Local Government Response to Fiscal Squeeze: Evidence from a Targeted Tax Reduction (Job Market Paper)

(with Jiayi Xu)

A rich literature examining the effects of intergovernmental grants to local governments has documented evidence of the “flypaper effect” in terms of overall and categorical expenditures. This paper considers this phenomena in the context of a budget shortfall generated by a targeted tax reduction that benefits a subset of the population. Specifically, we examine whether local government increases tax revenue from other sources to offset the shortfall, reduces expenditures that benefit the targeted group, and consider the net impact of these local responses on income and economic productivity. Identification comes from a ban on all agricultural taxes in China combined with differential revenue replacement levels determined by a national formula at the province level. Comparing nearly identical counties in adjacent provinces reveals that large differences in revenue shortfall are not offset by increased taxes on other subgroups, consistent with a strong flypaper effect. However, local government expenditure on agriculture is disproportionately reduced, attenuating the benefits to the targeted group. Further analysis reveals that farmers in counties that experienced larger revenue shortfalls suffered a loss of net income. These results shed light on how local governments respond to fiscal shortfall stemming from a targeted tax reduction, and how these responses may offset the intended benefits.

The Effect of Regulating Political Connections: Evidence from China’s Board of Directors Ban

(under review)

There is a great deal of variation in how countries regulate relationships between politicians and private sector firms, but little evidence about how such policies affect firm performance. In 2013, China passed a new regulation that banned politicians from serving on the boards of directors of companies. Using a novel data set that links board members, government officials, and forced resignations, I estimate the effect of the policy on the composition of corporate boards and subsequent changes in firm performance and stock price. The estimates reveal that firms complied with the policy and that the loss of a high level politician resulted in lower profits and a 9 percent reduction in stock price in the year after the policy was announced. Heterogeneity analysis reveals that the negative effect is increasing in the number of high-level politicians on a firm’s board, but that the loss of lower-level politicians generates little or no observable change in performance. The estimates provide important evidence about the efficacy and implications of one of the most commonly used policy tools for reducing political influence in the private sector.

Market Sensitivity to Product Quality, Personal Information, and Business Practice Scandals

(with Jiayi Xu)

The financial impacts of firm scandals have been widely studied in the literature. This paper exploits a unique setting that generates a large set of systematically reported scandals, allowing us to measure investors’ responses across scandal types. The “315 Night” is a Chinese TV program hold on March 15th annually and has exposed more than 200 firm scandals since 1991. We document the recent 10-year scandals revealed in the program and use an event study approach to estimate the effects across product quality, personal information breach, and business practice scandals as well as across sectors. We find that the effects are largest for defective products, but there are also strong negative responses to personal information breaches. We find little stock response to the revelation of deceptive business practices. Across sectors, consumer goods suffers the largest impacts, followed by the services sector, while the technology sector exhibits the smallest responses. The results shed light on how the market responds to various types of scandals.

How Fundamentalism Takes Root: A Simulation Study

(with Daniel Friedman, Jonathan Gair, Sriya Iyer, Bartosz Redlicki and Chander Velu. Resubmission requested by European Economic Review)

We report agent-based simulations of religiosity dynamics in a spatially dispersed population. Agents’ religiosity responds to neighbors via pairwise interactions as well as via club goods effects. A simulation run is deemed fundamentalist if the final distribution contains a sizable minority of very high religiosity together with a majority of lesser religiosity. Such simulations are more prevalent when parameter values shift from values reflecting traditional societies towards values reflecting the modern world. The simulations suggest that the rise of fundamentalism in the modern world is boosted by greater real income, lower relative prices for secular goods, less substitutability between religious and secular goods, and less time spent with neighbors. Surprisingly, the simulations suggest little role for the rise of long distance communication and transportation.

Education, Income and Intergenerational Persistence: An Evolutionary View

Research Statement
Please view my Research Statement here.