School to work transition and first marriage in the Caucasus and - PowerPoint PPT Presentation

School ‐ to ‐ work transition and first marriage in the Caucasus and Central Asia Presentation at the IUSSP IPC, Cape Town, South Africa Session 187 “The intersection between family formation, education and work” 2 November 2017 Prof. Dr. Michael Gebel Methods of Empirical Social Research (Chair) University of Bamberg, GERMANY

1. Motivation  The transition from education to work is a central stage in the individual life course (Buchmann/Kriesi 2011)  Finding a “good” job is often seen as an important precondition to form a family (marriage/childbirth) and leaving parental home  Education expansion (prolonged participation + higher attainment) and labor market insecurities (unemployment, temporary + informal work) are seen as factors contributing to the delay of family formation worldwide (Lloyd 2005)  Lack of research on the region of the Caucasus and Central Asia (CCA) Research question: How does education and economic activity status affect the timing of first marriage in CCA countries? (1) Motivation 2

2. The TEW ‐ CCA project “Opportunities and Barriers at the Transition from Education to Work. A Comparative Youth Study in Azerbaijan, Georgia and Tajikistan.” Facts Duration: 2015 ‐ 2018 Financed: 0.4 Mio € by Volkswagen Foundation Partners: ~20 researchers from 3 countries www.tew ‐ cca.de @TEW_CCA (2) The TEW ‐ CCA project 3

3. Theory and hypotheses The role of education participation and attainment  Hyp 1: Individual enrollment into education and attaining higher levels of education delay marriage timing.  Hyp 1a: The effects should be more pronounced for women than for men. Labor market insecurities: Men  Hyp.2: We expect for men that having a formal sector jobs increase the transition rate to first marriage compared to working in the informal sector. The lowest transition rates should be observed among unemployed or inactive men. Labor market insecurities: Women  Hyp 3: We expect faster transitions to first marriage among inactive women compared to women who are engaged in the labor market and no differences among women holding different job types.  Hyp 3a: These patterns should be particularly pronounced in Muslim CCA societies. (3) Theory and hypotheses 4

4. Research design Data Survey on Young People in Azerbaijan (N=2002), Georgia (N=2000), Tajikistan (N=2000) Using monthly retrospective life history data starting at age 15 Sample (“risk set”) 18 ‐ 35yo who left education during the last 10 years (retrospective; start age 15) Dependent variable Hazard rate of first legal marriage: entering first legal marriage (1=marriage; 0=staying at least one additional month unmarried) (not yet considered: religious marriage) Method  Discrete ‐ time logistic hazard rate model (Jenkins 1995) with piecewise constant specification of the base ‐ line hazard rate (5years age groups)  estimation of average marginal effects (AME) to allow model comparisons and easier probability interpretations (Mood 2009) (4) Research design 5

4. Research design Central independent variables of interest  Highest level of education attainment (time ‐ varying, measured after leaving education; category “being in formal education” to distinguish participation effect)  Monthly economic activity status (time ‐ varying) Independent variables as “control variables”  Gender (stratification variable)  Age (5y intervals)  Disability  Highest level of parental education, parental wealth, mother’s employment status (when respondent was 15yo); living with two parents (vs. less); number of brothers and sisters  Share of time lived in urban area until time of leaving education  Ethnicity; religiosity  Year of leaving education (4) Research design 6

5. Descriptive results Institutional context  Decline of VET and VET became unpopular; strong tertiary education expansion during post ‐ socialist time Highest level of education attainment (in %) AZE GEO TJK male female male female male female Basic secondary 6 6 14 9 9 19 Upper secondary 48 40 32 20 36 49 Initial professional 5 3 12 15 4 1 Secondary professional 10 24 3 5 9 11 Undergraduate: BA 28 25 30 37 19 8 Graduate: MA/PhD 4 3 9 14 23 12 (5) Descriptive results 7

5. Descriptive results Inactivity  Strong labor market orientation of women during socialist period  However, also strong influence of Muslim and Orthodox religion with traditional values  Pattern of “school ‐ to ‐ home transition” especially among women and in Muslim CCA societies; women often mention in AZE and TJK that they were “not allowed to work” Inactivity rates in the observed period after leaving education (in %) AZE GEO TJK male 7 10 8 female 40 25 48 (5) Descriptive results 8

5. Descriptive results Time until finding a first job, Kaplan ‐ Meier estimates 1.00 0.75 0.50 0.25 0.00 0 20 40 60 months since leaving education AZE male AZE female GEO male GEO female TJK male TJK female (5) Descriptive results 9

5. Descriptive results Type of first job  Labor market segmentation: formal versus informal sector  Gender ‐ specific pattern: strong formal sector orientation of women (due to preferences and occupational segregation)  Limited role of family businesses and self ‐ employment at the beginning of the working career Type of first job (in %) AZE GEO TJK male female male female male female formal/registered employee 43 77 58 77 48 64 informal/unregistered employee 40 15 24 18 43 24 employee/helper in family business 7 4 11 2 2 3 self ‐ employed/employer 11 3 6 3 7 9 (5) Descriptive results 10

5. Descriptive results Age at first legal marriage, Kaplan ‐ Meier estimates 1.00 0.75 0.50 0.25 0.00 15 20 25 30 35 age Azerbaijan male Azerbaijan female Georgia male Georgia female Tajikistan male Tajikistan female (5) Descriptive results 11

6. Multivariate results AZE GEO TJK Men Women Men Women Men Women Ref. formal/registered employee initial formal education ‐ 0.24*** ‐ 0.18 ‐ 0.03 ‐ 0.31*** ‐ 0.26** ‐ 0.90*** (0.07) (0.12) (0.11) (0.11) (0.11) (0.25) informal/unregistered employee ‐ 0.10** ‐ 0.34*** ‐ 0.03 0.04 ‐ 0.02 0.12 (0.05) (0.12) (0.06) (0.15) (0.09) (0.34) employee/helper in family business ‐ 0.14 0.24 ‐ 0.06 0.75 ‐ 0.04 ‐ 0.82* (0.09) (0.34) (0.07) (0.46) (0.45) (0.47) self ‐ employed/employer 0.23** ‐ 0.39* 0.05 ‐ 0.25 0.29* 0.15 (0.10) (0.23) (0.10) (0.20) (0.17) (0.49) apprentice/trainee n.a. n.a. n.a. 0.37 0.49** 0.29 (0.96) (0.19) (0.37) military service ‐ 0.27*** n.a. ‐ 0.14 n.a. ‐ 0.27 n.a. (0.05) (0.09) (0.34) unemployed ‐ 0.15*** ‐ 0.08 ‐ 0.06 0.20** 0.21* 0.18 (0.05) (0.09) (0.04) (0.08) (0.11) (0.28) inactive ‐ 0.30*** 0.94*** n.a. 0.94*** ‐ 0.23** 0.94*** (0.07) (0.15) (0.21) (0.11) (0.29) Remarks: AME; standard errors in parantheses; control variables included into the model but not shown. (6) Multivariate results 12

7. Conclusions  Great potential of analyzing youth transitions in CCA region with the new retrospective life history data on young people  Clear strengths and weaknesses of the data: Objective facts can be gathered via retrospective questions; but no retrospective subjective questions  Many patterns known either from post ‐ socialist countries or Muslim MENA countries  Both similarities and differences can be found  For most of the subgroups education participation delays marriage; effect stronger for women in GEO and TJK  Informal sector employment delays marriage in AZE, no effects in GEO and TJK  Inactivity delays marriage for men and speeds up marriage for women  Unemployment delays marriage for AZE men and speeds it up for TJK men and GEO women (7) Conclusions 13