Working Papers

    Health and Labor Market

• Long-term Impacts of Growth and Development Monitoring: Evidence from Routine Health Examinations in Early Childhood
  with Yinhe Liang and Xiaobo Peng
  CESifo Working Paper No. 10912
  Abstract

  This paper examines the long-term impacts of growth and development monitoring in early childhood. For this purpose, we evaluate a healthcare program, the Systematic Management of Children (SMC), which offers growth and development monitoring through routine health checkups for all young children (0-6 years) in China. Using data on the program’s county-bycounty rollout from 1950 to 2010, we find that full exposure to the SMC from birth increases lifetime income by 5%. We further provide evidence of several underlying mechanisms, including improved physical and mental health, better educational outcomes, increased cognitive skills, and sustained use of routine health checkups among adolescents.  Close

• Occupational Regulation and Reduction in Quality: Evidence from the Medical Residency Program in China
  with Vyacheslav (Slava) Mikhed and Xiangyu Li
  Abstract

  This paper examines the impact of occupational regulation on the quantity and quality of labor supply. For this purpose, we explore the implementation of the three-year medical residency training system in China in 2014 as a natural experiment. Using three unique administrative datasets provided by the Ministry of Education, we show that the implementation of the medical residency program leads to a 22\% decrease in the number of candidates opting for and passing the medical licensing exam. Meanwhile, there is a significant decline in college admission scores of medical majors compared to those of non-medical majors. These findings suggest that the medical residency program diminishes the overall labor supply of young physicians and deters high-quality contenders from pursuing a medical career, therefore reducing the overall quality of the profession.  Close

• The Impact of Local Alcohol Access on Health and Alcohol-related Crimes
  Abstract

  After the repeal of National Prohibition in 1933, 34 states passed local option laws, where county and municipal governments govern alcohol policies in terms of alcohol availability and distribution, mostly by popular votes (i.e., local option elections). Currently, 10% of U.S. communities still maintain a ban on some or all alcohol sales. Assessing the causal effects of local alcohol sales restrictions on alcohol-related outcomes, such as birth weight, is complicated by the potential non-random selection of liquor laws. This study uses detailed information of historic municipal-level local option elections from Texas for the period 1978-2004. Local option elections determining the types of alcoholic beverages which may be sold and how they can be sold by counties, cities, or individual justice of the peace precincts provide a source of variation that can be used to overcome the endogeneity and self-selection of local level alcohol restrictions. Comparing close losers with close winners (i.e., regression-discontinuity design) provides quasi-random variation in winner status, since for narrowly decided races the outcome is unlikely to be correlated with other district characteristics as long as there is some unpredictable component of the ultimate vote. Municipalities experienced substantial increases in the total number of establishments licensed to sell alcohol by the drink (e.g., bars and restaurants) after legalizing the local sale of alcohol to the general public (i.e., go from "dry" to "wet"). Also, legalizing the local sale of alcohol to the general public in Texas led to 5% more newborns with low birth weight (<=500g or 5.5lb). This indicates higher risk of maternal drinking during pregnancy when it is easier to purchase liquor locally.  Close

• Mistrust and Long-term Impacts of Quality Shocks
  with Haicheng Wang, Rui Xie, and Fushun Zhang
  Abstract

  Taking China’s dairy scandal in 2008 as a quasi-natural experiment, this paper shows that the domestic quality scandal pushed consumers to persistently substitute dairy products produced locally by oversea brands. To identify the scandal’s effects on the dairy imports, we use an interacted difference-in-differences model, comparing imports of non-dairy and dairy products, before and after the revelation of the scandal, in varying proximity to dairy firms with contaminated products. We find that the import of dairy products increased by over 60% after the disclosure of the scandal and continued to grow in following years with escalated mistrust of local dairy brands. As imported dairy products are at least 40 percent more expensive than domestic brands, consumers bear welfare losses of 20 to 30 percent of baseline spending on dairy products. These estimates suggest that quality shocks, especially the ones in developing countries, could have a long-lasting effect on consumer product choices and development of local industries due to mistrust to local firms and government agencies.  Close

    Public Transport

• Do Transit Fares Hurt the Poor? Evidence from Repeated Weekly Pass Commuters in New York City
  with Vyacheslav (Slava) Mikhed
  Abstract

  Exploiting details of transit card data during the period 2013-2015, we provide evidence that a large number of commuters on the New York City public transit system purchase a weekly (7-day) unlimited-ride transit pass every week and continue this purchase pattern for more than 11 months. As a monthly (30-day) pass costs roughly the same as three and a half weekly passes, these commuters waste a lot of money on transit fares than they would have with monthly passes while receiving the same level and quality of transit services. These repeated weekly pass commuters predominately live in low-income neighborhoods, which points to the importance of liquidity constraint. Low-income commuters may be unable to spend a large amount at one time for a monthly fare, opting instead to buy weekly fares that are cheaper in the short-term but much costlier in the long-term. Alternative mechanisms such as time inconsistency and job instability alone are unlikely to explain the observed behavior of recurring seven-day pass commuters. These commuters would benefit significantly from a monthly fare cap.  Close

• Money for MetroCards: How a New Card Fee Made New York Transit Riders Invest More and Lose More
  Abstract

  Since 1998, the New York City Metropolitan Transportation Authority (MTA) system has used prepaid cards (MetroCards) to collect subway and bus fares. In 2013, the MTA imposed a $1 card fee (surcharge) on new MetroCard purchases. After the card fee was imposed, riders started to put more money on new MetroCard purchases. The response to the card fee was greater in low-income neighborhoods and among riders who used cash or debit cards rather than credit cards. As a result, the net monthly outstanding balance from transit card deposits increased dramatically, with riders lending an extra $150 million, on an annual basis, to the MTA. Moreover, over $20 million of the increased balances in the first year were never redeemed and escheated to the MTA when these cards expired. I document these changes using a novel dataset with transaction-level deposit and card use information. The leading explanation highlights the importance of the cost of effort to remember to carry the same card for future periods. I pose a dynamic structural model to calibrate the effect of a new card fee. Counterfactual simulation predicts that the optimal new card fee is $4.35 to maximize the MTA's profit. These findings have implications for fiscal policy designs and fee structures of prepaid card industry. Close

• Will Consumers Voluntarily Undertake More Proenvironmental Actions When It is Easier? Evidence from Metrocard Usages in New York City
  Abstract

  Using a large subway transaction dataset with deposits and rides information in the New York City Metropolitan Transportation Authority (MTA) system, this paper examines the efficacy of various policy mechanisms to increase reuse of existing metrocards. Unlike findings in prior studies using self-reported survey data, commuters did not voluntarily reuse existing free transit cards more when cross-type refill options became available. In contrast, the surcharge fee on new card purchases reduced the overall demand for new metrocards by over 52 percent and prompted commuters to substitute to refill existing metrocards. These results suggest that accompanying better facilities with economic incentives can be much better effective in promoting proenvironmental actions than the implementation of better facilities alone.   Close

Publications

• "The Potential Causal Effect of Hukou on Health Among Rural-to-Urban Migrants in China" Journal of Family and Economic Issues, September 2021 Abstract

  A number of economic studies have shown a strong positive correlation between urban household registration status (hukou) and better health outcomes in China. The question at the center is whether the correlation implies causation. Change in the hukou system, in 1964, is used to test the causality between hukou and health. The regression-discontinuity (RD) design estimates suggest that urban hukou citizens have a much better chance of being in good health. The deleterious effects of rural hukou on health possibly work through mechanisms of income disparity, variations in educational attainment, and availability of health insurance.   Close

• "Dissecting China's Property Market Data" World Economics, March 2016 Abstract

  This paper analyses Chinese property market data to evaluate recent trends in the market and to make prognoses for the future. It considers whether or not the existence of high prices and at the same time an enormous rise in residential supply in terms of floor space under construction means that there is a ``bubble'' in China's property market which may burst, similar to what happened in Japan in the early 1990s. Evidence that the price of new homes moves almost perfectly with sales of new residential floor space rather than with completed floor space suggests that the housing market is behaving normally and follows mini boom and bust cycles like other industries. The analysis finds that there are low maintenance costs for buyers after purchase due to the lack of annual property tax and negligible depreciation of bare-shelled housing units which limits the risk of default. Although recently developers are under pressure to raise more revenue mainly due to high interest-rate borrowing from shadow banks, the author considers that the probability of a systemic collapse of housing market is minimal given existing taxation systems, easing monetary policy and the continuing urbanization process.  Close