We study how adverse economic shocks influence political outcomes in strong authoritarian regimes, by examining the export slowdown in China between 2013-2015. We exploit detailed customs data and the variation they reveal about Chinese prefectures’ underlying exposure to the global trade slowdown, in order to implement a shift-share instrumental variables strategy. Prefectures that experienced a more severe export slow-down witnessed a significant increase in incidents of labor strikes. This was accompanied by a heightened emphasis in such prefectures on upholding domestic stability, as evidenced from: (i) textual analysis measures constructed from official prefecture annual work re-ports; and (ii) data gathered on local fiscal expenditures channelled towards public security uses and social spending. The central government was subsequently more likely to replace the party secretary in prefectures that saw a high level of “excess strikes”, above what could be predicted from the observed export slowdown, suggesting that local leaders were held to account on yardsticks related to political stability.
This paper was distributed as Working Paper 25925, where an updated version may be available.
Chen, Chen, He, Liu, and Xie provide causal evidence for the value of asset pledgeability. Their empirical strategy is based on a unique feature of the Chinese corporate bond markets, where bonds with identical fundamentals are simultaneously traded on two segmented markets that feature different rules for repo transactions. The researchers utilize a policy shock on December 8, 2014, which rendered a class of AA+ and AA bonds ineligible for repo on one of the two markets. By comparing how bond prices changed across markets and rating classes around this event, Chen, Chen, He, Liu, and Xie estimate that an increase in the haircut from 0 to 100% would result in an increase in bond yields in the range of 40 to 83 bps. These estimates help the researchers infer the magnitude of the shadow cost of capital in China.
This paper was distributed as Working Paper 26520, where an updated version may be available.
China's housing prices have been growing rapidly over the past few decades, despite low growth in rents. Miao, Jiang, and Zhang study the impact of housing bubbles on China's economy, based on the understanding that local governments use land-sale revenue to fuel infrastructure investment. They calibrate their model to the Chinese data over the period 2003-2013 and find that their calibrated model can match the declining capital return and GDP growth, the average housing price growth, and the rising infrastructure to GDP ratio in the data. The researchers conduct two counterfactual experiments to estimate the impact of a bubble collapse and a property tax.
Are quid pro quo (technology for market access) policies effective in facilitating knowledge spillover to developing countries? Bai, Barwick, Cao, and Li study this question in the context of the Chinese automobile industry where foreign firms are required to set up joint ventures with domestic firms in return for market access. Using a unique dataset of detailed quality measures along multiple dimensions of vehicle performance, they document empirical patterns consistent with knowledge spillovers through both ownership affiliation and geographical proximity: joint ventures and Chinese domestic firms with ownership or location linkage tend to specialize in similar quality dimensions. The identification primarily relies on within-product variation across quality dimensions and the results are robust to a variety of specifications. The pattern is not driven by endogenous joint-venture network formation, overlapping customer base, or learning by doing considerations. Leveraging additional micro datasets on part suppliers and worker flow, the researchers document that supplier network and labor mobility are important channels in mediating knowledge spillovers. However, these channels are not tied to ownership affiliations. Finally, Bai, Barwick, Cao, and Li calibrate a simple learning model and conduct policy counterfactuals to examine the role of quid pro quo. Their findings show that ownership affiliation facilitates learning but quality improvement is primarily driven by the other mechanisms.
New Keynesian models often assume firms bear real resource costs to change product prices to maximize value for the firm as a whole. If, however, firms' large shareholders bear disproportionally high costs but enjoy few benefits, they will veto proposals in support of any price changes and, as a result, prices will be sticky. Xie, Shi, and Xu test this hypothesis using matched administrative customs and firm data that allows us to compare pricing strategies of the same product to the same destination. Their difference-in-differences framework exploits a mandatory corporate governance reform in China that differentially stimulates large shareholders' incentive to care about firm value. Following the reform, firms actively adjust producer-currency prices to stabilize buyer prices in response to real-exchange-rate fluctuations, and the effects are stronger when large shareholders' incentives are more stimulated. The results are mainly driven by product-destination markets in which flexible pricing strategies are costlier to implement, by firms that are more in need of external finance, and by SOEs that are more affected by the reform. The researchers conclude that the separation of ownership and control constitutes an important source of price stickiness.
Direct experience of a peer's punishment might make non-punished peers reassess the probability and consequences of facing punishment and hence induce a change in their behavior. D’Acunto, Weber, and Xie test this mechanism in a setting, China, in which they observe the reactions to the same peer's punishment by listed firms with different incentives to react - state-owned enterprises (SOEs) and non-SOEs. After observing peers punished for wrongdoing in loan guarantees to related parties, SOEs - which are less disciplined by traditional governance mechanisms than non-SOEs - cut their loan guarantees. SOEs whose CEOs have stronger career concerns react more than other SOEs to the same punishment events, a result that systematic differences between SOEs and non-SOEs cannot drive. SOEs react more to events with higher press coverage even if information about all events is publicly available. After peers' punishments, SOEs also increase their board independence, reduce inefficient investment, increase total factor productivity, and experience positive cumulative abnormal returns. The bank debt and investment of related parties that benefited from tunneling drop after listed peers' punishments. Strategic punishments could be a cost-effective governance mechanism when other forms of governance are ineffective.
State-owned enterprises (SOEs) continue to account for a large part of the Chinese economy, even though they are substantially less productive than their private counterparts. Wen proposes that one reason for SOEs' persistence is their role in promoting social stability through job provision. She documents that SOEs prevent widespread unemployment by hiring during natural disasters and periods of poor export demand. Motivated by these patterns, the researcher builds a model of unrest behavior and employment which predicts that SOE employment should increase in response to destabilizing shocks. Wen tests model predictions with a novel natural experiment using variation in an ethnic conflict in China's Xinjiang province between the Uyghur minority and the government. She uses a triple difference strategy for identification: unrest threat is higher following years with many conflict incidents in Xinjiang, in counties outside Xinjiang with a high share of Uyghur residents, and among minority men, the demographic most likely to participate in ethnic unrest. Using a combination of representative household data and original conflict data, which he constructs from archival sources, the researcher shows that SOE employment of male minorities differentially increases in response to ethnic unrest threat. Male minority wages also rise, which cannot be explained by a labor supply shift or increased prejudice in private hiring alone. Additionally, relief transfers to male minorities increase, but mostly to the non-employed, indicating that SOE employment and relief transfers are complementary stability policies. Quantifying SOE favoritism of male minority employees through the model implies that SOEs implicitly receive a 26% subsidy on male minority wages.
Greenstone, He, Jia, and Liu examine the introduction of automatic air pollution monitoring, which is a central feature of China's "war on pollution." Exploiting 654 regression discontinuity designs based on city-level variation in the day that monitoring was automated, they find that reported PM10 concentrations increased by 35% immediately post-automation and were sustained. City-level variation in underreporting is negatively correlated with income per capita and positively correlated with true pre-automation PM10 concentrations. Further, automation's introduction increased online searches for face masks and air filters, suggesting that the biased and imperfect pre-automation information imposed welfare costs by leading to suboptimal purchases of protective goods.
This paper studies the productivity consequences of favoritism in employee promotions within organizations. Using data from public high schools in four Chinese cities, Li first shows that teachers with hometown or college ties to the school principal are twice as likely to be promoted, after controlling for characteristics on their application profiles and their value-added in teaching. She then uses the results from a survey in which she asked teachers to select anonymous peers to promote from a pool of applicants applying for promotion to infer each teacher's revealed fairness views regarding promotion qualifications. Contrasting these with actual past promotions in turn allows me to measure if and when a teacher might have observed unfair promotions in her own school in the past. Exposure to unfair promotions adversely affects nonapplicant teachers' output, lowering their value-added and raising the probability that high-value-added teachers quit. The value-added effect appears to be driven primarily by teachers' social preferences for peer workers and the consequent erosion of their morale when peers suffer unfair treatment, while the quitting effect comes mainly from non-favored prospective applicants' career concerns as they learn about the principal's bias and leave due to poor promotion prospects. These adverse spillover incentive effects lead to a substantial reduction in school-wide performance. Finally, a transparency reform that required principals to disclose to their peers the profiles of promotion applicant teachers reduced the principals' bias and improved the overall productivity of schools.
Ho, Rysman, and Wang evaluate Chinese restrictions on the number of foreign movies distributed domestically, particularly an increase in the quota in 2012. They estimate a structural model of consumer demand for movies. The researchers solve a discrete choice model of consumer behavior that is dynamic in the sense that consumers may see movies only once. Ho, Rysman, and Wang find that the reliance on reduced form age profiles is greatly reduced in their dynamic model relative to standard static approaches. Counterfactual experiments show that consumer welfare increases by 5.3% due to the import liberalization, and that there is relatively little substitution between foreign and domestic movies.