Minsung Park

Date of Award

Spring 5-15-2023

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



This dissertation comprises three chapters that attempt to examine the effects of new infrastructure of transportation systems.

This first chapter "The Impact of High Speed Rail on Traffic Congestion" examines the impact of high-speed rail systems on highway congestion, using the case of South Korea. I use different strategies to address the endogeneity problem of evaluating the impact of transport infrastructure on traffic congestion. The results indicate that building high-speed rail lowers demand for intercity driving by 7.5\% to 30.9\%. However, the detailed data suggests that the reduction in congestion comes primarily from night time, during which high-speed trains do not run, and building high-speed rail does not have significant impacts on traffic demand during rush hour and daytime. These findings suggest that congestion at peak time may not be relieved. This study contributes to the literature on high-speed rail and congestion and is the first to estimate the effects of building high-speed rail systems on intercity traffic congestion.

The second chapter "Welfare Analysis on Travel Time Saving: from Intercity Driving to High Speed Train" attempts to answer whether building high-speed rail can recoup its tremendous costs of construction. The debate is rekindled in the United States because of the construction of the high-speed rail system linking San Francisco and Los Angeles. Transportation researchers haven't yet arrived at a conclusive cost-benefit analysis of the transport infrastructure. This stems from the difficulty of estimating first-order benefits such as travel time saving due to lack of data. To address this issue, this study employs novel intercity driving data and the construction of high-speed rail in South Korea as quasi-natural experiment to estimate the first-order benefits. Specifically, I exploits the difference in travel time between driving and high-speed trains for 146 intercity routes. Employing nested logit demand models, I find that the value of travel time is approximately 14 cents per minute. Also, the internal rate of return of adding a new rail line has a positive sign, even when revenue of the rail companies and rent increase for the land are not taken into account. These findings suggest that high-speed rail can be a viable option for improving transportation infrastructure.

The third chapter "Displacement and Substitution from Physical Barriers for Suicide Attempts" investigates substitution effects of installing physical barriers for suicide prevention. Physical barriers are considered as an effective way to reduce suicides at a given metro station. However, it remains unexplored whether such installation could eradicate the incentives to end life for those who are vulnerable to suicidal impulses. Using novel data on Metro in Seoul and suicide attempts at the municipality level, I find pioneering evidence showing that the physical barriers installation at metro stations can cause displacement to other stations without barriers or substitution into seeking another method of suicide. The installation substantially lowers the number of suicide attempts at the metro station but I reject the null hypothesis that all of the reduced suicides are prevented at the municipality level. Furthermore, taking types of physical barriers into account, I find an interesting pattern of substitution of each type based on economic theory. Search costs for alternative ways of suicide explains the substitution pattern of each type of physical barriers.


English (en)

Chair and Committee

Ian Fillmore

Committee Members

Gaurab Aryal, Marcus Berliant, Cecilia Diaz-Campo, Yongseok Shin,