Date post: | 11-Jan-2016 |
Category: |
Documents |
Upload: | meryl-west |
View: | 214 times |
Download: | 0 times |
Human Capital, Consumption and Human Capital, Consumption and Housing Wealth in Transition Housing Wealth in Transition
Jarko Fidrmuc ZU Friedrichshafen, Comenius University Bratislava, IOS
Regensburg
Matúš Senaj National Bank of Slovakia and Comenius University
Bratislava
Bratislava Economic Meeting
8th. Jun 2012 2
The views expressed are those of the authors and do not necessarily reflect those of the National Bank of Slovakia or the Eurosystem.
8th. Jun 2012 3
MotivationMotivation
Microeconometric analysis: households surveys on income, expenditures, and housing.
What is the impact of age, experience and human capital on disposable income, consumption and housing value.
The relationship between income, real estate wealth and private consumption (welfare).
Political economy implications with regard to support of economic reforms in Eastern Europe. Who are the winners and losers of economic reforms?
8th. Jun 2012 4
Human Capital and TransitionHuman Capital and Transition It is generally acknowledged that the CEECs
have a relatively efficient education system especially in field of natural and technical sciences.
„Conventional wisdom holds that if there is one area where the countries of Eastern Europe and the former Soviet Union are well served, it is with respect to their stocks of human capital.” (Campos and Coricelli, 2002, JEL, p. 808)
However, education of economics, business administration, marketing, and foreign languages were often insufficient.
8th. Jun 2012 5
Our ContributionOur Contribution
We estimate the returns to human capital in Slovak economy.
We augment this discussion by an analysis of real estate wealth and consumption.
We show that cohorts entering the labor market before or after 1989 have highly different equipments of human and physical capital, which corresponds to differential institutional settings in both periods.
Through the estimation of the consumption function we can get a proxy for the welfare effects.
8th. Jun 2012 6
Data DescriptionData Description We merge two data sets in our paper. (1) we use the household budget survey (HBS)
conducted annually by the Statistical Office. (2) we calculate the value of real estates owned
by the households using the database of housing prices published by the NBS.
The final data set representative data for more than 4500 households for every year in all Slovak regions.
Households are available only for 1 year. The recent data set covers ranges from 2004 to 2009,
which reflect the change of methodology of HBS in 2004.
Weights ensure that the results are representative.
8th. Jun 2012 7
Income Profile Income Profile 20
040
060
080
010
00M
onth
ly d
ispo
sab
le in
com
e o
f th
e ho
use
hold
20 40 60 80Age of the household head
8th. Jun 2012 8
Consumption Profile Consumption Profile 30
040
050
060
070
080
0M
onth
ly c
onsu
mp
tion
of t
he h
ouse
hol
d
20 40 60 80Age of the household head
8th. Jun 2012 9
Home Ownership Home Ownership .5
.6.7
.8.9
1H
om
eow
ner
ship
ra
te
20 40 60 80Age of the household head
8th. Jun 2012 10
Housing Wealth Housing Wealth 20
000
4000
060
000
8000
0H
ous
ing
wea
lth in
EU
R
20 40 60 80Age of the household head
8th. Jun 2012 11
Income EquationIncome Equation We estimate the income equation with the
following explanatory variables: Age and squared age of the household head; Household size: number of adults, number of
children, dummy for single households; Gender variable of the household head; Education: secondary, tertiary for household head
and the partner; Cohort variable – labor market entrance (age 26)
before 1990 Cohort education variable (dummies & years of
education) Regional and time dummies.
8th. Jun 2012 12
Income Equation – OLS I Income Equation – OLS I Income 1 Income 2 Income 3
Age 0.095*** 0.161*** 0.163***
Age squared -0.016*** -0.020*** -0.021***
No of adults 0.271*** 0.276*** 0.275***
No of children 0.058*** 0.052*** 0.053***
Female head -0.048*** -0.048*** -0.049***
Single parent -0.230*** -0.226*** -0.227***
Education primary -0.122*** -0.122*** -0.328***
Education tertiary 0.185*** 0.185*** 0.187***
Partner’s ed. primary -0.131*** -0.131*** -0.128***
Partner’s ed. tertiary 0.137*** 0.136*** 0.135***
Cohort 90 -0.071***
Cohort 90 x primary education 0.148***
Cohort 90 x secondary education -0.074***
Cohort 90 x tertiary education -0.076***
Constant 9.115*** 8.958*** 8.961***
Sample All households All households All households
No. of obs. 27,650 27,650 27,650
R2 0.641 0.642 0.643
8th. Jun 2012 13
Income Equation – OLS II Income Equation – OLS II Income 4 Income 5 Income 6
No of adults 0.244*** 0.292*** 0.275***
No of children 0.030*** 0.064*** 0.049***
Female head -0.076*** -0.019 -0.039***
Single parent 0.090 0.119*** -0.227***
Years of education 0.488*** 0.265***
Years of partner’s education 0.261*** 0.304***
Work experience 0.377*** -0.242**
Work experience squared -0.103*** -0.050
Years of education before 1990 0.415***
Years of education after 1990 0.494***
Work experience before 1990 -0.173***
Work experience after 1990 0.096***
Work exp. before 1990 squared 0.041***
Work exp. after 1990 squared 0.036***
Constant 8.135*** 8.197*** 8.604***
Sample Cohort of
younger h. Cohort of
older h. All
households No. of obs. 8,674 19,288 27,962
R2 0.464 0.665 0.637
8th. Jun 2012 14
Income - Summary of ResultsIncome - Summary of Results Very standard results. Concave age profile. Negative gender differential. Positive returns to education, especially for higher
education. Returns to education are also gender specific. Moreover, they differ between the cohorts. Labor
market entrants from before 1990 receive lower returns to education.
The difference for basic education is not significant. High regional and time effects, slower growth after
the financial crisis.
8th. Jun 2012 15
Household WealthHousehold Wealth
About 95% of household owns a house or an apartment.
For the cohort entering labor market before 1990, 97% of households own a real estate.
The main explanatory variables is income and the cohort variable.
There may be a truncation or selection problem. Therefore we use Heckman selection model.
8th. Jun 2012 16
Heckman Selection ModelHeckman Selection Model
We estimate a linear Heckman selection model. In the first step, we estimate the probability that a
respondent has house In the second step, we estimate the value of his
property Age variables are used as instrumental variables
for the identification of the selection equation.
8th. Jun 2012 17
Housing Wealth - Heckman Selection Housing Wealth - Heckman Selection Model Model
Housing 3 Housing 4 Housing 5 Housing 6
Disposable income 0.061*** 0.043*** 0.061*** 0.054***
No of family members 0.050*** 0.056*** 0.063*** 0.062***
Single-parent -0.058*** -0.045*** -0.060*** -0.058***
Cohort 90 0.133*** 0.093***
Constant 13.434*** 13.619*** 13.286*** 13.395***
Homeownership – selection model
Disposable income 0.370*** 0.336*** 0.360*** 0.336***
Single parent -0.199*** -0.238*** -0.206*** -0.238***
No of family members 0.042 0.019 0.034 0.019
Age 1.023*** 1.093*** 1.058*** 1.093***
Age squared -0.067*** -0.076*** -0.071*** -0.076***
Cohort 90 0.041 -0.000 -0.031 -0.000
Constant -5.301*** -4.931*** -5.178*** -4.931***
No. of observations 27,965 27,965 27, 965 27,965
No. of censored obs. 1,674 1,674 1,674 1,674
rho -0.447 -1.000 -0.275 -0.701
8th. Jun 2012 18
Consumption EquationConsumption Equation
Income and consumption are endogenous variables.
We instrument income with gender and small town, because these variables are not correlated with residuals from the consumption equation, as confirmed by Hansen J-test.
Housing ownership may also influence consumption trough lower savings and better access to credits.
We keep housing ownership as exogenous variable, although it could be endogenous.
8th. Jun 2012 19
Consumption Function – IV Consumption Function – IV Results Results
Consumption 1 Consumption 2 Consumption 3 Consumption 4
2SLS 2SLS GMM GMM
Disposable income 0.874*** 0.862*** 0.872*** 0.860***
Housing wealth -0.001 0.001 -0.001 0.001
Primary education -0.113*** -0.106*** -0.114*** -0.107***
Tertiary education 0.015* 0.017** 0.014* 0.016**
Constant 1.149*** 1.280*** 1.170*** 1.296***
Cohort 90 -0.047*** -0.046***
No of observations 27 965 27 965 27 965 27 965
Centered R2 0.620 0.623 0.620 0.624
Hansen J statistic 0.164 0.157 0.164 0.157
Hansen p-value 0.686 0.692 0.686 0.692
8th. Jun 2012 20
ConclusionsConclusions
Employees with education gained before 1990 have significantly lower returns to education.
By contrast, they have accumulated significantly higher stock of physical capital (housing).
In sum, the impact on household welfare (consumption) is ambiguous.
8th. Jun 2012 21
Thank you for your attentionThank you for your attention