Transcript Democracy, Development, and Distribution

Democracy, Development, and
Distribution
What we know, what we do not, and
how we learn?
5000
4000
0
0
1000
2000
3000
lowess level year
3000
2000
1000
lowess gdpcap_world year
4000
5000
Development in the long run
1500
1600
1700
1800
year
1900
2000
1500
1600
1700
1800
year
1900
2000
1000000
500000
lowess popm_gcm year
30000
20000
0
10000
lowess popm_gcm year
40000
1500000
Population in the long run
1500
1600
1700
1800
year
1900
2000
1500
1600
1700
1800
year
1900
2000
.7
.6
.5
.4
lowess legelec year
.8
.9
Elected legislatures in the long
run
1800
1850
1900
year
1950
2000
Plan
• Development and democracy
• Political regimes and development
• Inequality, political regimes, and
development
• What to teach?
Dynamics of Political Regimes,
1950-2000
• Development and Democracy (Lipset 1959)
• Transitions to Democracy
• Survival of Democracy
.6
.4
.2
Pr(dem)
.8
1
Development and Democracy
0
10000
20000
gdpcap
30000
40000
0.1
0.0
-0.1
-0.2
Estimated transition probability
0.2
Development and Transitions to
Democracy
0
5000
10000
Lagged GDP/cap
15000
0.05
0.0
-0.05
TDA
0.10
0.15
Development and Survival of
Democracy
0
1000
2000
3000
GDP.cap
4000
5000
6000
Conclusion
• The reason there are democracies in the
developed countries is that they never die at
high income levels, not that transitions to
democracy become more likely with
development.
Democracy and Development
• Rate of growth of per capita income = Rate of
growth of total income – Rate of growth of
population.
• Political regimes do not affect the rate of growth
of total income (GDP).
• Surprisingly, population growth is higher in
dictatorships, because of higher fertility per
woman.
• As a result, per capita income grows higher under
democracy.
Observed averages
Dictatorships
Democracies
Growth of total
output
4.27
3.68
Growth of
population
2.42
1.35
Growth of per
capita output
1.85
2.33
Methodological Issues
• Development = Regimes, Observed
Conditions + Unobserved Conditions
• Problem: Distinguish the effect of regimes
(generally institutions) from the effect of
observed and unobserved conditions.
• Different ways of doing it may generate
different results.
3.5
3
2.5
2
lowess yg gdpcap
4
4.5
For example, at different income
levels
0
5000
10000
15000
gdpcap
20000
25000
Estimated differences
YG
POPG
G
Observed
0.59
1.03***
-0.48
Match
0.44
0.37***
0.07
Heckman 2 0.59
0.48***
0.11
Heckman 1 0.32
0.59***
-0.27
2SLS
0.57**
-0.26
0.33
Conclusions
• Regimes do not make difference for
economic growth but for population growth.
• Do not conclude that political institutions
do not matter.
• Topic of intense research.
Which institutions?
•
•
•
•
Protect property rights
Coordinate investment
Make rulers accountable to citizens
Other?
Inequality
• Income distribution is very stable over long
periods.
• Major redistributions of assets are very rare and
result from cataclysmic events.
• Measured on individual/household basis,
inequality declined in the world in the past 20
years, because of China and India.
• But two-thirds of countries became more unequal.
14
12
4
6
8
10
lowess qratio gdpcap
10
8
6
4
lowess qratio gdpcap
12
14
Q5/Q1 by regime and gdp/cap
0
10000
20000 30000
gdpcap
40000
0
10000
20000 30000
gdpcap
40000
Selection corrected differences
Dictatorship Democracy Difference
N
93
238
Observed
10.50
9.19
Match
Heckman 2 11.27
1.31
3.84
10.75
0.52
Heckman 1
-1.38
2SLS
-1.25
2
0
-2
lowess yg qratio
4
6
Inequality and Development
0
10
20
qratio
30
40
10
8
6
lowess qratio gdpcap
12
But development and inequality
0
10000
20000
gdpcap
30000
40000
Conclusion
• We cannot tell what causes what.
• Moreover, the data are very bad.
• As important as inequality is, studying its
causes and effects is hard.
What we do not know?
• What makes political institutions change.
• What kinds of institutions, if any, affect
development.
• What is the impact of inequality on
development and democracy and of
development on inequality.
What to teach?
• How to learn.
• Basic knowledge: philosophy, history.
• Basic tools: models, statistics.
Why?
• Philosophy: Questions, ideas, concepts that
survived the test of time.
• History: Broaden perspectives, put each
case in context.
• Models: Clarify arguments; be sure they are
consistent.
• Statistics: Do the models explain reality?
Marginal value of income
1
0.8
0.6
0.4
0.2
0
10
20
y
30
40
50
Value of democracy
1
0.8
0.6
0.4
0.2
0
10
20
y
30
40
50
Why democracies survive in
wealthy societies?
1
0.8
0.6
0.4
0.2
0
10
20
y
30
40
50
The role of inequality
1
0.8
0.6
0.4
0.2
0
5
10
15y
20
25
30