### 风险厌恶程度

Figure 9.1: US expenditures of selected goods and services as share of Gross Domestic Product (19602004)
18
16
Percentage of GDP
14
12
10
8
6
4
2
0
1960
1964
1968
1972
1976
1980
1984
1988
1992
1996
2000
Year
Health
Food
Clothing and Shoes
Housing
SOURCE: US Census Bureau [2006, pp. 98, 443], and National Income and Product Accounts
(http://www.bea.gov/bea/dn/nipaweb/index.asp.
2004

• 政府提供的抵御风险的手段（养老、医疗、失业、工

• 美国的社会保险
• Medicaid
• Medicare
• Social Security
• Unemployment Compensation

• 风险规避倾向
• 保险费的决定
• Expected Value
• Expected value (EV) = probability of outcome 1) *
(Payout in outcome 1) + probability of outcome
2)*(Payout in outcome 2) + … + (probability of
outcome n)*(Payout in outcome n)

AB之间的距离表示风险

(A)
(B)
(C)
Income
if She
Stays
Healthy
Income if
She Gets
Sick
Expecte
d Value
Insurance
Options
Income
Probability
of Staying
Healthy
Probability
of Getting
Sick
Lost
Income if
She Gets
Sick
Option 1: No
Insurance
\$50,00
0
9 in 10
1 in 10
\$30,000
\$50,000
\$20,000
\$47,000
Option 2: Full
Insurance
(\$3,000
cover \$30,000
in losses
\$50,00
0
9 in 10
1 in 10
\$30,000
\$47,000
\$47,000
\$47,000
Actuarially Fair Insurance Policy
Utility

B
UB
UD
UC
D
C
• Expected
Utility
A
• Risk
Smoothing
UA
20,000
47,000 50,000
Income

• Risk Aversion

• Risk Aversion：风险厌恶程度，效用函数的形状弯曲程度

•
•
•
•
Insurance in a small population
Insurance in a large population
Law of large numbers大数定律

selection)
• 信息不对称问题
• 投保人比保险公司拥有更多信息
• 保险公司按照平均风险概率和支出制定保险费，由此，

• 有效率的做法应该是对高风险的人收高保费，对低风

(A)
Probability of
Insurance
Getting Sick
(B)
(C)

(D)
(E)
(F)
Expected
Benefit
Expected
Benefit
Expected
Benefit
Minus
Minus
\$3,000)
\$4,500)
Lost
Income
Expected
Minus
if Sick
Lost
Income
(Differential
Emily
1 in 5 (High Risk)
\$30,000
\$6,000
\$0
\$3,000
\$1,500
Jacob
1 in 5 (High Risk)
\$30,000
\$6,000
\$0
\$3,000
\$1,500
Emma
1 in 5 (High Risk)
\$30,000
\$6,000
\$0
\$3,000
\$1,500
Michael
1 in 5 (High Risk)
\$30,000
\$6,000
\$0
\$3,000
\$1,500
1 in 5 (High Risk)
\$30,000
\$6,000
\$0
\$3,000
\$1,500
Joshua
1 in 10 (Low Risk)
\$30,000
\$3,000
\$0
\$0
-\$1,500
Olivia
1 in 10 (Low Risk)
\$30,000
\$3,000
\$0
\$0
-\$1,500
Matthew
1 in 10 (Low Risk)
\$30,000
\$3,000
\$0
\$0
-\$1,500
Hannah
1 in 10 (Low Risk)
\$30,000
\$3,000
\$0
\$0
-\$1,500
Ethan
1 in 10 (Low Risk)
\$30,000
\$3,000
\$0
\$0
-\$1,500
\$0
-\$15,000
\$0
Insurer's
Net Profits

• 逆选择是否是政府干预市场的理由？市场是否有能力

• Experience rating：保险公司搜集信息，筛选客户，提

• Experience rating and equity：政府的强制性保险政策，

• Community rating：团体费率

•
•
•
•

Price per unit

Flat-of-the-curve medicine
P0
a
b
h
.2P0
0
Sm
Dm
M0
M1
Medical services per year

• The Elasticity of Demand for Medical Services道德风险

• Social experiments find that the elasticity of demand for
health care is -0.20.
• 道德风险还与共保率有关

• 道德风险引起的效率问题不是私人市场独有的，只要

• 政府并不比市场能更好地控制道德风险
• 医疗保险的作用：在分担风险和道德风险之间寻找平

• 信息问题：人们对疾病的知识有限；确定医疗质量的

• 外部性：如传染病预防
• 上述两方面的问题是政府干预的理由之一

Health Expenditures as Percentage of GDP
Figure 9.5: Expenditures on health care as share of Gross Domestic Product, selected countries
(1960-2004)
16
14
12
10
8
6
4
2
0
1960
1970
1980
1990
2000
2004
Year
Australia
France
Germany
Japan
United Kingdom
United States

• 可能的原因
• The “graying” of America – older populations require
more health care
• Income growth – health care is a normal good
• Third party payments – insurance coverage may have
changed
• Improvements in quality – treatments are very
different (better and more expensive) than in previous