Evaluating the Feasibility of an Investment for Improving Drinking Water Quality in South Korea

This dissertation aims to test the feasibility of improving the quality of drinking water in a target area in South Korea. The problem with drinking water quality is caused by pollution of the water environment. Most waterworks in South Korea are unable to handle problems like an unpleasant taste or odour in tap water. Improving raw water quality through prevention of water pollution is explored in pursuit of a long-term solution. However, this research focuses on advanced water treatment systems for a short-term solution and mainly tests the extent to which an investment in one chosen waterworks is feasible. Cost-benefit analysis (CBA) is used to test the feasibility of two advanced water treatment systems: granular activated carbon (GAC), and ozone plus GAC treatment. Three main steps are involved: measurement of the social benefits, cost estimation of the two alternatives and the CBA.

Choice experiments are chosen for measuring the benefits with three alternatives: the status quo, GAC, and ozone plus GAC. Four key attributes are selected: safety, taste and odour, colour and cost. The experimental design consists of 32 choice cards. Three types of treatment for hypothetical bias are used: budget constraint reminder, cheap talk and an honest priming task. In July/August 2015, 573 people participate in the survey; ineffective data, potential label heuristics, and outliers are filtered. Thus, 406 data items are examined for representativeness of the sample regarding socioeconomic factors and used in the analysis.

The marginal willingness to pays (MWTPs) is estimated using three types of logit models (multinomial logit, random parameter logit and latent class logit). To detect the effectiveness of the treatments for mitigating hypothetical bias, dummies and interaction terms are included in the models and the coefficients of the variables are examined. After measuring the social benefits using MWTP, the cost of installing the two alternatives is estimated. While the mean MWTP is the correct measure to use from the standpoint of economic efficiency, the median WTP is probably the more appropriate measure to assist a democratic decision-making

ii

process and can be considered a more cautious value to avoid hypothetical bias. Thus, the median values of WTP are also used for the CBA.

The economic feasibility is tested by comparing the costs and benefits of the two alternatives. Both net present values (KRW1 15.8 billion for GAC and 13.1 for ozone plus GAC) are larger than zero. Internal rates of return of the two alternatives are 8.97% for GAC and 7.46% for ozone plus GAC. The benefit to cost ratio of GAC, 1.389, is greater than of ozone plus GAC, 1.225. Note that the GAC seems to be a more robust option than ozone plus GAC in terms of the decision rules of three discount cash flow methods.

Concerning risk and uncertainty, sensitivity analyses are performed using several scenarios with the following factors: increase of discount rate, costs and construction period, and decrease of business life, benefits and beneficiaries. The worst case scenarios would likely be when the social benefits decrease to zero within the business life. In such a case, the three feasibility values cannot sustain the validity of the two alternatives.

In conclusion, the results of this research suggest that investment in the two advanced water treatment systems is feasible, but it depends on situations that may change in practice, such as reduction in the business life. The research also shows that prevention of water pollution can and should be a complementary approach for supplying safer and cleaner drinking tap water. Protecting the water catchment area along with the installation of the two advanced water treatment options should be considered for a more comprehensive and sustainable solution in the long run.