Economics Letters 120 (2013) 323–328
Contents lists available at SciVerse ScienceDirect
Economics Letters
journal homepage: www.elsevier.com/locate/ecolet
Female labour force participation and economic growth in the South
Mediterranean countries✩
Stella Tsani ∗ , Leonidas Paroussos, Costas Fragiadakis, Ioannis Charalambidis,
Pantelis Capros
E3M Lab, National Technical University of Athens, Department of Electrical and Computer Engineering 9, Iroon Politechniou Str., 15 773, Zografou
Campus, Athens, Greece
highlights
•
•
•
•
We study female labour participation and economic growth in the South Mediterranean.
A two-step approach of econometric and general equilibrium modelling is employed.
Econometric estimations confirm the U-shaped female labour force function.
Higher female labour force participation rates have a positive impact on growth.
article
info
Article history:
Received 23 November 2012
Received in revised form
18 April 2013
Accepted 28 April 2013
Available online 6 May 2013
JEL classification:
C1
J1
abstract
We investigate the relationship between female labour force participation and economic growth in the
South Mediterranean countries with a two-step methodology of econometric exercise and general equilibrium modelling. Econometric estimations on female labour participation confirm the U-shaped function and the presence of region-specific barriers. Estimations have been employed in a satellite manner
to a general equilibrium model for the simulation of (i) changes in female labour force participation as a
result of income developments and (ii) lowering of region-specific barriers to female labour force participation. The results suggest that while the first may lead to marginally lower economic growth, the second
may have a considerable positive impact on growth.
© 2013 Elsevier B.V. All rights reserved.
Keywords:
Female labour force participation rate
General equilibrium
South Mediterranean countries
1. Introduction
Female labour participation rates in the South Mediterranean1
countries remain low as compared to other developing and developed countries. Civil uprisings in the region are calling for social
and economic reforms and a boost of employment opportunities.
It is still an open question how the regions’ social and economic
future will evolve; however, developments might present an
opportunity to enhance women’s economic, social and political
✩ This paper was produced in the context of the MEDPRO project funded under
the Socio-economic Sciences & Humanities Programme of DG Research of the
European Commission’s 7th Framework Research Programme.
∗ Corresponding author. Tel.: +30 694 69 49 378; fax: +30 210 772 3630.
E-mail addresses: [email protected], [email protected] (S. Tsani).
1 The South Mediterranean countries studied in this paper are Algeria, Egypt,
Israel, Jordan, Lebanon, Libya, Morocco, Palestine, Syria, Tunisia and Turkey.
0165-1765/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.econlet.2013.04.043
inclusion. This paper assesses the impact of female labour force
participation changes on the economic growth of the South
Mediterranean countries. For this purpose a two-stage approach
of econometric and general equilibrium modelling is adopted.
At the 1st stage an econometric model of female labour force
participation is used. The derived estimations are employed in a
satellite manner to the 2nd stage where a computable general
equilibrium model, the GEM-E3-MEDPRO model,2 is employed for
the simulation of two alternative scenarios of female labour force
participation in the region. The first assumes lower participation
rates associated with income developments in the region projected
2 The GEM-E3-MEDPRO model is a version of the GEM-E3 model. E3MLab
and other contributors have extended the GEM-E3 in various directions, including development of model versions suitable for analysing growth and
structural policies. The model has been extensively used in a series of studies
completed for the European Commission and in several research projects. See:
http://ipts.jrc.ec.europa.eu/activities/energy-and-transport/gem-e3/.
324
S. Tsani et al. / Economics Letters 120 (2013) 323–328
Table 1
Reference scenario GDP and female labour force.
Source: Authors’ estimations.
Algeria
Egypt
Israel
Jordan
Lebanon
Libya
Morocco
Syria
Tunisia
Turkey
Palestine
GDP annual growth rate, in % (2010–2030)
GDP in 2030, in bn US $
GDP per capita in 2030, in US $
Female labour force in 2030, in % of total
labour force
3.03
4.69
4.03
4.64
3.09
4.13
4.05
3.97
4.78
4.09
4.23
286.5
379.1
412.1
49.9
52.1
163.4
186.9
108.5
100.5
1811.2
13.3
6,363
3,459
41,144
5,764
10,888
20,074
4,880
3,826
8,039
20,394
1,435
15.83
23.73
46.60
17.47
25.20
27.84
27.42
16.06
26.65
25.91
17.88
to occur in 2015–2030. The second simulates the effects of the removal of region-specific barriers to female labour force participation. The rest of the paper is organized as follows. Section 2 reviews
the employed methods and data. Section 3 discusses the findings.
The last section concludes with some policy considerations.
2. Methods and data
The literature offers a rich and well-established discussion
on the economic factors and personal characteristics affecting
female labour force participation.3 An equally rich set of work
documents the U-shaped relationship between female labour force
participation rates and economic growth.4 The existing literature
on the determinants and the characteristics of female labour
force participation has been employed for the development of the
econometric model summarized as follows:
FLPRi,t = b0 + b1 LGDP i,t + b2 LGDP 2i,t
+
k−1

bn Xn,i,t + bk MED11i,t + ei,t
(1)
n =1
where FLPRi,t is female labour force participation rate, LGDP i,t is the
log of the real GDP per capita, LGDP 2i,t its square, Xn,i,t is a set of n
variables controlling for education, fertility, urbanization, religious
norms and unemployment rates. MED11i,t is a dummy variable
controlling for the South Mediterranean countries. ei,t is the error
term capturing all other omitted factors, measurement errors and
possible misspecifications. Pooled time series cross-section data
are employed for 160 countries for the period 1960–2008. The
econometric model has been used so as to estimate the effects of
income changes and region-specific differentials on female labour
force participation. The quantification of the latter is important in
order to exogenously estimate the female labour force entering
the general equilibrium model for the simulation of alternative
scenarios.
The estimated coefficients on the relationship between income
and female labour force participation and on the dummy variable
controlling for the South Mediterranean countries, have been
3 Education, fertility rate, urbanization, unemployment and economic growth
have been suggested as factors determining female labour participation (see Kottis,
1990, Mishra and Russell, 2010 and references therein). With regard to the South
Mediterranean countries, an increasing volume of work has been pointing at
the importance of the interpretation of religious and cultural aspects and their
manifestation through different regulations, social and family norms regarding
women’s rights and opportunities (Youssef, 1978; Clark, 1991; Moghadam,
2004a,b). Persistence of the patriarchal family unit has also been argued to
have constituted a significant customary constraint on women’s mobility and
employment (Kandiyoti, 1988; Moghadam, 1993).
4 See Tam (2011) and references therein for instance.
employed in a 2nd step for the simulation of the alternative
scenarios on female labour force participation rates in the region.
The coefficients estimated on the dummy variable are assumed to
capture region-specific cultural and social norms, not captured by
other covariates, which may be manifested in family structures and
legal codes and constrain female labour force participation.
For the scenarios’ simulation GEM-E3-MEDPRO has been used.
GEM-E3-MEDPRO is a computable general equilibrium multicountry model, treating each country separately and linking them
through an endogenous trade of goods and services. The model includes multiple industrial sectors and economic agents. The version employed in the present work considers 19 countries/regions
and 23 economic activities5 and it is solved for the time period
2015–2030 following 5-year time steps. The model is calibrated on
the GTAP v.8 database (2007 base year).
In general equilibrium models it is the deviations from the reference scenario, which are the key for the evaluation of structural changes. The reference scenario relates to the development
of an economic outlook for each region and sector of the GEM-E3MEDPRO model. The present paper has made use of the reference
scenario developed by Paroussos et al. (2012). Table 1 summarizes
the growth assumptions and the GDP projections of the reference
scenario.6 In this scenario a total pool of the labour force is assumed
without making any assumptions on the male and female shares
in it. The present paper extends the work of Paroussos et al. (2012)
by making appropriate assumptions on female and male shares of
labour force and on the female labour force participation rates in
the reference scenario.
For this purpose total, male and female active population data
extracted from ILO, and active population growth rates provided by
Groenewold et al. (2012) have been used. ILO provides projections
up to 2020. For the period 2020–2030 it is assumed that trends
recorded over the period 2007–2020 continue to prevail. Data on
the female share to the labour force have been extracted from The
World Bank databank (2011 edition). The female share in the total
labour force has been assumed to stay close to the current levels
and not change significantly up to 2030 (Table 1).
Two alternative cases have been simulated and compared to the
reference scenario. In both cases the econometric estimations have
been employed so as to estimate the exogenously set female labour
force. The 1st alternative assumes that growth in per capita income
levels estimated in the reference scenario will be associated with
5 GEM-E3-MEDPRO regional and sectoral aggregation are summarized in the
Appendix, Tables A.1 and A.2 respectively.
6 The reference scenario assumes a continuation of current policies in the region
up to 2030. The development of the reference scenario has been based on the
growth projections for the South Mediterranean counties provided by Coutinho
(2012). Growth projections extend to 2030 and take into account the effect of the
Arab spring on the long-term growth prospects of the region.
S. Tsani et al. / Economics Letters 120 (2013) 323–328
325
Table 2
Econometric estimations. Dependent variable: female labour force participation rate (flpr).
lgdpcap
lgdpcap2
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
−39.92***
−13.52**
−42.67***
−86.60***
−41.68***
−84.39***
−73.24***
−48.30***
−31.99**
−77.97***
−35.81***
−33.890***
(4.876)
(6.363)
(5.173)
(9.292)
(8.566)
(9.576)
(4.476)
(4.666)
(4.636)
(9.323)
(9.348)
(9.361)
6.602***
(0.696)
2.891***
(0.883)
6.879***
(0.717)
12.32***
(1.246)
5.166***
(1.148)
12.01***
(1.288)
9.843***
(0.624)
7.398***
(0.663)
5.324***
(0.663)
10.16***
(1.221)
4.336***
(1.234)
4.117***
(1.234)
−0.161***
unempl
−0.131*
(0.059)
0.0302
(0.019)
urban
0.312***
(0.043)
primary_net
(0.094)
0.0402*
(0.024)
−0.104***
−0.0964***
(0.030)
(0.030)
0.395***
(0.044)
0.368***
(0.046)
0.133**
(0.051)
0.304***
(0.044)
tertiary_net
fertility
−5.262***
−4.647***
(0.208)
(0.463)
1.221
(0.803)
0.922
(0.815)
−0.152***
−0.116***
−0.102***
−0.085***
(0.009)
(0.013)
(0.015)
(0.017)
muslim80
−15.35***
med11
Observations
Adjusted
R-sq
−0.0386
(0.093)
0.329***
(0.028)
secondary_net
Constant
−0.0625
(0.070)
−5.111*
(1.068)
87.22***
(8.351)
1687
0.139
45.07***
(11.16)
1081
0.174
91.60***
(8.793)
1687
0.140
152.7***
(14.61)
712
0.195
93.73***
(15.17)
336
0.343
149.2***
(15.02)
634
0.190
179.5***
(8.191)
1500
0.382
108.8***
(8.080)
1591
0.257
76.77**
(7.917)
1687
0.234
(2.678)
180.3***
(15.91)
522
0.426
84.69***
(18.12)
307
0.460
82.64***
(18.07)
307
0.476
Standard errors in parentheses.
For variables’ sources and definitions see Appendix, Table A.3. Alternative regressions are run so as to test the U-shaped hypothesis and ensure robustness of findings.
*
p < 0.1.
**
p < 0.05.
***
p < 0.01.
lower female labour participation rates. This assumes that over the
period 2015–2030 women may not be able to compete with men in
the industrial and service sectors as employer preferences or cultural and social norms might hold them back from entering the
labour market. For this the econometric estimations on the relationship between income levels and female labour participation
rates are employed.
In the 2nd alternative it is assumed that region-specific barriers to the entry of females into the labour force seize to exist. This
simulates a state of social and economic development in the South
Mediterranean countries where cultural factors and family norms
regarding female participation in the labour force are lowered or
removed.
3. Findings
The econometric estimations confirm the arguments developed
in the literature to date on the determinants and the characteristics (U-shape relationship with economic development) of female labour force participation rates (Table 2). When controlling
for region-specific characteristics that may not be adequately captured by the model (med11 dummy) the estimations confirm that
the South Mediterranean countries record relatively lower female
labour participation rates when compared to the rest of the developed and developing countries included in the dataset.
The econometric estimations have been employed in the 2nd
step so as to exogenously set the female labour participation rates
in the alternative scenarios. In the first case female labour force
participation rates are assumed to be lower than in the reference
scenario. For this the estimated coefficients on the relationship
between per capita income and female labour force participation
rates (Table 2, column 12) have been employed.7 Thus a smaller
labour force pool applies to this scenario. Table 3 summarizes
the differences from the reference scenario. In the second case
female labour force participation rates are higher for all South
Mediterranean countries as compared to the reference scenario.
The econometric results suggest that removal of region-specific
barriers can lead to higher female labour force participation rates
by 5% (see Table 2, column 12, med11 variable coefficient). Hence
in the alternative case an equal increase is assumed to take place
in the South Mediterranean countries over 2015–2030.8 Table 4
summarizes the changes from the reference scenario assumed in
this case.
The simulation results indicate that the effects of lower female
labour participation rates may be small (Fig. 1). This is associated
7 We employ the estimation results derived from the extended form of the model
in an attempt to address omitted variable bias.
8 For Israel it is assumed that female labour force participation rate increases by
1% above the reference scenario as Israel records already relatively higher female
labour participation rates.
326
S. Tsani et al. / Economics Letters 120 (2013) 323–328
Table 3
Female labour force and total labour force when lower female labour force participation rates are assumed.
Source: Authors’ estimations.
Algeria
Egypt
Israel
Jordan
Lebanon
Libya
Morocco
Syria
Tunisia
Turkey
Palestine
Female labour force change from reference, in %,
cumulatively over 2015–2030
Total labour force, change from reference, in
million, cumulatively over 2015–2030
Total labour force, change from reference, in %,
cumulatively over 2015–2030
−2.76
−1.60
−1.30
−3.01
−1.51
−1.33
−1.61
−1.80
−2.33
−1.91
−0.84
−0.89
−2.04
−0.37
−0.18
−0.10
−0.16
−0.92
−0.35
−0.43
−2.41
−0.04
−0.44
−0.38
−0.61
−0.53
−0.38
−0.37
−0.44
−0.29
−0.62
−0.49
−0.15
Table 4
Female labour force and total labour force when higher female labour force participation rates are assumed.
Source: Authors’ estimations.
Algeria
Egypt
Israel
Jordan
Lebanon
Libya
Morocco
Syria
Tunisia
Turkey
Palestine
Female labour force change from reference, in %,
cumulatively over 2015–2030
Total labour force, change from reference, in
million, cumulatively over 2015–2030
Total labour force, change from reference, in %,
cumulatively over 2015–2030
22.69
8.09
1.39
16.58
9.74
7.14
8.40
13.67
9.33
8.95
9.73
7.34
10.34
0.40
0.98
0.66
0.86
4.78
2.66
1.72
11.31
0.41
3.59
1.92
0.65
2.90
2.45
1.99
2.30
2.20
2.49
2.32
1.74
Fig. 1. Impact on wages and GDP of lower female labour force participation rates,
change from reference scenario, in %, cumulatively over 2015–2030.
Source: GEM-E3-MEDPRO.
with the small changes in labour force occurring from changes in
female labour force participation rates as well as from the fact that
female labour participation is relatively low in most of the South
Mediterranean countries; hence any changes of a small magnitude
would not have any significant impact on macroeconomic variables.9 Lower female participation in the labour force constrains
labour supply. As a result wages increase. Increase in wages increases the cost of production. Exports of the South Mediterranean
countries become less competitive while imports reduce as domestic production contracts (Fig. 2). The effects on GDP are more
pronounced in countries like Tunisia and Morocco in which labour
intensive sectors, like textiles play an important part.
9 For Israel the effects are of greater magnitude as women are an equally
important share to the labour force as men and changes in their participation would
have a larger impact.
Fig. 2. Impact on trade of lower female labour force participation rates, change
from reference scenario, in %, cumulatively over 2015–2030.
Source: GEM-E3-MEDPRO.
In the 2nd alternative increased female labour force participation increases labour supply, thus the cost of labour and wages
fall (Fig. 3). This drives production costs down and makes exports
of the South Mediterranean countries more competitive in the international markets (Fig. 4). Reduced prices, resulting from lower
labour costs increase private consumption. Higher consumption
and investments push GDP to grow. Lebanon, Tunisia and Morocco
record the largest change in GDP cumulatively over 2015–2030
from the reference scenario. Oil-producing countries (Algeria and
Libya) record a relatively higher drop in real wages; nevertheless,
the impact on economic growth is limited as these economies are
dominated by capital intensive sectors.
The effects of higher female labour participation rates in the
South Mediterranean counties are not uniform across the region.
S. Tsani et al. / Economics Letters 120 (2013) 323–328
327
Table A.2
GEM-E3-MEDPRO sectoral aggregation.
Fig. 3. Impact on wages and GDP of higher female labour participation rates,
change from reference scenario, in %, cumulatively over 2015–2030.
Source: GEM-E3-MEDPRO.
No
Sector
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Agriculture
Animal products
Coal
Crude oil
Oil refining
Natural gas extraction
Gas distribution
Transmission and distribution of electricity
Water
Chemical products
Other energy intensive
Electric goods-other equipment goods
Transport equipment
Consumer goods industries-food
Consumer goods industries-rest
Textiles and clothing
Construction
Transport
Communication
Business-financial services
Public services
Recreational and other services
Dwellings
Nevertheless all countries benefit from an increase in their labour
force resulting from higher female participation. Countries where
labour intensive sectors are important and countries with significant export volumes like Tunisia, Turkey, Lebanon and Morocco,
benefit the most in terms of trade and GDP.
Fig. 4. Impact on trade and investments of higher female labour participation rates,
change from reference scenario, in %, cumulatively over 2015–2030.
Source: GEM-E3-MEDPRO.
Table A.1
GEM-E3-MEDPRO regional aggregation.
No
Country/region
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Algeria
Egypt
Israel
Jordan
Lebanon
Libya
Morocco
Palestine
Syria
Tunisia
Turkey
EU10 countriesa
New EU member statesb
South EU member statesc
Emerging Asian economiesd
Rest of OECD countriese
Rest of emerging economiesf
Rest of Middle Eastg (the Gulf region)
Rest of the world
a
Austria, Belgium, Denmark, Finland, Germany, Ireland, Luxembourg, Netherlands, Sweden and United Kingdom.
b
Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia,
Bulgaria, Romania.
c
Cyprus, France, Greece, Italy, Malta, Portugal and Spain.
d
China, Hong Kong, Taiwan, Indonesia, Malaysia, Philippines, Singapore,
Thailand, Vietnam and India.
e
Australia, New Zealand, Rest of Oceania, Japan, Korea Republic, Canada, United
States of America, Rest of North America, Switzerland, Norway and Rest of EFTA.
f
Mexico, Argentina, Brazil, Chile, Venezuela, Albania, Belarus, Croatia, Russian
Federation, Ukraine, Rest of Eastern Europe, Rest of Europe, Kazakhstan, Kyrgyzstan,
Rest of Former Soviet Union, Azerbaijan and Georgia.
g
Armenia, Bahrain, Iran Islamic Republic, Kuwait, Oman, Qatar, Saudi Arabia,
United Arab Emirates, Yemen and Iraq.
4. Conclusions and policy implications
This paper has studied the relationship between female labour
force participation and economic growth in the South Mediterranean countries. The econometric results confirmed the arguments on the U-shaped relationship between economic growth
and female labour force participation rates. With regard to the
South Mediterranean countries the estimation results suggested
that characteristics specific to the latter countries may play part in
explaining the low levels of female labour force participation in the
region. The simulation results of GEM-E3-MEDPRO suggested that
lower female labour participation rates may have marginal effects
on the macroeconomic outcomes in the region. In contrast if barriers to the entry of females in the labour force are removed and their
participation increases, the economic benefit might be significant.
The importance of the findings lies on their policy implications
suggesting that policies aiming at the removal of region-specific
barriers to female labour force participation might have a growth
enhancing effect. Policies in the region should promote further female labour force participation as this will benefit growth. Promotion of higher female labour force participation rates could be
achieved with the adoption of policies which can remove or lower
barriers to female labour participation. Policies should opt for alternatives and cooperation strategies which will promote the modernization of cultural norms and legal codes. These could include:
reform of legislation, change of curriculum in the education system, support and encouragement of female school enrolment, and
cultural exchange and cooperation programmes with developed
countries.
Appendix
See Tables A.1–A.3.
328
S. Tsani et al. / Economics Letters 120 (2013) 323–328
Table A.3
Variables: sources and definition.
Variable
Definition
Source
flpr
Female labour participation rate, defined as the number of female labour participants of age 15–64
divided by the total female population of the same age group
Log of GDP per capita. Authors’ estimations based on GDP data (constant 2000 US$) and total
population figures
Unemployment, total (in percent of total labour force)
Urban population (in percent of total)
School enrolment, primary, female (in percent of net)
School enrolment, secondary, female (in percent of net)
School enrolment, tertiary, female (in percent of net)
Fertility rate, total (births per woman)
Muslims, in percent of total population in 1980
Dummy variable taking the value of 1 if the country belongs to the South Mediterranean countries
group, 0 otherwise
International Labour Organization
lgdpcap
unempl
urban
primary_net
secondary_net
tertiary_net
fertility
muslim80
med11
References
Clark, R., 1991. Culture, gender, and labor force participation: a cross-national study.
Gender & Society 5 (1), 47–66.
Coutinho, L., 2012. Determinants of growth and Inflation in Southern Mediterranean
countries. MEDPRO Technical Report No. 10. The Centre for European Policy
Studies. http://www.ceps.eu/ceps/dld/6736/pdf.
Groenewold, G., de Beer, J., Huisman, C., 2012. Population scenarios for South
Mediterranean Countries 2010–2050. MEDPRO Report No. 2. The Centre for
European Policy Studies. http://www.ceps.eu/ceps/download/7226.
Kandiyoti, D., 1988. Bargaining with patriarchy. Gender & Society 2, 274–290.
Kottis, A.P., 1990. Shifts over time and regional variation in women’s labor force
participation rates in a developing economy. Journal of Development Economics
33, 117–132.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1999. The quality of
government. Journal of Law, Economics, and Organization 15 (1), 222–279.
Oxford University Press.
Mishra, V., Russell, S., 2010. Female labor force participation and total fertility rates
in the OECD: new evidence from panel cointegration and Granger causality
testing. Journal of Economics and Business 62, 48–62.
World Bank databank (2011)
s.a.a
s.a.a
s.a.a
s.a.a
s.a.a
s.a.a
La Porta et al. (1999)
Authors’ compilation
Moghadam, V., 1993. Modernizing Women: Gender and Social Change in the Middle
East. Lynne Rienner Publishers, Boulder, CO.
Moghadam, V., 2004a. Women’s economic participation in the Middle East: what
difference has the neoliberal policy turn made? Journal of Middle East Women’s
Studies 1 (1), 110–146.
Moghadam, V., 2004b. Patriarchy in transition: women and the changing
family in the Middle East. Journal of Comparative Family Studies 35 (2),
137–163.
Paroussos, L., Fragiadakis, K., Charalambidis, I., Tsani, S., Capros, P., 2012.
Quantitative macroeconomic scenarios for the MED11 region using the GEME3-MEDPRO general equilibrium model. E3M-Lab, Institute of Communication
and Computer Systems, National Technical University of Athens.
Tam, A., 2011. U-shaped female labor participation with economic development:
some panel data evidence. Economics Letters 110, 140–142.
The World Bank DataBank (2011 edition): http://databank.worldbank.org/ddp/
home.do.
Youssef, N., 1978. The status and fertility patterns of Muslim women. In: Beck, L.,
Keddie, N.R. (Eds.), Women in the Muslim World. Harvard University Press,
Cambridge, MA.