Transcript Unleashing The Potential Of Ethiopian Women: Trends

Unleashing the Potential of Ethiopian
Women: Trends and Options for
Economic Empowerment
Edited version of presentations to the Ministry of Finance
and Economic Development, Addis Ababa, July 2, 2008;
and to the Sixth International Conference on the Ethiopian
Economy of the Ethiopian Economic Association, held at
the United Nations Conference Center, Addis Ababa, July
4, 2008.
Caterina Ruggeri Laderchi and Hans Lofgren, World Bank
Outline of the report


Trends in women’s wellbeing indicators and
international comparisons
Determinants of economic disparities:





Labor markets (time use, returns to human capital)
Barriers to entrepreneurship
Access to land
Initial estimates of economic implications of
addressing gender disparities in economic
empowerment (MAMS modeling)
Policy options for addressing gender disparities
Why a focus on gender disparities in
economic empowerment




Multidimensional and interlocking nature of gender
disparities
Economic empowerment – “Unleashing the potential of
Ethiopian women”– is one of the priorities in PASDEP
Economic empowerment complements and reinforces
other measures for broader empowerment, such as
removing gender bias from the legal system and granting
women effective legal access
Recent and ongoing work at the WB offers an opportunity
to take stock of challenges women face in the economic
sphere – an area which has been less explored in both
policy oriented and academic literature
Key messages






Significant achievements in reducing gender disparities in some
dimensions but serious challenges remain
Heterogeneity of women’s conditions and progress over the decade
1995-2005 raises the question of whether “general development” is
enough to reduce disparities
International experience shows that some targeted measures can
complement “general development” efforts, but an evidence based
approach is needed
Analysis of determinants of economic disparities shows:
 importance of education
 yet pervasive influence of cultural and customary factors, also in
mediating policy effectiveness  challenge of implementation
The potential implications in terms of increased efficiency and
growth of addressing gender disparity – as other disparities – are
significant
This calls for continued efforts in mainstreaming gender, assigning
responsibilities for gender targets and careful monitoring
Recent trends and international
comparisons
International comparisons (2004-2006)
Ratio of female to male primary
enrollment
160%
Births attended by skilled health staff
(% of total)
120%
Under 5 mortality rate (per 1,000 )
80%
School enrollment, tertiary, female (%
gross)
40%
Life expectancy at birth, female
(years)
0%
Primary completion rate, female (% of
relevant age group)
Proportion of seats held by women in
national parliament (%)
Low Income Countries - LICs (Indexed to 100%)
Ethiopia relative to LICs
Ratio of female to male secondary
enrollment
Total fertility rate (births per woman)
Sub-Saharan Africa relative to LICs
On most indicators Ethiopia lags behind Low Income Countries averages
Smaller gaps with respect to LICs and SSA countries in primary
education and more significant challenges for example in secondary and
tertiary education and in delivery of health services.
International comparisons (1991-97)
Ratio of female to male primary
enrollment
120%
School enrollment, tertiary,
female (% gross)
Under-5 mortality rate (per 1,000)
80%
40%
Primary completion rate, female
(% of relevant age group)
Life expectancy at birth, female
(years)
0%
Proportion of seats held by
women in national parliament (%)
Ratio of female to male
secondary enrollment
Fertility rate, total (births per
woman)
Low Income Countries - LICs (Indexed to 100%)
Sub-Saharan Africa relative to LICs
Ethiopia relative to LICs


Significant progress on primary education and political participation
The disparity between Ethiopia and SSA in terms of ratio of female to male
secondary enrolment has grown
Selected trends: education
Profile of educational attainment, by age cohort, gender
and rural/urban location (HICES 2005)
Age group: 15-19
Age group: 30-39
1.00
1.00
0.90
0.90
0.80
0.80
0.70
0.70
0.60
0.60
0.50
0.50
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0.00
1
2
3
4
Rural males
Urban males
5
6
7
Rural females
Urban females
8
1
2
3
4
Rural males
Urban males
5
6
7
Rural females
Urban females
8
Selected trends: monetary poverty
Data show no signs of women being
increasingly over-represented among the
poor (the “feminization of poverty”)
 BUT: household poverty rates are likely to
underestimate the extent of poverty and
vulnerability experienced by women
 need to understand better the
intrahousehold allocation of resources

It seems likely that more women live in poverty
than captured by household based measures

There is evidence of:


Different spheres of control over household resources and
spending decisions
Systematic gender bias in consumption by adults, that
discriminate against female children
Gender bias in the consumption of selected adult goods
(HICE 2005)
0.005
Child age<15
0.004
Child age<13
0.003
0.002
0.001
0
Personal Ser.
Cloths
Coffee
Tobacco
Women’s heterogeneity



Across the rural-urban divide
Across regions
Across groups


Income groups
Female headed households
Heterogeneity: the rural-urban divide
Primary net enrollment rate
Secondary net enrollment
rate
Listening to radio at least
once a week
180
100
160
90
70
100
60
80
50
80
60
Rural
Urban
Women
Rural
Men
0
Urban
40.4
25.7
20
10.7
0
40.1
20
10
0
40
30
6.5 11.4
75.6
34.6
40
30.3
20
50.3
78.6
120
40
100
80
62.8
140
60
120
Rural
Urban
Source: DHS, 2005; WMS, 2004
Heterogeneity: Regional disparities


Best and worst performing regions change by indicator
Increasing disparities between best and worst
performing regions in at least some indicators
% of women who agree that
a husband is justified in
beating his wife
91.3
88.4
54.4
41.7
77.3
80
74.3
70
60
60
50
50
40
40
30
30
16.2
0
2000
2005
37
23.5
20
5.6
10
78.5
80
70
20
85.1
90
90
100
90
80
70
60
50
40
30
20
10
0
% of women with at least
one daughter circumcised
% of women who support
FMG
10
0
2000
Best performing region
2005
Worst performing region
2000
2005
Heterogeneity: differences across groups


Income groups differ in terms of access to primary school and
information
Other indicators such as those of women’s empowerment show less
variations across quintiles.
Listening to radio at least once a
week
Primary net enrollment rate
60
70
50
60
40
50
30
40
20
30
20
10
10
0
0
Bottom quintile
Bottom quintile
Top quintile
Women
Men
Top quintile
Heterogeneity: differences across groups

Female-headed households face specific challenges:
 disadvantages in household composition and size
 … compounded by the gendered division of labor prevalent in
agriculture

But are in themselves a heterogeneous category: e.g. marital
status matters:
 in urban areas, FHHs with never married heads have the lowest
probability of being poor (the poverty incidence rate is only 9
percent).
 in rural areas FHH with currently married heads have the lowest
probability of being poor
 widowhood increase women’s risk of being poor

FHH, while not more likely to be poor than MHH, are generally
more vulnerable
The Determinants of Gender Economic
Disparities
Gender disparities in the labour
markets



Women have lower activity rates in GDP/market
activities, lower employment rates and higher
unemployment rates than men.
They are disproportionately concentrated in unpaid
occupations or hold insecure jobs that offer lower
earnings
Earnings and opportunities are limited by:



Greater burden of household responsibilities
Lower education
Discrimination
The unequal burden of household responsibilities
Decomposition of the average household work hours per
week, LFS 2005
100
Urb women
Rur women
80
60
Urb men
Rur men
Rur women
Urb women
40
Urb men
20
Rur men
0
Incidence (%)
•
•
•
Duration
All household activities are predominantly performed by women.
Incidence of domestic work does not vary with number of hours spent on
market activities
Gender-based division of labor is much more acute in rural areas with a
a much heavier “double burden” of work.
Education associated with less difference in time spent in market
activities between women and men
Discrimination in returns to wage labor


Data allow exploration of discrimination for a limited
segment of the labor market (wage work in Ethiopia
accounts for about ¼ of all employment -- 11 percent in
rural areas)
Decomposition of the differences in earnings between
men and women reveals that
 about 45 percent can be explained by worker
characteristics
lower investments in human capital and less experience on the job
account for 25 to 39 percent of the gap. This captures women’s greater
concentration in the informal sector and lower concentration in betterpaying formal public and private jobs.

about 40 percent can be ascribed to “pure wage
discrimination effects”.
Gender disparities in entrepreneurial
activities
Percentage of female-owned enterprises in the formal sector
Ethiopia’s
female
ownership of
firms:
60
50
40
• one of the
lowest in the
region
30
20
10
Botswana
Cape Verde
Mozambique
Cameroon
Swaziland
Namibia
Burundi
Uganda
Burkina Faso
Gambia
Angola
Madagascar
Congo, Dem. Rep.
Lesotho
Malawi
Zambia
Guinea Bissau
Mali
Mauritius
Niger
Ethiopia
South Africa
Benin
Tanzania
Senegal
Nigeria
Kenya
Eritrea
0
• increasing
trend
Source: World Bank Enterprise Surveys, 2002-2006. Data reported
in The Africa Competitiveness Report 2007.
Barriers to female entrepreneurship in large
and small scale activities

In the formal urban sector women business owners




face a more hostile environment than their male counterparts
are more educated and successful in running their businesses
than male business owners, but are more vulnerable to crime and
corruption
are more likely to run businesses in partnerships
In rural areas their likelihood to be involved in nonagricultural rural activities is



driven by “push factors” rather than “pull forces” from farm
activities
directed towards activities with low barriers to entry
resulting in lower profitability than either female-owned
enterprises in small urban areas or male-owned enterprises in
rural areas
Gender disparities in land access



Land ownership traditionally seen as specifically male:
dissolution of a marriage might result in women’s destitution
(FHH)
Males
Females
Number of owners (in millions)
9.6
2.3
Land per capita (ha)
1.12
0.71
Informal institutions and customary mediate land access and
effectiveness of specific programs– also related to heterogeneity
by region or by ethnic group
These limitations compounded by other traditional gender
roles/customs
The Economic Implications of
Addressing Disparities in Women
Economic Empowerment
General question



What are the long-run effects of stronger efforts to
foster the economic empowerment of women?
To answer this question, we “engendered” MAMS, a
CGE model for development-strategy analysis,
including MDG and education strategies.
Simulations focus on two kinds of interventions:


promoting women’s access to and returns from productive
assets
reducing costs to women of their household roles
Why use a CGE model?

The effects of efforts to promote the empowerment
of women depend on and influence the evolution of
the economy as a whole. For example, if education
is expanded, other things that matter and can be
considered in a CGE model include:




How is the expansion in education financed?
To what extent is the labor market gender blind?
To what extent is it possible to reduce time spent on “home
services”?
How rapid is GDP growth?
25
Engendering MAMS

As opposed to standard MAMS, the engendered
version of MAMS:



covers full time use (not only GDP labor time);
adds leisure and “home services” (cooking, cleaning, child
care, fetching fuel, shopping, …) to the production activities
(i.e. these are no longer limited to GDP activities);
disaggregates population in working age and their time use
by gender (not only by education) – their time use is a
“production input.”
Labor nesting in GDP production
A
g
g
r
e
g
a
te
L
e
s
s
th
a
n
c
o
m
p
le
te
d
s
e
c
o
n
d
a
r
y
M
a
le
F
e
m
a
le
C
o
m
p
le
te
d
s
e
c
o
n
d
a
r
y
M
a
le
F
e
m
a
le
C
o
m
p
le
te
d
te
r
tia
r
y
M
a
le F
e
m
a
le
27
Wage discrimination against women

Across all GDP activities, for females:



wage paid < marginal value product (MVP);
surplus (the gap) paid to male labor.
Treatment justified by need to consider:


the fact that economic benefits of increasing female
employment > financial benefits reaped by female workers;
impact of reduced discrimination (direct on earnings;
indirect on broader indicators, considering differences in
male and female spending patterns).
31
Treatment of leisure and home services

For both leisure and home services:






commodities disaggregated by gender and education;
only demanded by the household;
each commodity produced with the related labor type as input.
For each leisure type, a “subsistence” quantity demanded defined on
the basis of the total size of the related labor type; its total quantity
demanded depends on price and income effects.
For each home-service type, subsistence quantities scaled to keep
fixed total per-capita home service time; when productivity improves,
these subsistence quantities are scaled down for all labor types.
This non-neoclassical treatment is justified by the special nature of
leisure and home services:


norms important in time allocation by gender and education;
leisure produced and consumed by the same person.
32
Time use by gender (%), 2005
60
50
40
GDP
30
Home Services
Leisure
20
10
0
Male
Female
Employment shares by labor type, 2005 (%)
60
50
40
30
20
10
0
Male, <
completed
secondary
Female, <
completed
secondary
Male,
completed
secondary
Female,
completed
secondary
Male,
completed
tertiary
Female,
completed
tertiary
Determinants of GDP growth


Growth in factor employment in GDP activities; for
labor (including female labor), the higher its level of
education, the higher the MP;
Growth in the TFP of production activities, with two
components:


endogenous: depending on economic openness and
growth in government infrastructure stocks
exogenous part captures what is not explained in model
(institutions, new technologies, ….)
Simulations: period and description


Period: 2005-2030.
Description on the following table …
Description of simulations
Name
Description
business-as-usual scenario with 6% annual growth
in real GDP at factor cost
base
tax-financed expansion (increased quality) in
education after 1st primary cycle
edtx
same as edtx except for that financing is provided
by foreign grants (gains from aid financing?)
ed
ed + high male-female labor substitution elasticities
in GDP activities (wage impact of more "genderblind" hiring decisions?)
ed+el
ed+el + increased productivity growth in home
service production (wage and growth impact of
releasing women to labor market?)
ed+el+hp
ed+el+hp + increased productivity growth in private
GDP sectors (impact of higher GDP growth on the
labor market?)
ed+el+hp+pp
ed+el+hp+pp + removal of wage discrimination
ed+el+hp+pp+rd against females (extent of female wage gain?)
37
GDP at factor cost and private
consumption (% growth per year)
8
base
edtx
7
ed
ed+el
6
ed+el+hp
ed+el+hp+pp
5
GDP at factor cost
Private consumption
Gross Enrollment Rate, secondary (%)
60
50
40
Male
30
Female
20
10
0
2005
base
edtx
ed
ed+el+hp+pp
Employment growth by labor type (%)
12
hp
+p
p
ed
+e
l+
ed
0
Male, completed
tertiary
ed
tx
4
Female, completed
secondary
ba
se
8
Male, completed
secondary
Female, completed
tertiary
47
Wage growth, secondary (%)
4
Male
Female
3
2
e
ba s
ed tx
ed
el
hp
pp
+
+rd
+
+
d
l
p
p
e
e
p
h
p+
ed + d +el+
h
+
l
e
e
ed +
Wage income growth, secondary (%)
10
Male
Female
8
6
e
ba s
ed tx
ed
p
p
el
rd
ed + d +el+h l+hp+p p+pp+
e
e
h
ed + d +el+
e
Conclusions (1)

Main results:


Expanded higher education (with strong gains in
female education) accelerates GDP growth and
raises private consumption (if financed by aid);
Female wage growth is positively related to:


growth in educated labor demand (which depends on
GDP growth); and
reduced discrimination against women in wage and
employment decisions.
50
Conclusions (2)

Future work (drawing on emerging micro evidence):




improved database;
incorporate links between incomes under female control
and the allocation of spending across different types of
consumption and savings;
add female education indicators to the determinants of
health and education outcomes.
Such extensions make it possible to consider
additional channels through which female
empowerment contributes to human development.
51
References


Bourguignon, Francois, Carolina Diaz-Bonilla, and
Hans Lofgren. 2008. “Aid, service delivery and the
Millennium Development Goals in an Economywide
Framework,” pp. 283-315 in François Bourguignon,
Maurizio Bussolo, and Luiz A. Pereira da Silva, eds.
The Impact of Macroeconomic Policies on Poverty and
Income Distribution: Macro-Micro Evaluation
Techniques and Tools. Washington, D.C.: World Bank.
World Bank Report. Unleashing The Potential Of
Ethiopian Women: Trends And Options For Economic
Empowerment. June 2008 Draft. Poverty Reduction
and Economic Management 2. Africa Region.
Policy Options for Addressing Gender
Disparities in Economic Empowerment
Policy options
Menu of options based on international experience to help
translating PASDEP’s vision into concrete policy options
Broad based and targeted policies



Scope for improving wellbeing indicators of both women and men,
particularly in rural areas
Yet heterogeneity of women’s situations means that certain groups
are particularly vulnerable or face specific challenges which call for
targeted measures
Further targeted measures can help addressing strong cultural
barriers.
Evidence-based approach

Different national and international experiences can be scaled up
after piloting and evaluation
Policy options [2]

Focus on a limited set of interventions: three
main areas



Promoting gender equality in access to productive
resources – focus on human capital and its
utilization
Reducing the costs to women of their household
roles
Strengthening women’s voice and representation
NB separate work will look specifically into land issues
1. Access to productive resources

Access to human capital
 Untargeted measures: continuing ongoing efforts in improvement
in school facilities and expansion of village schools
 Targeted measures for girls’ enrolment and attendance:






secondary school scholarships for girls;
stipends for female students linked to school distance;
conditional cash transfers targeting poor households;
school-feeding and take-home rations reserved for girls.
Promoting safety in schools and gender awareness eg Girls'
Education Advisory Committees
Earning capacity and opportunities

training and employment creation programs
1. Access to productive resources [2]

Access to financial services:



micro-credit with complementary services: business literacy
training, management capacity building, product and
processing training
additional services such as saving options.
Monitoring of gender outcomes and accountability:



actions for improving awareness of women’s rights especially
among agents delivering services;
legal aid, incentives and support for access to formal courts;
clear responsibility for gender outcomes
2. Reducing the costs to women of their
household roles

Time-saving technologies and services especially in
rural areas:


Provision of child care:



improved stoves and modern cooking fuels, water and
transportation services, opportunities offered by
decentralized energy generation
investment in child care facilities
low cost solution such as wawa-wasi program in Peru
Considering women’s time use in program design
and implementation
3. Strengthening women’s voice and
representation

Examples already in major Government programs


But challenges in translating plans into practice:


Quotas for Local Committees were respected by only 20
percent of them. Women account for a small number of
representatives also in PSNP structures.
More training and sensitization:


E.g. statutory gender quotas in Local Committees for land
registration; women representation in PSNP structures
Qualitative evidences point out women’s low awareness of
their rights or fear of speaking in public
Strengthening monitoring of key aspects of women’s
participation