- Email: [email protected]
Female labour force participation and economic growth in the South Mediterranean countries
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.
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.