Transcript Explaining Inequality the World Round: Cohort Size, Kuznets Curves

Explaining Inequality the World Round:
Cohort Size,
Kuznets Curves, and Openness.
By Matthew Higgins and Jeffrey G. Williamson
The three hypothesis
• Inequality and cohort size
• Inequality and openness
• Kuznet curve hypothesis
Cohort size hypothesis
• „Fat age cohorts tend to get low rewards“
• Important where the fat cohorts lie in the demographic structure
• Cohort size variance across countries vs. within countries over time
Inequality and openness
• Rising inequality correlated with increasing globalization
• Trade (Heckscher- Ohlin argument)
• Immigration
• Labor supply shifts important
• Globalization as cause for falling inequality in developing countries
and increasing inequality in developed countries?
Kuznets curve hypothesis I – Strong version
• Inequality first decreases and then declines during economic
development
• Demand for different skill levels causes the inequality development
Kuznets curve hypothesis II – Weak version
• Demand forces not necessarily the reason for Kuznet movements
• Other forces possible
• Forces of some demographic transition (cohort size)
• Different policies (attitude towards liberal policies, schooling,…)
• Natural resource endowments
EMPIRICS
• Pooled dataset of 111 countries, 1960-1990ies
• Variables:
•
•
•
•
•
Decadal average of Real GDP per worker (RGDPW)
RGDPW2
GINI-coefficent
Log of (Q5/Q1) income ratio (GAP)
8 dummies, including decadal dummies
Problems
• Simultaneity bias - Instrumental variables?
• Omitted variable bias
• Heteroskedasticity - heteroskedasticity-robust-estimators
Monotonically
declining
Inverted U,
but
imprecise
estimations
Extension of the model
• MATURE: share of population with age between 4059
• OPEN: Sachs-Warner index
(i) a black market premium of 20 percent or more for
foreign exchange
(ii) an export marketing board which appropriates most
foreign exchange earnings
(iii) a socialist economic system
(iv) extensive non-tariff barriers on imports of intermediate
and capital goods.
Quite high
Lacks
degrees of
freedom!
• M3/GDP is a measure for financial depth (how sophisticated are
financial institutions in a country, how many have access to these…)
• FREEDOM geometric average of two indices, one measuring civil
liberties and one measuring political rights
Alternative measures for openness:
• quantitative and tariff restrictions on imports,
• the share of imports plus exports in GDP
• Natural level of openness:
the logs of country size, population, per capita income, per capita crude proven oil
reserves, the average distance from trading partners, and two dummy variables describing,
respectively, whether a country is an island or is landlocked
Globalization and Inequality
• Milanovic (2005) proxies openness by the ratio of trade to GDP, finds
negative effect of openness on inequality in poor countries
• Ravallion (2001) as well
• Dollar and Kraay (2000, 2002) find that openness has no impact on
inequality
Alternative demographic measures:
•
•
•
•
•
total fertility rate
population growth rate
labor-force growth rate
the infant mortality rate
life expectancy at birth
Fertility rate
• De La Croix and Doepke (2003) find negative correlation between high
fertility and growth, via education
• Barro (2000): High correlation between inequality and fertility
Fixed country effects
• Dummy variable for every country
• Unbiased and consistent under the assumption that unobserved
effects are correlated with the explanatory variables
• Loss in efficiency
• Serial correlation -> LDV
Explaining cohort size effect on inequality
• Three channels
• 1) altered age structure, leaving age-earnings profile constant (+)
• 2) different age groups with different within inequality levels (+)
• 3) age-earnings structure changing, different experience premium (-)
Simulations - I
• Three key sets of parameters
• 1) Age profile of labor productivity over the lifecylce
• 2) Age profile of the variance of earnings over the lifecylce
• 3) Elasticity of substitution between different age groups
Simulations - II
• Treat estimated mean age-income as representing the age-profile of
labor productivity
• Select various values for the elasticity of substitution across age
groups
• Evaluate inequality indexes associated with various steady-state
population growth rates
Simulations - III
• Perfect substitutability
• Small effect by changing mix between older and younger workers
caused by different wage levels (32.5 to 32.1)
• Larger effect caused by different variance within age groups (43.1 to
39.7)
• Taking mean-earnings and variance effects together: similar effect
Simulations - IV
• Imperfect elasticity of substitution necessary for cohort size effect to
be true
• 3.0 : offsets first two channels
• suggests lower elasticity of substitution
• Barro, R. J. (2000). Inequality and Growth in a Panel of Countries.
Journal of economic growth, 5(1), 5-32.
• De La Croix, D., & Doepke, M. (2003). Inequality and growth: why
differential fertility matters. The American Economic Review, 93(4),
1091-1113.
• Dollar, D., & Kraay, A. (2002). Growth is Good for the Poor. Journal of
economic growth, 7(3), 195-225.
• Milanovic, B. (2005). Can we discern the effect of globalization on
income distribution? Evidence from household surveys. The World
Bank Economic Review, 19(1), 21-44.
• Ravallion, M. (2001). Growth, inequality and poverty: looking beyond
averages. World development, 29(11), 1803-1815.