GENDER INEQUALITY IN THE INDONESIAN LABOUR MARKET

Transcription

1 GENDER INEQUALITY IN THE INDONESIAN LABOUR MARKET Lisa Cameron, University of Melbourne. 24 July 2018

2 OVERVIEW 1. Female labour market participation; 2. Gender wage gap; 3. Women s Labour Market Transitions. Joint work with Diana Contreras-Suarez, University of Melbourne. A collaboration with the Australia Indonesia Partnership for Economic Governance (AIPEG)

3 IMPLICATIONS OF GENDER GAPS FOR ECONOMIC GROWTH Barriers to women s employment reduce the pool of talent from which employers can draw, reducing the average ability of the workforce (Esteve-Volart, 2004) Employment and earnings increase women s bargaining power within the household. - greater investment in next generation s human capital (health and education) - increased saving rates (Seguino and Flores, 2003; Stotsky 2006) Empirical evidence: total output estimated to be 6.4%-8.7% lower (across 37 European countries) as a result of barriers to women s entry, Cuberes and Teignier (2016), effect on entrepreneurial ability. Moving from the country with the lowest value of the UN gender inequality index to the highest value is associated with a decrease of about 4 percentage points in the growth rate, Amin et al. (2015) using cross-sectional data from 107 countries.

4 1. FEMALE LFP IN INDONESIA IS LOW

5 ..AND UNCHANGING. Despite large changes in the structure of the economy Less reliance on agriculture Increasing industrialisation and urbanisation Increases in incomes And Increases in educational attainment Declines in fertility rates Female-ILO Male-ILO Female-Susenas Male-Susenas

6 DATA & METHODOLOGY SUSENAS for 1996, 2000, 2007, 2011 and 2013 Total of 2,323,615 individuals aged 15 to 64 born between 1934 and ,113,031 females, and 1,150,584 males Estimate: where: LFP i = β 0 + X 64 i β + σ j= δ j a ji + σ j=1943 η j C ji + ε i LP i =1 if the person i is participating in the labour market, 0 otherwise. Xi is a set of individual, household and village characteristics, provincial unemployment rate and province fixed effects; a ji is an age indicator variable; C ji is a year of birth (cohort) indicator variable; and ε i is a random disturbance term.

7 Participation probability KEY FINDINGS (AGE EFFECTS) The gender gap in LFP widens during child-rearing years Effect of marriage on FLFP Age Effect Age Females Males

8 Participation probability Participation probability FEMALE LFP BY EDUCATION AND AGE OF CHILD 1 Age Effects 1 Age Effects Age Age Primary Lower-secondary Upper-secondary Tertiary 0 to 2 y.o. 3 to 6 y.o. 7 to 11 y.o. 12 to 17 y.o.

9 COHORT EFFECTS Urban Areas Rural Areas Cohort Effects Cohort Effects Year of Birth Females Males Year of Birth Females Males

10 2. GENDER WAGE GAP The wage gap has been declining in the formal sector but remains large with women earning less than 80% of the male wage. 0.8 Wage workers Female/Male Ratio * * ᵜ Source: * Feridhanusetyawan et al., (2001) using Sakernas. Pirmana, (2006) usign Sakernas. Siegmann, (2003) using Susenas. ᵜ Taniguchi et al., (2014) using Sakernas. Authors' calculations using Susenas. The figures for 2011 and 2013 include wages for

11 WAGE GAP (CONT.) How does the gender wage gap vary across the wage distribution 2011 National Socioeconomic Survey (SUSENAS) in 2011 Formal and Informal Sector Formal: Employer assisted by permanent and paid workers; employees Informal: Self-employed; employer with casual and unpaid workers; casual workers; unpaid workers Individuals aged 15-64: 332,718 (Formal: 161,040; Informal: 171,678) Estimation of the wage gap using unconditional quantile estimation (Firpo, Fortin & Lemieux, 2009)

12 HOURLY WAGES (IN LOGS) BY GENDER Wage Density Susenas 2011 Formal Sector Wage Density Susenas 2011 Informal Sector ln(hourly wage) ln(hourly wage) Density Females Density Males Density Females Density Males

13 DIFFERENT CHARACTERISTICS: EDUCATION Formal Informal Male Female Male Female No school Primary Junior HS Senior HS Diploma I/II Diploma III/IV/S Years of experience Career interruptions due to children Years of experience: Age Years of education N children born 5

14 CHARACTERISTICS DIFFERENCES: INDUSTRY Formal Informal Male Female Male Female Mean Mean Mean Mean Industry: Agriculture Industry: Mine Industry: Manufacture Industry: Trade Industry: Service

15 METHODOLOGY Returns to productive characteristics (OLS) W i,g = X i,g β g + ε i,g ε i,g = 0, g = male, female W i,g is the log of the hourly wage for individual i of gender g X i,g are productive characteristics (Years of experience, educational attainment, vocational training, computer skills, health status, geographic indicators, industry type, status of employment and marital status) Blinder-Oaxaca Decomposition at mean Firpo, Fortin and Lemieux (2009) method for decomposition across the distribution.

16 1 Wage Gap Decomposition at Mean % 68% 62% 38% Formal Explained Unexplained 50.2% 75% 25% Informal

17 THE GENDER WAGE GAP PREVAILS AND IT IS LARGER FOR THOSE WITH THE LOWEST EARNINGS 0.6 Formal Sector 0.6 Informal Sector % 63% % % 0 Q10 Q20 Q30 Q40 Q50 Q60 Q70 Q80 Q90 Wage quantile Endowments Discrimination 13% 50% % 0 Q10 Q20 Q30 Q40 Q50 Q60 Q70 Q80 Q90 Wage quantile Endowments Discrimination 32%

18 EXPLAINED COMPONENT

19 KEY FINDINGS We find strong evidence of sticky floors in the formal sector Relative to the total gap, explained proportions are constant along the distribution Explained components Formal: experience explained differences in wages while education reduces the gap Informal: Industrial segregation plays an important role Career interruptions account for an important part of the gap particularly in the informal sector

20 3. WOMEN S LABOUR MARKET TRANSITIONS We look at how women s labour market activity changes as they get married and have children. Main Questions 1. How does child-rearing affect women s participation in the labour market? 2. Do women move between the formal and the informal sectors? Across industries/occupations? If so, what do these transitions look like? Key Transitions 1. Marriage 2. First child, second child,

21 DATA Indonesian Family Life Survey (IFLS) 1993, 1997, 2000, 2007 and 2014 panel data. Track individuals across time and observe how their labour force activity changes when they get married, have a child etc. N = 9075 women aged between 10 and 49. N=3781 women who we observe 1 year before and after marriage; 1 year before and after birth of first child, 3 years after the birth of first child.

22 FEMALE LABOUR FORCE PARTICIPATION DROPS WITH MARRIAGE AND FIRST CHILD Female labour force participation by age 2 nd Child Marriage 1 st Child Age Single Married 1st Child 2nd Child

23 THE SHARE OF WORKING WOMEN IN FORMAL EMPLOYMENT DECREASES WITH MARRIAGE AND EACH CHILD. FOR EXAMPLE, 88% OF THE NEVER MARRIED WORKING WOMEN AGED 24 ARE IN FORMAL EMPLOYMENT WHILE ONLY 50% OF WOMEN WITH 1 CHILD ARE Share of formal employment by age 17 pp 25 pp Single Married Age 1st Child 2nd Child

24 NOT MUCH MOVEMENT BETWEEN THE FORMAL AND INFORMAL SECTORS WITH MARRIAGE Self Employed Wage worker Unpaid Not working Working Transition Matrix 42% of the women working in the formal sector before marriage are not working 1 year after marriage Not working 80% Unpaid 60% Wage worker Self Employed Before Marriage After Marriage Self Employed Wage worker Unpaid Not working 40% 20% 0%

25 NOR BEFORE AND AFTER HAVING CHILDREN. 40% OF WOMEN WHO ARE INITIALLY IN WAGE EMPLOYMENT TRANSITION TO NOT WORKING AFTER MARRIAGE AND CHILDBIRTH. Working Transition Matrix: 3 years after There is a positive net loss from the formal sector. 16% leaving and only 8% entering 80% 60% 40% Before 1 st child 20% Self Wage Employed worker 3 years after 1st Child Unpaid Not working 0%

26 PARTICIPATION IN THE FORMAL SECTOR SHRINKS

27 LARGE REDUCTIONS IN LABOUR SUPPLY FOR MODERATELY EDUCATED WOMEN, WHILE TERTIARY EDUCATED WOMEN LARGELY CONTINUE TO WORK % Senior High-School Female labour force participation by age Age Single Married 1st Child 2nd Child Tertiary Education Age Single Married 1st Child 2nd Child

28 TERTIARY EDUCATED WOMEN MORE LIKELY TO STAY IN WAGE WORK

29 KEY FINDINGS More than 40% of women are not working one year after the birth of their first child. 8.6 million women aged between 20 and 24 are not working. This is a major loss of Indonesia s productive capital. Marriage and childbearing reduce both women s participation and formality of employment for all women except tertiary educated women. Women leave the formal sector in large numbers and very few of these women then take up self-employment. Instead they leave the labour market. Institutional change is needed to increase the retention of women in the formal sector.

30 POLICY IMPLICATIONS the growth of FLFP will continue to be slow or non-existent unless policies are developed to allow support married women to juggle family and work responsibilities. The main issue appears to be that the formal sector is not equipped to retain women once they have a family (and is possibly not even trying to retain them). Promote cultural change that encourages companies to develop policies to retain women. Family-friendly work: Flexible hours, compressed work week; Part- time work with the same benefits as full-time work; Telecommuting/working from home Job sharing Work-based child care provision, in particular for women with levels of education lower than tertiary. These measures are likely to increase firm profitability and generate economic growth as the economy better utilises and benefits from the skills of half its population.

31 THE END Thank you.