DescriptionThis dissertation aims to improve the understanding of environmental policy mechanisms and provide estimates of the costs of toxic pollution (in the air and water) and ambient particulate matter air pollution in Korea. The results will be useful for future evaluations and formulations of Korean environmental policies and regulations.
In Chapter 2, businesses' incentives to participate in voluntary environmental programs are explored through the 30/50 program conducted from 2004 to 2013 in Korea. The industrial facilities that joined the 30/50 program promised to reduce their toxic emissions by 30% within 3 years and 50% within 5 years. Unlike mandatory regulations, the voluntary 30/50 program did not include sanctions for noncompliance. However, most of participating facilities achieved the emission reduction goals (although any implication regarding the causal impact of the program cannot be made without formal study), and the linear probability model estimation results indicate that public recognition and strong regulatory background threats (or the prospect of relaxed regulatory oversight) are important predictors of businesses' participation in the program. A notable finding is that facilities owned by chaebols, i.e., family-owned conglomerates, showed a considerably higher probability of participation. Chaebols with great economic and political power in Korea might have wished to participate in order to take advantage of any opportunities to influence future environmental regulations and relax the greater regulatory pressures usually placed upon them. Chaebols' higher participation rates reveal the importance of studying VEPs within a country's specific political, economic, and social structures.
In Chapters 3 and 4, estimates of the costs of toxic air and water pollution as well as ambient particulate matter air pollution are provided. Chapter 3 estimates the effects of pregnant mothers' exposures to toxic emissions on newborns' health. Using 2004-2007 birth data merged with data on toxic emissions in the mothers' counties of residence, linear probability models for the probabilities of babies being born with low birth weight (<2.5kg) and very low birth weight (<1.5kg) are estimated. The estimation results indicate that toxic and carcinogenic water emissions increase the probability of low birth weight and toxic water emissions increase the probability of low birth weight more than non-toxic water emissions do. It is also found that heavy metal air emissions have the most harmful effects on the probability of low birth weight and are the only type (and medium) of chemical emissions found to increase the probability of very low birth weight. Chapter 4 examines Korean households' willingness to pay to avoid ambient PM10 (particulate matter 10 micrometers or less in diameter) air pollution using a hedonic approach. Here, the transaction records for all apartments sold during 2006-2016 in Korea are used. The OLS estimation, even when including apartment-complex fixed effects, yields statistically significant and positive point estimates of the effects of local ambient PM10 concentrations on apartment prices. This suggests that there must be bias coming from an omitted time-varying factor(s). An IV approach using Asian dust events as an instrumental variable for the annual average PM10 concentrations was also utilized. Although a naturally occurring Asian dust storm is likely to satisfy the exclusion restriction, it forms a weak instrument, and the weak-instrument robust Anderson-Rubin test cannot identify the parameter estimate of the effect of local ambient PM10 concentration on apartment prices. No conclusive evidence regarding Korean households' willingness to pay to avoid ambient PM10 air pollution can be found using these approaches.