Transcript Presentation

Company
LOGO
New Paradigm for
International Insurance Comparison:
With an Application to Comparison of Seven Insurance Markets
Wei Zheng, Peking University
Yongdong Liu, China Academy of Sciences
Yiting Deng, Peking University
Outline
1. Introduction
2. Comparison of Insurance Growth Level
3. Comparison of Insurance Growth Structure
4. Economic and Institutional Factors in Insurance Growth
5. Conclusion
2
1. Introduction
 Commonly used methods for international insurance
comparison
 premium income method
 insurance density method
 insurance penetration method
 Limitations of the above methods
 They fail to take into consideration the relationship between insurance
penetration and economic development stage
 A new paradigm is proposed
 BRIP: comparison of Insurance Growth Level
 Trichotomy: comparison of Insurance Growth Structure
3
1. Introduction
 Application to seven markets
 U.S.
 Japan
 U.K.




Brazil
Russia
India
China
 Sometimes we also refer to data of OECD average, BRIC
average and world average.
4
2. Comparison of Insurance Growth Level
2.1 New Method: BRIP
2.2 Ordinary model of insurance growth
2.3 Comparison of Ranking Results under the New
Method and the Traditional Methods
5
2.1 New Method: BRIP
 Benchmark Ratio of Insurance Penetration
actual penetration
BRIP 
100%
benchmark penetration
 “benchmark penetration” refers to “the world average insurance
penetration at a country’s economic level”
 “actual penetration” refers to a country’s actual penetration
 The central idea here is to measure the “benchmark-adjusted
insurance growth level” instead of a traditional one.
 The difficulty is how to get the benchmark penetration.
6
2.2 Ordinary model of insurance growth
 Carter & Dickinson (1992) and Enz (2000) developed a
logistic model to depict the relationship between insurance
penetration and GDP per capita.
1
Y

X
C1  C 2  C3
 Y: insurance penetration
 X: GDP per capita
 C1, C2, C3: three parameters
 ε: residual
 This paper uses the data of 95 countries (regions) over the past
27 years (1980-2006) as the sample.
7
Estimates of “Ordinary Growth Model”
Life
Insurance
Non-Life
Insurance
Insurance
Industry
C1
24.37***
(16.59)
35.45***
(47.53)
14.47***
(33.35)
C2
111.03***
(12.83)
62.72***
(19.93)
42.07***
(17.32)
C3
0.8671***
(68.14)
0.8276***
(51.46)
0. 8592***
(81.33)
R2
0.5362
0.8115
0.7393
Adjusted-R2
0.5356
0.8112
0.7389
Number of Observations
2,052
2,071
2,011
The Robust t-statistics is in parentheses. The term of “***” means the level of
significance is 1%.
8
Regression Curves of “Ordinary Growth Model”
14%
12%
Penetration
10%
8%
6%
4%
2%
0%
10
1000
100
10000
100000
GDP per Capita (US Dollars)
Real Insurance Penetration
Non-Life Insurance Growth Curve
Life Insurance Growth Curve
9
Insurance Growth Curve
Why use BRIP ?
The international insurance comparison will make
more sense only when it is based on the comparable
“benchmark-adjusted insurance growth level”.
BRIP is such a “benchmark” adjustment to insurance
penetration.
So, BRIP represents a more reasonable indicator for
the international insurance comparison.
10
What’s the economic implications of BRIP ?
BRIP = 1: the country’s actual penetration is equal to
the world average penetration at that country’s
economic development stage
BRIP < 1: the actual penetration is less than the
average
BRIP > 1: the actual penetration is greater than the
average
There is a positive correlation between the BRIP and
the relative insurance growth level of that country.
11
2.3 Comparison of Ranking Results (2006)
Traditional methods
Market
BRIP
premium
Insurance
density
Insurance
penetration
GDP
per capita
U.S.
26
1
6
14
15
Japan
14
2
9
7
16
U.K.
4
3
1
1
26
Brazil
36
19
49
44
86
Russia
52
22
52
56
80
India
5
15
76
31
157
China
27
9
70
47
122
12
To sum it up,
We should have a new recognition for the insurance
growth level of each country:
 the benchmark-adjusted insurance growth level of the
emerging countries is not as low as what traditional
methods indicate
 the benchmark-adjusted insurance growth level of the
developed countries is not as high as what traditional
methods imply
 Put it in another way, for the year 2006, the ranking of the
growth potential of the seven countries would be like this
(from large to small): Russia, Brazil, China, US, Japan,
India and UK
13
3. Comparison of Insurance Growth Structure
3.1 Introduction to “Trichotomy”
3.2 Adjusted model of insurance growth
3.3 Comparison of Growth Structure
14
3.1 Introduction to “Trichotomy”
Insurance growth can be decomposed into three parts
 Regular growth
• Insurance growth accompanying the economic growth
assuming the insurance penetration is unchanged
 Deepening growth
• Insurance growth brought about by the increase of insurance
penetration induced by economic growth
 Institutional growth
• The remaining part of the growth, which is brought about by
the institutional factors after the economic factors have been
deducted
15
Trichotomy of Insurance Growth Structure
Penetration
D
Adjusted Growth Curve of
World Insurance
C
B
A
GDP per Capita
16
Trichotomy of Insurance Growth Structure
Penetration
D
Adjusted Growth Curve of
World Insurance
C
B
A
GDP per Capita
17
Trichotomy of Insurance Growth Structure
Penetration
D
Adjusted Growth Curve of
World Insurance
C
B
A
GDP per Capita
18
3.2 Adjusted model of insurance growth
94
1
Y '
  i Di  
'
'X
C1  C2  C3
i 1






Y : insurance penetration
X : GDP per capita
C’1, C’2, and C’3 : three parameters
Di（i=1,…94）: country dummy with respect to country i
λi（i=1,…94）: coefficient for Di
ε: residual
19
Estimates of “Adjusted Growth Model”
Life
Insurance
Non-Life
Insurance
Insurance
Industry
C1
10.76***
(24.22)
40.09***
(14.07)
8.49***
(26.23)
C2
154.27***
(5.24)
155.35***
(5.31)
76.65***
(6.93)
C3
0.8408***
(110.54)
0.7367***
(28.82)
0. 8505***
(126.74)
R2
0.9079
0.9508
0.8771
Adjusted-R2
0.9033
0.8112
0.7389
Number of Observations
2,052
2,071
2,011
The Robust t-statistics is in parentheses. The term of “***” means the level of
significance is 1%.
20
Regression Curves of “Adjusted Growth Model”
14%
12%
Penetration
10%
8%
6%
4%
2%
0%
10
100
1000
10000
100000
GDP per Capita (US Dollars)
Real Insurance Penetration
Life Insurance Growth Curve
21
Non-Life Insurance Growth Curve
Insurance Growth Curve
3.3 Comparison of Growth Structure
Economic Factors (%)
Institutional Factor(%)
Regular growth
Deepening growth
U.S.
78
37
-15
Japan
69
27
4
U.K.
34
24
41
Brazil
24
6
71
Russia
25
8
67
India
22
2
76
China
5
9
86
OECD Average
63
34
3
BRIC Average
15
20
65
World average
63
6
31
22
To sum it up,
In developed countries, the insurance growth is
mainly driven by the economic factors (including
regular and deepening factors)
In emerging countries, the insurance growth is largely
driven by the institutional factors
23
4. Economic and Institutional Factors in Insurance Growth
4.1 Comparison of Two Growth Models
4.2 Discussion on “Institutional Factors”
4.3 Discussion on Developed and Emerging Countries
24
4.1 Comparison of Two Growth Models
Ordinary growth model
 combines both the economic factors and institutional
factors that influence the insurance growth
Adjusted growth model
 separates the country-specific institutional influences and
the common economic influences
25
Comparison of Two Models for Life Insurance
14%
12%
Penetration
10%
8%
6%
4%
2%
0%
10
100
1000
10000
GDP per Capita (US Dollars)
Real Life Penetration Adjusted Growth Curve Ordinary Growth Curve
26
100000
Comparison of Two Models for Non-Life Insurance
14%
12%
Penetration
10%
8%
6%
4%
2%
0%
10
100
1000
GDP per Capita (US Dollars)
Real Non-Life Penetration Adjusted Growth Curve
27
10000
Ordinary Growth Curve
100000
Comparison of Two Models for Insurance Industry
14%
12%
Penetration
10%
8%
6%
4%
2%
0%
10
100
1000
GDP per Capita (US Dollars)
Real Insurance Penetration Adjusted Growth Curve
28
10000
Ordinary Growth Curve
100000
Comparison of Two Models for Insurance Industry
 In the figure
 Ordinary growth curve: combines both economic and institutional
factors
 Adjusted growth curve: reflects only pure economic factors
 When GDP per capita is low, the ordinary curve is higher than
the adjusted curve, which indicates that institutional factors
facilitate the growth of the insurance industry to some degree.
 When GDP per capita is high, the ordinary curve is obviously
lower than the adjusted curve, which indicates that institutional
factors markedly restrain the growth of the insurance industry.
29
4.2 Discussion on “Institutional Factors”
Major institutions
 social security system (systematic institution)
• dominantly affects the life insurance
 legal system (systematic institution)
• dominantly affects the non-life insurance, with its most
typical components being the compulsory insurance and
liability insurance
 culture (non-systematic institution)
 religion (non-systematic institution)
30
Effects of institutional factors on life insurance
 Relationship between life insurance and social security
 usually substitutable
 the better developed the social security system is, the more the life
insurance growth is restricted
 Relationship between social security and GDP per capita
 usually positive correlation
 low GDP per capita countries: social security system is usually underdeveloped
 high GDP per capita countries: social security system is usually welldeveloped
 Thus, as the GDP per capita increases (with the improvement
of social security system), the negative effects of institutional
factors on life insurance would gradually increase.
31
Effects of institutional factors on non-life insurance
 Relationship between non-life insurance and certain legal
policies
 usually complementary
 the more compulsory insurance and liability insurance are implemented,
the more growth opportunities will be created for the non-life insurance
 Relationship between certain legal policies and GDP per capita
 usually no direct relation
 the government’s decision of whether to adopt those legal policies (the
compulsory insurance and liability insurance) is mainly based on the
consideration of social policy (such as equity and justice), and
generally is not related to GDP per capita
 Thus, no matter how large GDP per capita is, institutional
factors will always bring positive effects to the growth of nonlife insurance.
32
Effects of institutional factors on insurance industry
When GDP per capita is low
 institutions have some positive effects on both the life
insurance and the non-life insurance
 with its net effects on the insurance industry being positive
When GDP per capita is high
 institutions have remarkably negative effects on the life
insurance and some positive effects on the non-life
insurance
 with its net effects on the insurance industry being negative,
and the negative effects are notable
33
4.3 Discussion on Developed and Emerging Countries
 For the emerging countries
 institutional factors facilitate the growth of the insurance industry to
some degree
 For the developed countries
 institutional factors notably restrain the growth of the insurance
industry
 It could also imply that as the economy develops, the
contribution of the institutional factors to the insurance growth
would gradually decrease, and the economic factors would
play a more active role in driving the insurance growth.
34
4.3 Discussion on Developed and Emerging Countries
 This implication suggests that, for those emerging countries,
after the insurance industry having experienced a period of
“taking-off”, its growth will gradually change from being
“driven by both economic and institutional factors” to being
“driven mainly by economic factors”.
 Following this judgment, it is extremely important for the
insurance industry in the emerging countries to upgrade its
growth strategy from the extensive developing pattern to a
refined and sustainable developing pattern, for the former one
will lose its foundation for surviving.
35
5. Conclusion
1. We should have a new recognition for the insurance growth
level of each country.
 BRIP gives a different and probably more reasonable answer
2. The insurance growth in developed countries is mainly driven
by the economic factors, while that in emerging countries is
largely driven by the institutional factors.
3. As the economy develops, the contribution of the institutional
factors to the insurance growth would gradually decrease, and
the economic factors would play a more active role in driving
the insurance growth.
36
Company
LOGO
Thank you for your kind attention！
Comments are welcome！
[email protected]