연구제목 : 미세먼지 저감 정책에 따른 조기사망 확율 감소의 간접비용 (통계적생명가치:VSL) 변화분석

일정: 2018년 12월 21일 금요일 오전10시-오후12시

장소: 국회예산정책처 경제분석팀 세미나실

연구 목적

미세먼지가 인간에게 미치는 영향은 노출과 질병 발병 사이의 상당한 잠복을 포함 할 수 있으며 조기 사망에 이르는 장기간의 고통스러운 질병을 수반 할 수 있음. SSK 연구팀의 과제3의 목적은 미세먼지가 소비자의 평생주기 비용에 미치는 영향을 밝혀내는 것이며, 본 회의에서는 통계적생명가치(VSL)과 지불의사액(WTP)를 이용한 논문 두 편을 소개함

연구방법 (Case study)

(1) Valuing Health Risks from Air Pollution: A Review of the Literature and a Conceptual Model (Hong Sok Kim and Up Lim)

Addressing quesiton:

1. To what extent do people discount future morbidity and mortality

2. Do people place a greater value on avoiding an increased risk of death that is preceded by a considerable period of debilitating morbidity than for an increased risk of a quick death?

3. How do people value avoiding long term health risks for exposed children?

4. How do people value avoiding genetic damage that poses health risks to future generations?

Conceptual Theoretical Model : The first model derives the VSL for a single individual with no family or heirs. The second model explores the VSL for each individual family member where both the father and mother share altruistic concern for the well being of a child. Finally, we present a model, again for a single individual, which combines morbidity and mortality.

(2) Valuing Health Risks from Air Pollution: A Review of the Literature and a Conceptual Model (Brian H.S. Kim, Euijune Kim, Jeasun Lee, Minsu Son)

Applying a single WTP for all mortality risks assumes that the value is invariant with the remaining life expectancy of the person at risk. WTP to reduce mortality risk is related to the remaining life expectancy for those at risk, exactly how WTP to reduce mortality risk varies with the remaining life expectancy of the individual has not been determined empirically. Remaining life expectancy is, of course, closely related to age.

WTP Estimation Model ( Choice Model for Attributes Survey )

1) Characteristics of the Individual: Socio-demographic (age; gender; ethnicity; household income; size of household) . Individual’s personal health risks (personal health history; family health history)

2) Characteristics of the Health Risk: Timing (time until onset of illness; time from onset until death or recovery) · Discomfort and impairment (level of discomfort during illness) · Probabilities (of being affected by the illness; of dying if affected)

• The results indicate that PROB_GET_DIE and COST are significant and of the expected sign. Other coefficients are not significant probably because of the correlation of the attribute levels and small sample size. Length of illness and time to onset have the expected signs.

• Assuming time until onset is zero, length of illness is zero, and impairment is zero (which is irrelevant when length of illness is zero), the only relevant variables are probability of getting ill and then dying and the cost. This information represents the individual’s tradeoff between the probability of dying now and cost, similar to accidental death VSL estimates.

• Dividing the marginal utility of a 1 in 100,000 chance of getting ill and dying (PROB_GET_DIE) by the marginal utility of money (COST) and multiplying by 100,000 provides an estimate of the VSL: (-0.06620/ -0.00107) * 100,000 = $6,186,000.