Presentation on theme: "The parental leave reforms and childbearing in Poland Findings from Cox proportional hazard model for transitions to the first and second child for Polish."— Presentation transcript:
The parental leave reforms and childbearing in Poland Findings from Cox proportional hazard model for transitions to the first and second child for Polish women Anna Kurowska, PhD Institute of Social Policy, University of Warsaw LPR Network annual seminar 18-19 September 2014 Tallinn, Estonia
Motivation (policy context): In 2013 a major reform of Polish parental/maternal leaves: introduction of 52 weeks of parental leave paid at 80% of previous earnings (in fact: extension of maternity leave); One of the policy goals was to increase the fertility rate in Poland. Poland has been experiencing the lowest low fertility rate (TFR around 1,3) for over ten years.
Research question: Can we expect an increase in fertility in Poland as a result of the extension of maternity leave? Research for other countries bring rather positive results, e.g.: Rønsen (1999, 2004) positive impact of the length of maternity leave on childbearing in Finland and Norway; Lalive i Zweimuller (2005) also positive results for Austria. No similar analysis for Poland.
Can we say something about the impact of the length of maternity/parental leave on childbearing in Poland from the past? Kkkkkkkk kkkkkkkkk Number of paid maternity leave weeks entitlement For the first child For the next child
Cox proportional hazard model on GGS data for Poland (wave 1): transitions to the first and second child Generations and Gender Survey 2010/2011: 20 000 respondents, part of the Generations and Gender Programme (UNDP) Subsample: 5501 women born in 1960 or later: 1620 childless, 1390 - 1 child; 1666 – 2 children, 825 – 3 children or more Model: Independent variables: -length of the maternity leave to which a woman was entitled at each moment in time; -controls 1 (variables that changed in time): marital status, employment status at each moment in time; level of education at each moment in time; -controls 2 (fixed characteristics): religiousness of women’s parental home; background place of residence; age at first birth (for the transition to the second child)
Results: *** significant at : ***p<0,01, **p<0,05, *p<0,1 Variable Transition to the first child Transition to the second child estimatehazard ratioestimatehazard ratio length of maternity leave (entitlement) -0,0080,992-0,045***0,956 Control variables: employment status 0,513***1,6681,018***2,767 level of education -0,024**0,976-0,0180,982 importance of religion at responden’s family home -0,0240,976-0,085***0,918 marital status 1,969***7,1680,546***1,726 age of first birth XX-0,083***0,921 place of residence in childhood -0,103***0,901-0,236***0,790
Conclusions and discussion: The length of the maternity leave in Poland up to 2011 did not have a positive impact on childbearing: neither for the transition to the first, nor to the second child. It is employment that plays a crucial role: it seems that the employment is a condition for many women to have the first and the second child (results in line with Matysiak 2009 and 2011). Driving from this results it is hard to expect a new reform to positively impact childbearing in Poland. However, the previous changes to the length of maternity leave much smaller that in case of the present reform. Future research: the evaluation of the “day after” effects of the reform based on the 2014 2 nd wave of GGS-PL.
More on the presented results: Kurowska A., Słotwińska-Rosłanowska E., Zatrudnienie kobiet a pierwsze i drugie urodzenia wśród kobiet w Polsce, Studia Demograficzne nr 1 (163), pp 37-51. (link to be found at my profile at: Academia.edu) An updated version in English: in preparation.