Determinants of Female Labour Force Participation in Jordan Alma - - PowerPoint PPT Presentation

Determinants of Female Labour Force Participation in Jordan

Alma Boustati UNU‐WIDER SOAS

Objective

• To place the question in its historic, economic, and cultural context
• To survey competing theories and methodologies.
• To present an integrated approach to answering the question.

Content

• Overview of Jordan & the Paradox of Female Labour Force

Participation

• Economic Theory
• Methodology
• Qualitative
• Descriptive
• Multivariate
• Conclusion and moving forward

Jordan: an Overview

• Small lower middle income country with an open economy.
• Vulnerable to oil price fluctuations but NOT resource rich.
• Agriculture is 4.2% of GDP. Urban population is 84% of the country.
• The services is 66.2% of GDP and manufacturing is apparel and

pharmaceuticals.

• The majority are Muslims and half the population is of Palestinian
• rigins.
• Syrian refugees make up 20% of the Jordanian population (recent).

Female labour force participation in Jordan is the second lowest in the world…

5 10 15 20 25 30 35 40 45

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Labor force participation rate, female (% of female population ages 15+)

Jordan Lower middle income Middle East & North Africa (excluding high income)

Source: World Development Indicators

Closer inspection shows declining female labour force participation rate and low male labour force participation rates

(age 15+)

12.8% 16.2% 16.0% 16.7% 16.6% 16.5% 15.8% 14.4% 13.1% 63.8% 66.0% 65.6% 66.3% 65.1% 64.7% 63.4% 62.0% 60.7%

2006 2007 2008 2009 2010 2011 2012 2013 2014

Female Male

Source: Employment and Unemployment Survey, DoS 2006‐2014

Women are doing better than men when it comes to education…

0.2 0.4 0.6 0.8 1 1.2 1.4

School enrollment, tertiary (gross), gender parity index (GPI)

Jordan Lower middle income Middle East & North Africa (excluding high income)

0.2 0.4 0.6 0.8 1 1.2 1.4

School enrollment, primary and secondary (gross), gender parity index (GPI)

Jordan Lower middle income Middle East & North Africa (excluding high income)

Gender equality

Source: World Development Indicators

…and there has been important gains in women’s health and a decline in fertility rates

1 2 3 4 5 6 7 8 9 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

Fertility Rates (Births per Woman)

Jordan Lower middle income Middle East & North Africa (excluding high income)

100 200 300 400 500 600 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Maternal mortality ratio (modeled estimate, per 100,000 live births)

Jordan Lower middle income Middle East & North Africa (excluding high income) Source: World Development Indicators

Economic theory predicts the following:

Long (1958) – Mincer (1962) – Becker (1965):

• Gains in productivity and increase in wages

means the opportunity cost of “household production” increases.

• Fertility rates decline &“human capital

investment” increases for women Female labour force participation rate increase

…so what happened in Jordan?

The most invoked reason is “conservative social norms” stemming form Islam or Arab culture:

• Identity Economics (Akerlof and Kranton 2000) is used by incorporating an identity term in

the utility function of a woman. (For MENA: Chamlou et al. 2011; Hayo and Caris 2013…)

• Empirically, identity is captured by subjective questions on religiosity, “traditional identity”

& gender roles.

• Policy recommendations keep calling back to more education for women.

BUT it’s a rather simplistic, one‐dimensional approach

• Endogeneity: Conservative social norms –> female labour force participation

But also female labour force participation –> Conservative social norms

• Mechanism: FLFP = f(education, conservative social norms)

But also Conservative social norms= f(education) And… Education= f(conservative social norms)

• Coefficient: Explains only 2%‐5% of variation and dwarfed by other, more mainstream factors.
• Identification: Biases in subjective questions.

Of course conservative social norms matter … but a more nuanced approach is required.

• The transformation of the “one breadwinner family” is a recent

phenomenon

• What economic factors allow it to persist?
• What economic factors allow it to changes?
• Demand‐side factors & supply‐side factors matter
• the initial conditions prior to structural change matters (Bosserup 1970)
• So does the path taken to industrialization (Goldin 1995)

=> Socio‐cultural factors are endogenous to the process of development.

Examples for MENA:

• Breadwinner Family: Oil revenues => high wages => "one‐breadwinner family” remains
• sustainable. (Karshenas 1997)
• Dutch Disease: Oil revenues => exchange rate too high => export/labour‐intensive industries

that “typically employ women” stopped growing. (Assaad 2005 – on Egypt)

Methodology

Following Ilkkaracan’s (2012) in Turkey

• Descriptive
• Economic structure, evolution & FLFPR
• The cultural context
• Bivariate
• Cross‐sectional data to look at the ”stoke” (EUS)
• Retrospective data to look at the ”flow” (JLMPS)
• Multivariate
• Using EUS: ”Classic” logit regression of female labour force participation
• Independent variables include personal and household characteristics

1960s – 1970s: Import Substitution Industrialization Strategy

Oil revenue => maintenance of high family wage Expansion of public sector => employment opportunities for educated women Foreign female labour ‐ low demand for less‐educated Jordanian women. (Moghadam 2005)

1980s – 1990s: “Economic Reform”

Oil price crash and Gulf War Curtailment of public sector => negative impact on job opportunities for educated female labour. (Assaad

2014)

“Implicit policy of discouraging female employment” (Moghadem 1998; Amawi 2000) Increase demand for foreign female labour–but why?(Frantz 2008).

2000s: Export Oriented Industrialization Strategy

 Structural adjustment period => high economic growth BUT high unemployment rates.

(to be continued)

Economic Structure and Evolution in Jordan

The cultural context: “patriarchal social contract”

• Men’s economic responsibility is explicitly stated in the law.
• “Rights of men to support their families” during ISI (Olmsted 2005)
• Social security/safety nets is contingent on employment => most women are excluded (except

through their male next‐of‐kin) (Olmsted 2005).

• Jordan ratified CEDAW (World Bank 2013) but there are still some discriminatory articles in the

labour:

• Wife to get permission from husband to work.
• Women excluded from some industries/time shifts.
• In 2010, 1 in 3 women in Jordan has ever worked
• Low FLFPR is as much a retention issue as it is a participation issue.
• It also means that “rejection of women in the public sphere” isn’t necessarily a problem.

Quantitative Analysis: Scope and Limitations

• EUS (2006‐2014) & JLMPS (2010)
• Representative household surveys
• EUS is a pooled cross‐section data.
• JLMPS is unique because of retrospective questions.
• FLFPR does NOT change in Jordan even after accounting for informality and agriculture (Flynn 1999;

Assaad 2012).

• Underestimation of the number of non‐Jordanians because these are household‐based surveys.

The low female labour force participation is a problem only for lower‐ educated women.

0% 10% 20% 30% 40% 50% 60% 70% 80%

None Primary Secondary Post‐secondary University Postgrad

Female Labour Force Participation Rate by Educational Attainment

2007 2010 2014

Source: Employment and Unemployment Survey, DoS 2006‐2014

The “retention issue” is the result of women dropping out upon marriage and not coming back

0% 2% 4% 6% 8% 10% 12% 14%

‐10 ‐8 ‐6 ‐4 ‐2 2 4 6 8 10 12 14 16 18 20

Percentage of Women Who Ever Worked Year Since Marriage

Change in Labour Force Status of Women by Year Since Marriage in LF ‐> not in LF Not in LF ‐> in LF in LF ‐> in LF women who drop out of the labour force the year they get married Very small return to the workforce after marriage.

Source: Jordan Labour Market Panel Survey 2010

The unemployment rate of women is twice that of men and the problem of discouraged workers predominantly affect women

0% 5% 10% 15% 20% 25% 30% 35% 2006 2007 2008 2009 2010 2011 2012 2013 2014

Rate of Unemployment

Men Discouraged Men Women Discouraged Women

Source: Employment and Unemployment Survey, DoS 2006‐2014

Unemployment is highest among university graduates and the situation is especially bad for women with university degrees.

0% 5% 10% 15% 20% 25% 30%

None Primary Secondary Post‐secondary University Postgrad Total

Unemployment rate in 2014

Men Women

Source: Employment and Unemployment Survey, DoS 2006‐2014

Average hours worked per week are extremely high for non‐Jordanian women, exceeding the maximum amount of HpW of work allowed by the law.

10 20 30 40 50 60 70 None Primary Secondary Post‐Secondary University Postgraduate

Average Working Hours per Week for Women by Education and Nationality

non‐Jordanians Jordanians 10 20 30 40 50 60 70 None Primary Secondary Post‐Secondary University Postgraduate

Average Working Hours per Week for Men by Education and Nationality

non‐Jordanian Jordanian

Maximum hours of work per week (48) Source: Employment and Unemployment Survey, DoS 2006‐2014

Lessons to apply to the Multivariate Analysis

• Disaggregating the analysis by educational attainment.
• Accounting for family structure beyond married and single.
• Analyzing the situation of female foreign workers in Jordan.
• Acknowledging the role of demand‐side factors and opportunity structures.

Determinant of women’s labour force participation: EUS logit regression results (marginal effects)

Controls include: age, age squared, urban/rural, whether the woman is the head of her household, and the educational attainment of the household head

VARIABLES 2011 2014 Married ‐0.0793*** ‐0.0725*** (0.00309) (0.00304) No education ‐0.00138 0.00504 (0.00505) (0.00535) Secondary 0.0173*** 0.00709* (0.00358) (0.00405) Post‐secondary 0.168*** 0.162*** (0.00295) (0.00302) University or higher 0.255*** 0.233*** (0.00238) (0.00233) Presence of Children under 4 years old ‐0.0299*** ‐0.0235*** (0.00271) (0.00268) Presence of Children between 5 & 11 ‐0.0233*** ‐0.0187*** (0.00229) (0.00225) Household head is unemployed 0.0159** ‐0.00328 (0.00625) (0.00721) Non‐ Jordanian 0.133*** 0.0779*** (0.00452) (0.00415)

Education is the most important determinant of participation Marriage is a more important deterrent than the presence of children

None Primary Secondary University VARIABLES 2011 2014 2011 2014 2011 2014 2011 2014 Married ‐0.00210 ‐0.00181 ‐0.0509*** ‐0.0277*** ‐0.0540*** ‐0.0434*** ‐0.189*** ‐0.272*** (0.00368) (0.00338) (0.00329) (0.00283) (0.00667) (0.00634) (0.0148) (0.0148) Presence of Children under 4 years old ‐0.0195*** ‐0.0167*** ‐0.0238*** ‐0.0257*** ‐0.0190*** 0.00537 ‐0.0478*** ‐0.0262* (0.00392) (0.00403) (0.00305) (0.00290) (0.00591) (0.00517) (0.0132) (0.0134) Presence of Children between 5 & 11 ‐0.00404 ‐0.0130*** ‐0.0220*** ‐0.0301*** ‐0.0407*** ‐0.0102** 0.0203* 0.0376*** (0.00334) (0.00321) (0.00262) (0.00242) (0.00505) (0.00457) (0.0112) (0.0112) Household head is unemployed 0.0159** ‐0.00335 0.00770 ‐0.00805 0.0276** ‐0.0146 0.0390 0.133*** (0.00799) (0.00862) (0.00630) (0.00661) (0.0116) (0.0159) (0.0349) (0.0423) Non‐ Jordanian 0.0582*** 0.0385*** 0.105*** 0.0552*** 0.0474*** 0.00142 ‐0.318*** ‐0.340*** (0.00335) (0.00312) (0.00352) (0.00276) (0.0103) (0.0104) (0.0348) (0.0364)

Determinant of women’s labour force participation: EUS logit regression results (marginal effects)

Marriage has a stronger negative effect for educated women and its increasing

low‐paid foreign female workers are still present There is an added‐worker effect for educated women

Controls include: age, age squared, urban/rural, whether the woman is the head of her household, and the educational attainment of the household head

Conclusions

• Education is the most important determinant of whether a woman will participate. Possible mechanisms include
• Substitution effect.
• Access to work opportunities.
• “Reservation working conditions”. (Assaad 2015)
• Marriage is a much more important deterrent of labour force participation than children
• “Patriarchal contract” matters more than “burden of care”.
• Educated women do NOT necessarily change conservative perception towards gender roles.
• Being non‐Jordanian positively affects participation at the lower educational levels
• The presence of low‐paid foreign female workers is still an important factor (though not completely captured by the survey)
• Foreign female workers may be working under conditions that are not accepted by Jordanian women.
• Or they are preferred by the employer (lower minimum wage and no max hours)
• The “added worker effect” is much larger for educated women either because
• They’re more likely to find employment opportunities
• Or they earn a higher wage.