Research
Dissertation Chapters
"Consolidation in Medicare Part D Market and its Effects on Patients’ Utilization" (Job Market Paper)
Abstract: The Medicare prescription drug program (Part D) launched in 2006. Since then, there has been a series of consolidation, driven by mergers and acquisitions (M&A) and regulatory changes, concentrates the Medicare Part D market. We examine how the industry consolidation affect Part D utilization by merging the individual-level data from HRS with the information of market structure at CMS market level. In particular, we examine how industry consolidation affects Part D utilization through reducing plan variety and competition. We find that a lower plan variety decreases their Part D participation and that less competition increases their out-of-pocket (OOP) prescription drug costs. Our main results are robust to taking into account the effect of unobserved individual heterogeneities and region-specific time trends, and to the use of alternative outcome variables. To strengthen identification of causal effects, we exploit a quasi-experimental design to compare markets that affected by duplicated Part D plans and a nation-wide M&A with those unaffected, and find consistent results. Further, in response to industry consolidation, well-off and healthy individuals are more likely to adjust their Part D participation, while younger and long-term Part D participants pay more OOP. Finally, we find that a lower plan variety decreases their Part D participation is related to a quality deterioration and reduced plans with full subsidy among remaining plans available. The rise in OOP driven by a higher concentration is sometime partly offset by the plan composition offered in the market.
"Sleep Behavior, Mental Health, and Productivity"
Abstract: Sleep is an important input in labor productivity as well as residents’ mental health. I use data from the American Time Use Survey (ATUS) and the Behavioral Risk Factor Surveillance System (BRFSS) to examine the causal effects of sleep on labor productivity and mental health. The potential endogeneity of deciding how many hours to sleep is addressed by using an instrumental variable estimation technique. The results suggest that a one-hour increase in the weekly number of hours of sleep raises the wage rate by up to 1.8% and brings down the odds of developing depression symptoms by 1.7% on average.
Research Papers in Progress:
“Daylight Saving Time Transition and Bicycle Accidents”