors but especially in agriculture, and generating a large impact in aggregate outcomes.