GENDER AND RACE IN THE LABOR MARKET
Sponsored by a Grant TÁMOP-4.1.2-08/2/A/KMR-2009-0041 Course Material Developed by Department of Economics,
Faculty of Social Sciences, Eötvös Loránd University Budapest (ELTE) Department of Economics, Eötvös Loránd University Budapest
Institute of Economics, Hungarian Academy of Sciences Balassi Kiadó, Budapest
2
Author: Anna Lovász Supervised by Anna Lovász
June 2011
Week 9
Women in the labor market Literature for next week
• Roma in Hungary
• Kertesi–Kézdi 2010 (BWP): Roma Employment in Hungary After the Post- Communist Transition
Women’s labor market situation: trends
• The employment situation of women has improved significantly in the last decades:
the female-male difference decreased.
• The employment of women decreased in many transitional countries, especially for low-skilled women.
• The positive international tendency has stopped: since 2000 women’s employment has fallen in many countries.
• The female-male wage gap has decreased to about half in the last decades, and
3 leveled off.
• After the transition the wage gap fell in Eastern European countries, and increased in former Soviet countries.
Female employment, 2000–2007, EU
4
Female employment and unemployment, EU
5
USA – gender wage gap
(Institute for women’s policy research)
6
Gender pay gap, EU
Country Gender Pay Gap Country Gender Pay Gap
Belgium 9.0 Luxembourg 12.4
Bulgaria 13.6 Hungary 17.5
Czech Republic 26.2 Malta 9.2
Denmark 17.1 Netherlands 19.6
Germany 23.2 Austria 25.5
Estonia 30.3 Poland 9.8
Ireland 17.1 Portugal 9.2
Greece 22 Roumania 9.0
Spain 17.1 Slovenia 8.5
France 19.2 Slovakia 20.9
Italy 4.9 Finland 20.0
Cyprus 21.6 Sweden 17.1
Latvia 13.4 United Kingdom 21.4
Lithuania 21.6
What are the main factors contributing to the gap and its changes?
7
Women – other measures: business
Women in leadership
2008 Women Men Gap
Sex distribution of leaders of businesses, in 2008 (6)
28.6 71.4 42.8
2009 Women Men Gap
Members of single/lower houses of national parliaments in EU Member States (MS)
11 89 78
Eurostat
8
Differences and trends – causes
Female roles
Significant differences remain in the division of labor
2005 Women Men Gap
Domestic and family work
Paid employment
Domestic and family work
Paid employment
Domestic and family work
Paid employment Average time spent
per week in domestic and family
work and in paid employment
26.3 43.1 10.0 49.6 16.2 –6.5
2008 Women Men Women Men
without children with children without children with children Difference Difference Employment rate of
women and men (aged 25–49) with or without children
80.5 52.2 81.4 85.7 –28.3 4.2
9
2008 Women Men Gap
Educational attainment (at least upper secondary school) of women and men aged 20–24
85.5 81.7 –3.8
• Education: women’s average schooling has improved.
• Women are the majority in higher education
10
• Occupational and industry-level segregation has decreased, but still significant
• Female jobs pay lower wages
2007 Women Men Gap
Male- dominated sectors*
Female- dominated sectors**
Male- dominated sectors*
Female- dominated sectors**
Male- dominated sectors*
Female- dominated sectors**
Share of employed persons in male or female dominated sectors
24.5 38.5 49.7 16.2 25.2 –22.3
Male- dominated occupations
***
Female- dominated occupations ****
Male- dominated occupations ***
Female- dominated occupations
****
Male- dominated occupations
***
Female- dominated occupations
****
Share of employed persons in male or female dominated
18.3 63.3 54.4 29.9 36.1 –33.3
Source: Eurostat
11
Flexible work forms – Flexicurity
• Flexicurity: policies that promote a flexible and secure labor market
• Flexible work forms:
• Part time work/shared work
• Distance work/work from home
• Flexible work hours
• Dangers:
• Full time jobs may be turned into part time jobs – this is not the goal, but rather expansion of jobs.
• These jobs are less secure – increases the poverty risk of women.
• Steps to aid the situation of women:
• Expansion and increased flexibility of childcare provision
12
• Provision of adult training during maternity leave
• Decreasing the administrative and tax burdens of flexible employment for employers
• Ensuring the availability of information regarding employment forms, job search databases
Part time work, EU
Differences and trends – causes:
preferences
• Women’s decisions regarding employment may differ:
• Balancing work/family obligations: gender differences lead to different
13 choices of time use.
• – Empirical results: time spent on housework impacts wages negatively (compensating wage differentials for flexible jobs)
• Different expectations and labor market ties lead to differing human capital investment decisions.
It is possible that labor market differences are due to women’s decisions.
Hersch (2006): even after controlling for all things related to work/family decisions, an unexplained gender wage gap remains
Differences and trends – causes:
discrimination?
• Hersch (2006): summary of studies analyzing the unexplained wage gap:
• Remains even after controlling for detailed worker characteristics
• Occupational segregation does not explain most of the gap
• Remains after taking time spent at a workplace and the frequency of exits into account
• Studies on the effect of family status and time spent on housework found that these do not explain the wage gap
• Compensating wage differentials also do not explain it
• Research on schooling decisions find that these also do not lead to the wage gap
• Studies measuring workers’ true productivity do not find an explanation for the gap
Overall, we cannot exclude discrimination as a major cause of the gender wage and employment gaps
14
Effects of psychological traits – previous results
• Labor market effects of psychological traits:
• Friendly/talkative people choose occupations in which they are in contact with others (Borghans et al 2008, Krueger&Schadke 2008)
Lately technological advances and the prevalence of teamwork have benefitted workers with such characteristics.
Individual psychological traits influence employment decisions (Filer 1988).
• Wage effects of psychological traits:
• Leadership skills influence wages and promotions positively, even after controlling for cognitive abilities (Kuhn&Weinberger 2005)
• For young people, pessimism and low self-esteem influence schooling decisions negatively, and later labor market outcomes as well (Waddel 2006)
• Locus of control beliefs (destiny) affect expectations regarding wages, and schooling decisions (Coleman&DeLeire 2003)
Psychological traits affect labor market outcomes through productivity and preferences (Braakmann 2009)
Effects of psychological traits by gender – previous results
• Gender differences in psychological traits:
• Experiments regarding altruism: men donate more if the cost is low, women donate more if the cost is high (Andreoni&Vesterlund 2001)
• Women perform worse in competitive settings, while men perform the same
15 in any situation (Gneezy et al 2003)
• Overconfidence: more typical of men based on stock market data (Barber&Odean 2001)
• Gender differences in the labor market effects of psychological traits:
• The market values the traits of men and women differently, and this explains 7-16% of the wage gap (Mueller&Plug 2006)
• Differences in self-confidence and views on the importance of family/work affect the wage gap as well (Fortin 2008)
• Non-cognitive skills have different returns by gender (Heineck 2007)
Effects of psychological traits – Braakmann 2009
• Estimation of the effect on wages and employment
• Methodology:
• German SOEP database 2005, individual and labor market characteristics
• Psychological traits: based on 3 questions, averaged on a 7 point scale:
• Openness: curiosity, creativity, appreciation of new ideas
• Conscientiousness: self-discipline, sense of duty, preference for planned events vs. unexpected
• Extraversion: social behavior, engagement
• Agreeableness: ability to get along with others, cooperation
• Neuroticism: emotional instability, stress, sorrow, anger
• External locus of control: life is governed by fate not me
• Reciprocity: willingness to return favorable/hostile acts
• Risk aversion – experimentally validated measure
• Problems:
• Causality: current labor market status may influence traits
• Measurement error, lack of structural model: not evidence, just trends
16
Questions for measuring psychological traits
17
Differences by gender (Braakmann 2009)
18
Results – employment effects
Results – wage effects, summary
• Differences in psychological traits explain about 18% of the wage gap.
• They explain about 7% of the employment gap.
• Bigger effect on wages than employment.
Mainly influenced by: agreeability, neuroticism, and conscientiousness.
19
Effects of psychological traits in testing – Weichselbaumer 2000
• Unexplained wage gap contains unobserved productive differences.
• CV-based tests do not make use of the opportunity to filter out the effect of characteristics that are generally not observable.
• Certain typically male characteristics (dominant, confident, go-getter, competitive) may be important for high-paying jobs.
Testing, where these differences are controlled for via the CV-s: can have female applicant with typically male characteristics, and vice versa.
If there is still a difference: taste-based discrimination.
Categorization of traits by gender
20
Female applicants – signaling typically male and female characteristics
Testing the signals of male/female traits
Survey of 119 students
21
Results: typically male occupations
Results: typically female occupations
22
Summary – Weichselbaumer 2000
• In typically male jobs, men had a significant advantage over both types of women, no difference between two women.
• No significant difference when labor supply is low: suggests taste-based discrimination.
• In female jobs, women have a significant advantage.
If there were no discrimination, the results of the masculine female applicant would be closer to the male’s results.
Results suggest that employers (taste) discriminate based on gender, not statistics.
• Problems with the methodology?
Gender and competition – Booth 2009
• Wage increases and promotions are often based on competition it’s possible that women don’t like to compete as much
– Datta Gupta et al 2005: experiments where subjects have a choice between payment based on piece rates or on competition – women tend to choose the less competitive option.
• Question: is this an innate difference, or the effect of the environment (nature v.
nurture)?
– If innate: no need to address labor market differences, these are due to preferences.
– If environmental: consequence of prejudice, and it is important to address it, as it may lead to loss of talent.
• Empirical test: how often boys/girls from mixed gender or all girl/boy schools take part in competitive tournaments
– Girls in mixed gender schools tend to act more according to traditional female roles – and compete less.
23
Attitudes towards competition: innate or environmental effect?
• Results: in all girl schools, girls choose to compete equally often as boys, in mixed gender schools they do not.
Learned behavior, not innate characteristics
• Other studies on competitive attitude:
• Paserman 2007: based on results from tennis tournaments, women make more mistakes during critical points in the match.
• Gneezy et al 2007: in patriarchal societies men, in matrilinear societies women choose to compete more often.
• Gneezy-Rustichini 2002: 9 year old children first run alone, then in pairs
• Alone: no significant difference between the results of boys and girls
• In pairs: boys ran faster
Competition led to better performance for boys, not for girls.
