Cairo, Egypt July 26, 2015 ERF workshop on Education in the ERF Region

1. Riham Rizk
The British University in Egypt(BUE)
Hala Abou-Ali
FEPS, Cairo University
Household and Individual Determinants on off-spring’s Educational
Expenditure : Evidence from Arab countries
ERF Workshop on education in The ERF Region

2. Overview
Motivation
Data and methodology
Empirical results
Stylized facts
Conclusion and recommendations

3. Motivation
 Education is at the top of the reform agenda for many policymakers in
the Arab countries especially in Egypt.
 Education is considered an antidote for reducing poverty levels and
promoting economic growth, both at the national and household levels.
 Education is seen as a solution against authoritarian governments and
a condition for achieving democracy.

4. Where is the problem, then ? (1/2)
 The average level of education among MENA population is still very low
compared to East Asia and Latin America.
 The average gross enrollment in the secondary school was: about 75% in MENA, 78% in East
Asia and 90% in Latin America.
 The average gross enrollment in higher education reached about 26% in 2003 in MENA which
represent two-thirds of the other two regions average.
 Spending on education as a percentage of GDP reaches 5% compared to 3.6% and 3.9% in East
Asia and Latin America, respectively.
 This could be attributed to the free education applied to secondary and
higher education compiled. Altogether, lead to low level of Arab countries
human capital.
*Reference :World Bank ( 2008)

5. Where is the problem, then ? (2/2)
 There is a huge concern to public expenditure on education while
continuing ignoring the household expenditure quantum proves to lead
to incorrect long run government educational policies.
 The relation between public and private spending could be either
complementary or substitute.

6. Objectives
 To what extent the amount of household income affects the demand on
education in Arab countries?
 What are the patterns of household education expenditures in Arab
countries? And, to what extent it differs with household and individual
characteristics ?
 What are the various factors that determine household and individual
expenditure on education in Arab countries? Do they differ from one
country to another in the region? And why?

7. Data
ERF-OAMDI Harmonized Household Income and
Expenditure Surveys 2010/2011 for
Egypt,
 Jordan,
 Palestine,
 Tunisia,
 and Sudan 2009.

8. Methodology (1/2)
 To estimate individual and household determinants of
education expenditure , we use Robust OLS model to reduce
Heteroscedasity problem. Thus , logged annual household expenditure
was regressed on various family characteristics such as
 Household income, Father’s education , Mother’s education, Father’s occupation, Mother’s
occupation, number of children at each different age categories and geographic residence.
 The conceptual model follows Mincer’s schooling model :Mincer’s schooling model :
Log HHEX=Log HHEX=αα ++ββggww++γγXXii++ЄЄ

9. Methodology (2/2)
 FirstFirst, Pooled sample model including all individualistic characteristics for the
family is estimated.
 SecondSecond, the full sample was divided into income quintiles and the same model
was estimated separately for each subsample, in order to examine at each
income level the effects of family characteristics on household’s expenditure
pattern on education following Quang (2012).
 ThirdThird , we grouped the sample by children’s age to study the link between the
economic circumstances and patterns of educational allocation among the
school-aged children following Hannum (2005) and Qian & Smyth (2010). The
model was estimated separately for each subsample with pre- and primary
school aged children, secondary-aged and college-aged children focusing
on income effect and geographic location.

10.  FinallyFinally, the Mincerian model is regressed again on the various
household characteristics was namely, household income, head’s
education, head’s occupation, household’s number of children and
household's geographic location.

11. Results highlight(1/4)
The main findings that emerge from this study are three folds:
 Household income has significant effects on the magnitude of
educational expenditure
 Education and occupation of household “head” “Mother’s and father’s”
matter.
 Households place of residence has a significant impact.

12. Results highlight(2/4)
Individual Determinants..
A)Using log Total Household educational expenditure
Log (HH Educational expenditure) Egypt Tunisia Jordon Palestine Sudan
Father's education level
Primary 0.175*** 0.055 0.113 0.290*** -0.026
Secondary 0.302*** 0.171*** 0.283** 0.416*** 0.209***
Tertiary 0.397*** 0.214*** 0.704*** 0.751*** 0.599***
Mather's education level
Primary 0.278*** 0.162** 0.039 0.074 0.279***
Secondary 0.390*** 0.140*** 0.099 0.454*** 0.538***
Tertiary 0.485*** 0.201*** 0.512*** 0.403*** 0.623***
Household decomposition
No. Pre-primary school age children -0.194*** 0.015** -0.445*** -0.227*** -0.200***
No. of Secondary-age children 0.103** -0.125** -0.536** -0.157** 0.081**
No. of college -age children 0.715*** 0.669*** 0.488*** 0.234*** 0.529***
Household income
Income quintile 2 0.290*** 0.234*** 0.024 0.036 0.072
Income quintile 3 0.423*** 0.490*** 0.190* 0.229*** 0.275***
Income quintile 4 0.499*** 0.721*** 0.170* 0.294*** 0.365***
Income quintile 5 1.002*** 1.094*** 0.394*** 0.418*** 0.899***

13. Egypt Jordon Palestine Sudan
Father’s occupation
Cadres 0.327*** -0.712 0.439*** #
Middle professional 0.137** 0.377*** 0.218***
Lower professional -0.038 0.283*** 0.180***
Blue collar and service -0.014 -0.015 0.082
Unemployment 0.015 -0.649*** -0.032
Mother’s occupation
Cadres 0.028 -0.441 0.541**
Middle professional 0.029 0.219* 0.551***
Lower professional 0.035 -0.49 0.185***
Blue collar and service 0.128*** -0.653 -0.079
Unemployment -1.106 -0.054 -0.194
Household Place of residence
Lower Egypt North West Bank Western
-0.371*** -0.397*** 0.105*** 0.155***
Upper Egypt South Eastern
-1.215*** -0.598*** 0.123**
Northern
0.032
N 10276 3046 11394 6144
R-square 0.441 0.459 0.324 0.377

14. Results highlight(3/4)
Table (2): Using Log Share of Educational Expenditure
Log (Education share ) Egypt Jordon Palestine Sudan
Father's education level
Primary 0.169*** 0.09 0.276*** 0.007
Secondary 0.298*** 0.219** 0.430*** 0.220***
Tertiary 0.442*** 0.638*** 0.762*** 0.689***
Mather's education level
Primary 0.301*** 0.094 0.039 0.201***
Secondary 0.403*** 0.175 0.402*** 0.410***
Tertiary 0.590*** 0.572*** 0.360*** 0.538***
Household decomposition
No. Preschool age children -0.198*** -0.443*** -0.228*** 0.017
No. of secondary age children 0.706*** 0.497*** 0.232*** 0.276***
No. of college -age children -0.032** 0.504*** 0.515*** 0.425***
Household income
Income quintile 2 -0.03 -0.388*** -0.722*** -0.800***
Income quintile 3 -0.121*** -0.495*** -0.936*** -1.127***
Income quintile 4 -0.286*** -0.869*** -1.235*** -1.504***
Income quintile 5 -0.269*** -1.235*** -1.735*** -1.837***

15. Egypt Jordon Palestine Sudan
Household Place of
residence
Lower Egypt North West Bank Western
-0.346*** -0.367*** 0.105*** 0.288***
Upper Egypt South Eastern
-1.162*** -0.543*** 0.014
Northern
-0.003
N 10276 3046 11394 6144
R-square 0.441 0.382 0.314 0.267

16. Results highlight(4/4)
Household Determinants of Education Expenditure
Egypt Tunisia Jordon Palestine Sudan
Dependent variable : Log educational Expenditure
Household Income
Second quintile 0.367*** 0.259*** 0.401*** 0.142 0.217***
Third quintile 0.542*** 0.502*** 0.755*** 0.243** 0.341***
Fourth quintile 0.711*** 0.776*** 0.862*** 0.573*** 0.465***
Fifth quintile 1.190*** 1.117*** 1.694*** 0.768*** 0.985***
Household's head
education
Primary 0.230*** 0.042 0.560*** 0.297*** 0.237***
Secondary 0.490*** 0.186*** 1.275*** 0.570*** 0.519***
College 0.812*** 0.303*** 1.521*** 0.800*** 0.986***
Number of school -aged
children
No. of pre-Prim school-
aged children 0.126*** 0.213*** -0.166*** -0.133*** 0.003
No. of secondary school
aged children 0.438*** 0.304*** 0.09 0.055 0.316***
No. of college-aged
children 0.090*** 0.223*** 0.305*** 0.527*** 0.378***
N 4233 6406 1881 2659 2624
R-square 0.391 0.229 0.257 0.229 0.255

17. Stylized Facts (1/3)

18. Stylized Facts (2/3)

19. Stylized Facts (3/3)

20. Recommendations
Conditional cash transfers
Subsidies

21. Thank you
REHAM RIZK
ASSISTANT PROFESSOR
BRITISH UNIVERSITY IN EGYPT
REHAM.RIZK@BUE.EDU.EG