Fertility, household models and labour market outcomes in EU countries. An analysis of the gender gap in parenthood penalty and the moderating role of family policies

[TITLE] Fertility, household models and labour market outcomes in EU countries An analysis of the gender gap in parenthood penalty and the moderating role of family policies Piotr Lewandowski, Cristiano Perugini, Fabrizio Pompei, Laurène Thil, Maryna Tverdostup, Wojciech Szymczak Grant agreement no. Deliverable: D3.2 Due date: 01.2024 PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 2 Document control sheet Project Number: 101061388 Project Acronym: WeLaR Work-Package: [3] Last Version: [24/01/2024] Issue Date: [ ] Classification Draft Final X Confidential Restricted Public X Legal notice This project, WeLaR, has received funding under the Horizon Europe programme. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or the European Research Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 3 Table of contents 1. Introduction ....................................................................................................................................... 6 2. Gender gaps in paid work and housework in the EU ................................................................. 9 2.1. Data and methods .................................................................................................................................. 9 2.2. Descriptive evidence on gender disparities in time use ................................................................. 12 2.3. Time allocation gender disparities in European couples ................................................................ 15 2.4. Within-couple time allocation gaps and macro-level factors ........................................................ 17 3. Mapping labour market parenthood penalties in the EU ......................................................... 23 3.1. Data and methods ................................................................................................................................ 23 3.2. Gender gaps and parenthood across the EU .................................................................................. 29 3.3. Gender gaps and parenthood across household types ................................................................. 37 4. Policies, reforms, and gender disparities in parenthood penalty across the EU ................. 39 4.1. Data and methods ................................................................................................................................ 39 4.2. Policies and motherhood penalty in labour supply and in employment ...................................... 45 4.3. Policies and motherhood penalty in employment characteristics ................................................ 53 5. Summary and concluding remarks .............................................................................................. 54 References ........................................................................................................................................... 58 Appendix .............................................................................................................................................. 64 PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 4 Acknowledgements The authors are grateful to Mikkel Barslund and Ludivine Martin for their comments on a previous version of the paper, which contributed to its overall coherence and clarity. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 5 Abstract In this paper, we analyse the interactions between fertility, parenthood, household characteristics, labour market outcomes and institutions through the lens of gender inequality. In the first part of the report, we use HETUS data for ten EU countries for the period 2008-2015 to provide a crosscountry descriptive analysis of the disparities in time allocation to paid work, housework and childcare within the household. In the second part, we analyse gender asymmetries in labour market outcomes in relation to parenthood across the EU and different household types. To this aim, we use the European Union Survey on Income and Living Conditions (EU-SILC) data to assemble a longitudinal dataset at the demographic group (gender, age, education) level for twenty-three EU countries over the period 2006-2018. This dataset is also used, in the third part of the report, to analyse how gender asymmetries in the effects of parenthood on labour market outcomes are moderated by an array of family-related public policies. Our results provide a widely informative mapping of the asymmetries in parenthood penalty in labour market outcomes across the EU and suggest that only some of the existing policies help reduce this gender gap in labour supply (at the intensive and the extensive margin), employment and some job attributes. However, none of them plays any role in moderating gender disparities in labour remuneration related to parenthood. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 6 1. Introduction Understanding the drivers and evolutions of labour supply is key to assessing labour market performance and designing policy interventions. Individual labour supply choices are not isolated from decisions and circumstances shaped by non-strictly economic factors. In particular, the household structural features related to the presence and rearing of children have been investigated as one major driver of labour market behaviour. As such interlinks are not symmetric across genders, the interest in these factors extends well beyond the labour market spheres, reaching the domain of gender economic and social inequality. In the EU, the still large and persistent gender imbalances in formal and informal work and the heterogeneous (across countries) fertility dynamics represent key areas of debate and policy intervention. This calls for a comprehensive effort to analyse how the distribution of housework responsibilities in general, and the presence of children in particular, shape labour market outcomes and which policy/institutional settings moderate the link. The last decades marked an unprecedented shift in the work and family roles of women, which has materialised in a significant improvement in female labour market performance. Increased labour market participation, especially among mothers with small children, narrowing gender inequalities in wages, thinning of the glass ceiling effect 1 and more gender equality in job promotion and career progression are among the main domains in which achievements have been observed (Schröder and Burow, 2016; Bertrand et al., 2015; Greig and Bohnet, 2009; Albanesi and Olivetti, 2009; Aguiar and Hurst, 2007; Fuwa, 2004; Álvarez and Miles, 2003). This transformation of female labour market position and diminishing gendered allocation of paid work, has fostered women’s sounder employment commitment, higher career aspirations and stronger bargaining power at the workplace and within the family. Despite this progress, women still perform most of the housekeeping and childcare work (Sánchez et al., 2021; Zamberlan et al., 2021; Lee et al., 2021; Sullivan and Gershuny, 2016; Bianchi et al., 2012; Blau and Kahn, 2007). Female achievements in the labour market have indeed not translated into an equivalent increasing role of male spouses in at-home labour and, more generally, into a more egalitarian gender division of housework and care (Mandel and Lazarus, 2021; Fuwa, 2004); this situation places an additional burden on women, often referred to as "double days". Within-family division of housework loads along traditional 1 The term "glass ceiling" implies that while individuals from underrepresented groups (such as women) may be able to see the opportunities for advancement, they are hindered by an invisible barrier that prevents them from reaching top-level positions within an organization. This barrier can manifest in various forms, including discriminatory hiring practices, unequal pay, lack of mentorship and sponsorship opportunities, and biased promotion decisions. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 7 gendered lines is documented to be extensive and persistent, with unpaid care work being placed at the core of gender inequality throughout Europe (Gálvez-Muñoz et al., 2011). This genderasymmetric time allocation pattern holds across countries, individual (men and women) profiles, household types and female bargaining power levels, measured by earnings (Sevilla-Sanz et al., 2010; Gupta, 2007). The presence of children magnifies gender-uneven allocation of household work (Kimmel and Connelly, 2007) and inevitably reverberates on labour supply decisions. This link appears so much disproportionally stronger for women that the presence of children is today seen as the main driver (if not the only remaining one) of labour market gender inequality (see Juhn and McCue, 2017; Vladisavljević et al., 2023). Numerous studies have highlighted that the association between children and labour market outcomes, referred to as child or parenthood penalty, is complex and depends on a variety of factors. Parenting can impact labour market outcomes by shaping labour supply decisions, employment opportunities, and labour returns. Regarding the first domain, extensive evidence exists that childbirth decreases participation rates and hours worked only for mothers (OECD, 2007; Schönberg and Ludsteck, 2014; Brewer and Paull, 2006). This effect is observed even after accounting for the possible endogeneity of fertility and for adverse selection (e.g., Angrist and Evans, 1998; Jacobsen et al., 1999; Cruces and Galliani, 2007). This loss is often paralleled by a penalty in the wage rate (e.g., Lundborg et al., 2017; Adda et al., 2017), especially when mothers experience substantial interruptions in employment (Lundberg and Rose, 2000). Mothers accumulate less job experience and, due to continuing responsibilities in child rearing, face more challenging career/family conflicts in coping with long hours, heavy travel commitments and inflexible work schedules. As a result, they tend, more often than men, to choose family-friendly jobs and to be less competitive for higher-paid jobs (Bertrand et al., 2010; Kleven et al., 2019a; Perugini and Pompei, 2023). An interesting branch of the literature has identified several individual and household attributes that can mitigate or exacerbate the negative effects of childbirth. Among the individual attributes, age, education and the type of occupation pre-birth emerge as relevant in one direction or the other depending, to a significant extent, on the socio-economic and institutional context (see Sigle-Rushton and Waldfogel, 2007; Davies et al., 2000). Household characteristics (income, age/employment composition) have been less explored, despite their ability to shed light on aspects related to gender role beliefs and stereotypes. Interestingly, a few contributions focus on the role of spouses’ attributes. Bertrand et al. (2010) show that US graduate mothers with lowerearning spouses suffer only a modest and temporary penalty compared to those with higherearning spouses, who tend to reduce their labour supply considerably more. Fernàndez et al. (2004) focus on the role of the family model in which the man grew up and find that the spouses of men whose mothers worked are themselves significantly more likely to work. Kleven et al. (2019a) find PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 8 that the child penalty for mothers in Denmark is strongly related to the labour supply history of maternal grandparents: women whose mothers worked very little compared to their father suffer a larger child penalty when they become mothers. Various studies have devoted attention to the role of institutions and policies in shaping the consequences of parenthood on labour market outcomes; most of them refer to a specific country or a limited set of countries, due to the demanding nature of data needed to identify the key relationships. They suggest that high fertility rates are associated with a decrease in women's labour force participation, especially in countries with inadequate childcare support (e.g., Herbst, 2010). In contrast, family-friendly work policies, such as well-paid parental leave and flexible working hours, have been shown to support the labour force participation of individuals with children and positively impact fertility rates (Del Boca, 2015). As for the EU context, recent empirical evidence on single countries about fertility and labour supply suggests that there is a negative relationship between fertility and women's labour force participation, particularly those with lower levels of gender equality and a more generous infrastructure of childcare and family policy provision (see Fehr and Ujhelyiova, 2013; Neyer, 2006). In some contexts, access to high-quality, affordable childcare has been associated with higher levels of labour force participation among women with children (e.g., Gehringer and Klasen, 2017). Despite the abundant body of knowledge produced in the last decades, a fine-grained, EU-wide, and updated analysis of gender asymmetries associated with the presence of children is not available; similarly, further studies are needed to understand the interplay between family policies and labour market outcomes in the presence of children (Blau and Winkler, 2017). This paper aims to provide an overview of the relationship between parenthood and labour market asymmetries across genders in EU countries; it also aims to shed light on the moderating effects of institutional and policy settings on the extent and asymmetry of the child penalty across genders. To this aim, we assess the work-life balance by organising our research into two main stages, devoted to different domains of gendered division of labour: housework and childcare. As a first step, we provide a cross-country analysis of the disparities in time allocation to paid work, housework and childcare within the household. Specifically, we rely on the Harmonized European Time Use Survey (HETUS) micro-level data from ten European countries (in the reference year 2010) to analyse the scale and cross-country variation in within-couple gender gaps in employment, housework and childcare, in relation to a broad range of spousal and household characteristics and of country-level indicators of gender equality. In the second part of the analysis, we focus on gender asymmetries in labour market performance related to the presence of children. To this aim, we use microdata from the European Union Survey on Income and Living Conditions (EU-SILC) to assemble a longitudinal dataset at the demographic group (gender, age, education) level for twenty- PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 9 three EU countries over the period 2006-2018. We first map the gender gap in labour market outcomes related to children across country groups and household characteristics. We then assemble a multi-level dataset by matching the demographic group database with higher level information (at country-level) to analyse how changes in a broad array of institutional settings and policies (such as length and generosity of parental leave, childcare services, child-related benefits, family benefits, work-life balance, and gender-balanced parenting) affect the asymmetry in fertilityrelated labour market outcomes. The rest of the paper is structured as follows. Section 2 provides an overview of gender disparities in the allocation of paid work and housework in selected EU countries. After a presentation of the data and the methods of analysis, we provide a descriptive picture of gender disparities in time use, the role of different individual and household contexts and country-level institutional factors. Section 3 maps the magnitude and variability of the parenthood penalty in the EU in a broad set of labour market outcomes: labour force participation, employment, hours worked, type of employment and earnings. After an illustration of the data and methods used, the presentation of the outcomes is detailed by groups of geographically contiguous countries (Southern, Continental, Northern, and Central-Eastern countries) and by household typologies. In Section 4 we analyse the moderating effects of various policy measures and reforms on the gender asymmetry in the parenthood penalty in labour market outcomes, identifying those measures that alleviate women’s disadvantage compared to men. To this aim, we first describe the country-level data assembled and the empirical approach; we then highlight which policy and reform contexts can alleviate the gender asymmetry in the parenthood penalty in labour force participation, employment, and job characteristics. Section 5 summarizes and concludes. 2. Gender gaps in paid work and housework in the EU 2.1. Data and methods This analysis uses Harmonized European Time Use Survey (HETUS) round 2 data, with reference year 2010 2 . Apart from a range of core socio-demographic, household, and employment characteristics of respondents, HETUS collects information, by means of a self-recorded diary by all household members, on how individuals spend their time in various activities such as work, 2 For more information on HETUS data, see: https://ec.europa.eu/eurostat/documents/203647/10397147/HETUS_variables+description.pdf/54efe947-767f-2e4609ee-fc13688f04ac PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 16 gaps turn out not significantly different from zero (see panel (i)). In Estonia the gender disparity in relative worktime turns statistically insignificant upon controlling for a full set of individual and household characteristics. This evidence suggests that, in Finland and Estonia, the worktime gap between wives and husbands in dual-earner couples stems from observed disparities in individual demographic and employment profiles, with the type of employment contract (partor full-time) emerging as the main driver (see Tables A1 and A2 in Appendix). In all other countries, wives work drastically fewer hours than their husbands, even demographic and employment profile being equal, with adjusted gender gaps in relative worktime ranging from 9.3 p.p. in Luxembourg (implying that husband’s adjusted average share of worktime is 54.7% and wife’s is 45.3%) to around 3 p.p. in France, Poland, and the UK (51.5% and 48.5% are the average worktime shares of husbands and wives, respectively). Panel (ii) of Figure 1 reveals even stronger gender-based segregation. In all sample countries, the time wives invest in housework exceeds the time invested by their husbands enormously, even when demographic, employment and household characteristics are controlled for. Albeit crosscountry variation in the magnitude of the gender gap in housework is remarkable, no country appears gender-equal in terms of within-couple allocation of time in housework. In Finland – the country with zero gender gap in relative worktime – wives invest, on average, 13.4 p.p. more time in housework than their husbands, implying that within-couple share of wives’ housework time is around 56.7% and husbands’ is around 43.3%. In Estonia – the other country posting a nonsignificant adjusted gender gap in relative worktime – adjusted gender disparity in relative housework climbs to 31.9 p.p.; this means that 66% of the couple’s joint total housework time is allocated to the wife, as opposed to 34% of the husband. The absolute largest within-couple gender inequality in housework is recorded in Greece, where wives are allocated 80.5% of the couple’s housework time and husbands only 19.5%. These findings indicate that, even if women achieve equality in terms of their labour market commitment and work comparable hours as their husbands, like in Finland, the workload at home is still disproportionally on them. Panel (iii) of Figure 1 displays the unadjusted and adjusted gender gaps in relative childcare. The results reveal a stark variation in within-couple equality in childcare across the sample countries. Finland appears to be the most equal country in terms of within-couple division of childcare, where wives and husbands assume equal shares of childcare once observable characteristics are controlled for. However, all other sample countries deviate from this equality, yet to varying extents. In Greece, Luxembourg, and Poland, wives do, on average, 60% of childcare time. Germany, Estonia, and the UK appear to be the most unequal, with the wife’s average contribution to childcare time ranging from 70% to 75%. The stark within-couple childcare inequality in Estonia appears rather surprising, given the insignificant wife-husband gap in worktime (panel (i) of Figure 1). PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 17 Figure 1. Gender gaps in relative worktime, housework and childcare, by country Notes: Tobit regression estimates based on HETUS wave 2010 data. The dependent variable is relative worktime (panel i) and relative housework (panel ii) censored at 0 and 1. The point estimates are reported with 95% confidence intervals. The unadjusted gap is estimated controlling for gender and year, month and day of the week fixed effects. The adjusted gap is estimated controlling for gender, age group, education level, migration status (being born in the survey country), household size, number of kids aged 0 to 6, number of kids aged 7 to 17, household net income band, full-time employment, industry of employment, as well as year, month, and day of a week fixed effects. The estimates account for combined individual response and day weight. Country sample sizes are as follows: Belgium – 790; Germany – 1632; Estonia – 532; Greece – 420; Finland – 600; France – 2992; Luxemburg – 426; Poland – 3132; Romania – 3752; the United Kingdom – 694. 2.4. Within-couple time allocation gaps and macro-level factors The magnitude of gender inequalities in within-couple time allocation is related, to a certain extent, to the overall level of gender equality in the labour market and in society. Earlier studies highlighted a significant association between country-specific macro-level factors and the level of gender equality in time use (Mandel and Lazarus, 2021; Grunow, 2019; Fuwa, 2004); similarly, several studies emphasised the importance of prevailing religion and cultural background for gender equality in time use (Burda et al., 2013). PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 18 Our sample covers an array of European countries having a heterogenous degree of gender equality, religious background, and stringency of gender norms 5 . In this section, we investigate whether such differences reverberate in the level of asymmetry in time use allocation. To this aim, we use six macro-level indicators extracted from the Eurostat database broadly capturing the level of gender equality and gender disparities in labour market participation and conditions: (i) the gender gap in employment, measured as the difference between the employment rates of men and women aged 20-64; (ii) the gender gap in part-time employment, defined as the difference between the share of part-time employment in total employment of women and men aged 20-64; (iii) unadjusted gender wage gap, measured as the difference between average gross hourly earnings of male paid employees and of female paid employees as a percentage of average gross hourly earnings of male paid employees; (iv) childcare enrolment, defined as the percentage of children (under 3 years old) cared for by formal arrangements other than by the family; (v) female representation in top management positions, defined as the share of female board members and executives in the largest publicly listed companies; (vi) female representation in executive government positions, defined as the proportion of women in national parliaments and national governments. Table A3 in the Appendix reports the levels of the sex indicators for the countries covered in our analysis in the years around the reference year of the HETUS data used here (2010). Figure 2 plots country-average estimates of the wife’s relative worktime against the macro-level indicators. The results suggest that the overall gender gap in part-time employment is, not surprisingly, the only macro-level indicator from our selection having a strong negative association with the wife’s average relative worktime (Spearman’s correlation coefficient 𝜌 = 0.6 , 𝑝 < 0.1 ). What seems more surprising is that all other macro-level characteristics have a statistically weak association with the within-couple gap in worktime, even though the small sample size makes the identification of statistically significant correlations difficult. The magnitude of the correlation coefficients between a wife’s average relative worktime and higher gender wage gap on one side ( 𝜌 = 0.418 ) and a higher share of children aged under 3 in pre-school education on the other ( 𝜌 = 0.479 ), signal that a significant association might exist. However, we lack statistical power to precisely estimate it. 5 According to wave 6 (2010-2014) of the World Value Survey, only 19% of respondents in Estonia find religion important in their lives, as opposed to 52% of Romania. Similarly, only 38% of respondents in Estonia see no problem in a wife’s income surpassing her husband’s, while in Poland the corresponding share is 65% (see: https://www.worldvaluessurvey.org/WVSOnline.jsp). PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 19 Figure 2. Country-level correlation of wife’s relative worktime and gender equality indicators Source: Macro-level indicators are available at https://ec.europa.eu/eurostat/web/main/data/database. Average estimates of the wife’s relative worktime are estimated using HETUS wave 2010 data. Notes: The correlation between relative worktime and macro indicators is estimated using Spearman's rank correlation coefficient (rho). Each panel depicts country-level correlation between the average female within-couple relative worktime and (i) the male-female gap in employment; (ii) the female-male gap in part-time employment; (iii) the male-female unadjusted wage gap; (iv) the percentage of children (under 3 years old) cared for by formal arrangements other than by the family; (v) share of female board members and executives in the largest publicly listed companies; (vi) the proportion of women in national parliaments and national governments. Country sample sizes are as follows: Belgium – 790; Germany – 1632; Estonia – 532; Greece – 420; Finland – 600; France – 2992; Luxemburg – 426; Poland – 3132; Romania – 3752; the United Kingdom – 694. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 20 Figure 3 plots the same associations but for the wife’s average relative housework. Two out of six macro-indicators have a significant association with the country’s average share of housework done by wives. The gender gap in employment is strongly and positively related to within-couple housework disparity ( 𝜌 = 0.83 , 𝑝 < 0.01 ). A similar association between female labour market participation and the average gap in housework was documented by Mandel and Lazarus (2021). Figure 3. Country-level correlation of wife’s relative time spent on housework, including childcare, and gender equality indicators Source: Macro-level indicators are available at https://ec.europa.eu/eurostat/web/main/data/database. Average estimates of wife’s relative worktime are estimated using HETUS wave 2010 data. Notes: The correlation between relative worktime and macro indicators is estimated using Spearman's rank correlation coefficient (rho). Each panel depicts country-level correlation between the average female within-couple relative time spent on housework, including childcare, and (i) male-female gap in employment; (ii) female-male gap in part-time employment; (iii) male-female unadjusted wage gap; (iv) the percentage of children (under 3 years old) cared for by formal arrangements other than by the family; (v) share of female board members and executives in the largest publicly listed companies; (vi) the proportion of women in national parliaments and national governments. Country sample sizes are as follows: Belgium – 790; Germany – 1632; Estonia – 532; Greece – 420; Finland – 600; France – 2992; Luxemburg – 426; Poland – 3132; Romania – 3752; the United Kingdom – 694. Yet, the association grew weaker over the last decades suggesting a strengthening position of women in within-couple bargaining over housework even if their relative advantage on the labour market sees no major improvements. Furthermore, we document an important correlation between PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 21 a share of women in top-level management positions in the country and the average within-couple gender gap in housework. In countries with higher representation of women in top management, within-couple division of housework appears more balanced ( 𝜌 = −0.802 , 𝑝 < 0.01 ). While the gender gap in employment is a direct indicator of female labour market attachment, the share of women in top managerial positions is an indicator of gender-unbiased labour market, providing equal career growth opportunities to both men and women. Figure 4. Country-level correlation of wife’s relative time spent on childcare, and gender equality indicators Source: Macro-level indicators are available at https://ec.europa.eu/eurostat/web/main/data/database. Average estimates of wife’s relative worktime are estimated using HETUS wave 2010 data. Notes: The correlation between relative childcare and macro indicators is estimated using Spearman's rank correlation coefficient (rho). Each panel depicts country-level correlation between the average female within-couple relative time spent on housework, including childcare, and (i) male-female gap in employment; (ii) female-male gap in part-time employment; (iii) male-female unadjusted wage gap; (iv) the percentage of children (under 3 years old) cared for by formal arrangements other than by the family; (v) share of female board members and executives in the largest publicly listed companies; (vi) the proportion of women in national parliaments and national governments. Country sample sizes are as follows: Belgium – 790; Germany – 1632; Estonia – 532; Greece – 420; Finland – 600; France – 2992; Luxemburg – 426; Poland – 3132; Romania – 3752; the United Kingdom – 694. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 22 Thus, the association we have identified signals that overall gender equality in society and in the labour market translates into the couple-level arrangements related to house chores. The share of women in local governments and in the parliament reveals economically meaningful, yet statistically weak association presumably due to insufficient sample size, with wife’s average relative housework ( 𝜌 = −0.455) . Nonetheless, the direction and strength of the association appear sufficient to argue that representation of female in political decision-making contexts captures the level of female emancipation on societal level, which shapes an environment conducive to more balanced within-couple housework arrangements. Figure 4 illustrates the relationship between the wife's average relative housework and the six macro-level indicators. None of the association is statistically significant. The gender gap in parttime work and wages has an economically meaningful, yet statistically weak, correlation. The direction of the latter associations is rather expected, with the wife's average relative childcare being higher in countries where part-time work is more prevalent among women and wages are more unequal. However, there is a weak, but economically non-negligible positive association between the share of children aged under 3 in childcare and the wife's relative childcare, which is surprising. This suggests that the accessibility of early age childcare does not reduce within-couple inequality in childcare. Our results suggest that within-couple gendered disparity in worktime is narrowing and, in some European countries, including Finland and Estonia, it turns insignificant once spousal and housework characteristics are controlled for. This finding indicates that the discrepancy between the wife’s and the husband’s workhours is gradually vanishing, and that the wife’s labour market attachment is strengthening. Nevertheless, housework and childcare remain divided along traditionally gendered lines within a couple, with the wife assuming a larger share of the household chores and childcare duties even in dual-earner couples in all the sample countries. Thus, if women achieve equality in terms of their labour market commitment, as is the case in Finland and Estonia, they still contribute to housework and childcare disproportionally more than their husbands. However, within-couple housework disparity does appear to be smaller in countries with higher female labour market attachment and higher degrees of overall gender equality on the labour market and in the society. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 23 3. Mapping labour market parenthood penalties in the EU 3.1. Data and methods In this section, we use data from the European Union Survey on Income and Living Conditions (EUSILC) to provide a descriptive picture of the gender gap in labour market outcomes associated with parenthood. EU-SILC is a household and individual data collection that provides comparable data on income, poverty, social exclusion and living conditions in European countries, along with detailed individualand household-level demographic, socio-economic and labour market information. EU-SILC provides two types of microdata: (i) cross-sectional data over a given time or a certain period with variables on income, poverty, social exclusion, and other living conditions; (ii) longitudinal data on individual-level changes over time, observed periodically over a 4-year period. Unfortunately, the length of the longitudinal dimension is not sufficient to carry out the analysis of the effects of the parenthood penalty using an event-study approach, which is standard in this type of research (see Kleven et al., 2019a and 2019b). For this kind of analysis, we would indeed need to observe each individual over a time interval ranging from a few years before the event (birth of a child) to some years after. Due to the limited availability of appropriate (longitudinal) datasets, such studies are indeed normally conducted for a single or a limited set of countries: the US (Bertrand et al., 2010; Cortés and Pan, 2020), Sweden (Angelov et al., 2016), Denmark (Kleven et al., 2019a), Russia (Vladisavljević et al., 2023), and a set of six developed economies (five European countries plus the US) (Kleven et al., 2019b). Restricting the sample to the limited set of EU countries for which adequate and accessible longitudinal data are available (basically, the four countries used by Keleven at al., 2019a, i.e., Sweden, Denmark, Austria and Germany) proved to be not functional to the main aim of our analysis. To assess how policies and institutional settings moderate the association between parenthood and labour market outcomes (see section 4) we indeed need to observe policy variability across countries and, more importantly, over time. Restricting the analysis to a few countries and for the time intervals available would have limited the scope of the analysis in terms of policy changes considered and posed serious limitations to the generalization of results. As a second-best solution, we use EU-SILC microdata to assemble a pseudo-panel dataset in which the unit of observation is a demographic group (see Doorley et al., 2023). Specifically, for each country, we identify 90 demographic groups defined by gender (men and women), education level (basic, secondary, tertiary), age (five 10-year age groups: 20-29, 30-39, 40-49, 50-59, 60 or more years-old) and number of children (zero, one, two or more). Our sample includes 23 out of 27 EU countries; Bulgaria, Romania, Malta and Croatia have been excluded due to unavailability of data for some crucial variables. For descriptive purposes, we group the countries of the sample as follows: Austria, PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 24 Belgium, France, Germany, Netherlands, and Luxembourg (Continental European countries); Denmark, Finland, Sweden, and Ireland (Northern European countries); Cyprus, Greece, Italy, Spain, and Portugal (Southern European countries); Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovak Republic (Eastern European countries). Although the grouping is geography-based, it also reflects some common historical, cultural, and institutional features relevant to the aims of the analysis 6 . The analysis covers the years from 2006 to 2018, a period long enough to include a variety of policy changes implemented before the outburst of the labour market effects of the Covid-19 pandemic, which might be a confounding factor difficult to handle. EU-SILC data provides a rich set of variables that can be used as metrics of labour market performance and as their drivers (see Table A4 in the Appendix). To the aims of this analysis, we consider the following set of labour market outcomes at the demographic group level: labour force participation rate, employment rate, weekly hours worked, employment status (self-employed, employee), type of contract (permanent/temporary and part-time/full-time), and hourly remuneration (real hourly wage and real hourly earnings). The set of individual/household characteristics used as drivers and controls include self-reported health status, marital (married on consensual union) status, migration, whether the individual is the respondent of the survey, household disposable equivalent income, household size, household dependency ratio, total time spent in caregiving activities by the household members. As all variables are defined as shares or averages at the demographic group level and to guarantee their reliability, we restrict the sample used for the whole empirical analysis to those demographic groups in which we observe at least 10 individuals. To control for country-specific structural features and for the macroeconomic cycle, we also include country-level controls for the unemployment rate, per capita GDP, and employment shares in the secondary and tertiary sectors. Table 2 provides a descriptive picture of labour market outcomes in the total sample and in subsamples identified by the individual characteristics that define the demographic groups. 6 The group of the Northern European countries is probably the most heterogeneous due to the presence of Ireland, where the characteristics and the generosity of family policies differ substantially from the other three Nordic countries. Nonetheless, as the measurement of the child penalty (and of the moderating effects of policies) is carried out by means of a regression approach, the allocation of Ireland into one of the macro-groups was needed to keep the number of observations adequate. As all regressions include time and country fixed effects, the otherwise unobserved specificities of single countries are accounted for in the estimation of the child penalty. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 25 Table 2. Descriptive labour market outcomes by gender, education and age (23 EUcountries, 2006-2018) lab force employed hours full time permanent self employed wage h earn h Total Sample 0.766 0.679 48.022 0.862 0.899 0.138 10.695 12.125 Gender Men 0.841 0.754 52.459 0.951 0.914 0.175 11.370 13.246 Women 0.688 0.601 43.009 0.765 0.882 0.099 9.939 10.856 Education Low 0.608 0.466 48.122 0.830 0.836 0.196 6.869 7.512 Medium 0.771 0.673 48.932 0.849 0.897 0.130 8.479 9.708 High 0.838 0.790 47.062 0.886 0.922 0.128 13.704 15.572 Age 20-29 0.735 0.608 46.954 0.851 0.802 0.064 7.902 8.480 30-39 0.870 0.776 47.930 0.879 0.889 0.111 9.650 10.666 40-49 0.875 0.795 48.340 0.863 0.926 0.144 10.923 12.530 50-59 0.829 0.744 49.067 0.870 0.942 0.179 13.230 14.945 600.182 0.162 46.757 0.813 0.937 0.250 12.240 14.878 Source: Own elaborations on EU-SILC data Notes: The unit of observation is a country/year-specific demographic group defined by gender/age/education/number of children. For the variables’ definition, see Table A4 in the Appendix. The first interesting piece of information relevant to our purposes regards the gender gaps that, as expected, are large in all labour market metrics. However, the significant heterogeneity also existing across education and age groups confirms the importance of controlling for such characteristics to avoid biased results and interpretations of gender disparities. The demographic groups with low levels of education are clearly in a weaker labour market position in terms of labour market performance and remunerations; they also exhibit higher self-employment rates, probably related to small businesses in specific sectors (such as farming or trade). The distribution of labour market participation and employment is also unequal by age group with the youngest and the oldest classes exhibiting, as expected, lower levels; average hourly wage and earnings increase with age, reflecting the role of experience and seniority. In Table 3 we provide a description of the gender gaps in the labour market indicators used, in the rest of the analysis, in association with parenthood. Men have on average a 15.2% positive gap in terms of labour force participation rates and employment compared to women; they work 9.5 hours more per week and significantly more on a full-time (18.5%) and permanent (3.1%) basis and as self-employed (7.6%). The unadjusted gender wage and earning gaps are also significant (about 14% and 22%, respectively). Those gender gaps calculated for the whole sample exhibit a remarkable heterogeneity across subsamples describing different household compositions related to children. The presence and number of children exacerbate all labour market gender disparities. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 32 Table 5. Baseline estimates: gender gap in employment and parenthood (1) (2) (3) (4) (5) (6) VARIABLES total male female no_child one_child two_child_more female -0.093*** -0.057*** -0.178*** -0.215*** (0.003) (0.003) (0.006) (0.007) one_child 0.095*** 0.081*** -0.068*** (0.008) (0.009) (0.010) two_child_more 0.124*** 0.071*** -0.225*** (0.020) (0.020) (0.024) sec_educ 0.040*** 0.033*** 0.040*** 0.035*** 0.072*** 0.080*** (0.005) (0.005) (0.006) (0.005) (0.008) (0.010) ter_educ 0.121*** 0.065*** 0.144*** 0.120*** 0.131*** 0.148*** (0.009) (0.010) (0.011) (0.010) (0.011) (0.013) age30_39 0.185*** 0.140*** 0.180*** 0.190*** 0.114*** 0.159*** (0.006) (0.007) (0.006) (0.009) (0.006) (0.007) age40_49 0.223*** 0.140*** 0.233*** 0.203*** 0.143*** 0.185*** (0.007) (0.008) (0.009) (0.011) (0.007) (0.009) age50_59 0.128*** 0.075*** 0.093*** 0.104*** 0.090*** 0.123*** (0.009) (0.010) (0.013) (0.013) (0.008) (0.010) age60_ -0.428*** -0.512*** -0.481*** -0.461*** -0.313*** (0.011) (0.014) (0.013) (0.016) (0.031) health -0.019*** -0.056*** -0.017* -0.038*** -0.048*** -0.022* (0.007) (0.007) (0.009) (0.008) (0.008) (0.012) migrant -0.045** -0.010 -0.076*** -0.072*** -0.049*** -0.095*** (0.019) (0.021) (0.018) (0.020) (0.018) (0.024) partner_house 0.011*** 0.058*** 0.021*** 0.021*** 0.034*** 0.041*** (0.004) (0.005) (0.007) (0.005) (0.010) (0.013) respond -0.014 0.088*** -0.013 0.070*** 0.054*** -0.123*** (0.012) (0.017) (0.021) (0.011) (0.016) (0.019) rel_disp_eq_income 0.080*** 0.083*** 0.098*** 0.084*** 0.043*** 0.057*** (0.012) (0.011) (0.015) (0.013) (0.010) (0.012) nhousehold -0.036*** -0.026*** -0.040*** -0.021*** -0.007 -0.057*** (0.004) (0.004) (0.005) (0.006) (0.007) (0.007) household_d -0.047** 0.043** 0.198*** 0.093*** 0.107*** -0.112*** (0.021) (0.020) (0.023) (0.027) (0.038) (0.019) n_care_hh -0.010*** -0.005*** -0.016*** -0.006*** -0.023*** -0.013*** (0.001) (0.001) (0.002) (0.002) (0.002) (0.002) ur -0.005*** -0.006*** -0.004*** -0.005*** -0.007*** -0.004*** (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) lr_pc_gdp -0.008 0.031 -0.036* 0.009 -0.039 -0.015 (0.017) (0.019) (0.021) (0.015) (0.031) (0.033) s_emp_sec 0.185 0.133 -0.089 0.050 -0.269 0.323 (0.209) (0.194) (0.261) (0.193) (0.306) (0.324) s_emp_ter -0.165 -0.340** -0.391** -0.225 -0.683*** -0.443 (0.166) (0.165) (0.185) (0.167) (0.230) (0.274) Constant 0.778*** 0.692*** 1.031*** 0.682*** 1.482*** 1.360*** (0.168) (0.183) (0.191) (0.171) (0.248) (0.286) Observations 17,575 8,929 8,646 7,871 5,359 4,345 R-squared 0.927 0.955 0.934 0.944 0.716 0.800 Source: Own elaborations on EU-SILC data Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Pooled sample of demographic groups for 23 EU-countries from 2006 to 2018. For the variables’ definition, see Table A4 in the Appendix. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 33 Table 6. Gender gap in other labour market outcomes and parenthood (EU 23 countries) (1) (2) (3) (4) total no_child one_child two_child_more (1): hours female -6.195*** -5.537*** -10.171*** -11.094*** (0.379) (0.406) (0.840) (0.990) (2) Full-time female -0.100*** -0.087*** -0.185*** -0.244*** (0.009) (0.008) (0.016) (0.020) (3) Permanent female -0.000 -0.002 -0.018*** -0.021*** (0.003) (0.003) (0.004) (0.004) (4) Self-employment female -0.063*** -0.060*** -0.066*** -0.067*** (0.003) (0.003) (0.004) (0.004) (5) Hourly wage female -0.075*** -0.053** -0.150*** -0.238*** (0.018) (0.021) (0.046) (0.051) (6) Hourly earnings female -0.098*** -0.078*** -0.141*** -0.198*** (0.017) (0.020) (0.037) (0.038) Source: Own elaborations on EU-SILC data Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Pooled sample of demographic groups for 23 EU-countries from 2006 to 2018. Complete estimates are available upon request. The full-time employment gap amounts to 8.7% for childless women indicating that, besides childcare, part-time positions are a crucial mean to cope with other asymmetric work burdens (as highlighted in the previous section). Panel (3) of Table 6 suggests that parenthood is the only factor preventing gender-equal access to permanent employment, as the gender gap is significant only for groups with children (2% lower permanent employment rate for women). Self-employment, especially if without employees, has been often seen as offering a potential solution to work-family conflict, due to greater flexibility and control over the timing and conditions of work (Bari et al., 2021; Goldina and Katz, 2011). As such, encouraging women's self-employment and reducing the gender gap in participation have become policy priorities at the EU level, presented as a way to improve both labour market participation of women and gender equality more broadly (European Commission, 2015; Fackelmann and De Concini, 2020; Tervo and Haapanen, 2010; Georgellis and Wall, 2005). Panel (4) of Table 6 suggests that a gender gap in self-employment exists (around 6%); however, parenthood is not associated with a significant change in the disparity. This is not the case for hourly wage and earnings (which include returns from self-employment). Parenthood emerges as a significant driver of the adjusted gender wage gap, as in the presence of one or more children the gap is three times higher (15%) and almost five times higher (24%), respectively, compared to the gender wage disparities between non-parents (5.3%). Although on a smaller scale, the same pattern is confirmed if we consider return from dependent and self-employment jointly. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 34 Figure 5 illustrates the heterogeneity in gender gaps in various labour market outcomes across four groups of countries (Southern, Continental, Northern, and Central-Eastern countries). Figure 5. Gender labour market gaps in macro-groups Source: Own elaborations on EU-SILC data Notes: The dots are the coefficients obtained using weighted OLS (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Pooled sample of demographic groups for EU countries from 2006 to 2018. The coefficients (with statistical significance level) and the standard errors of the female dummy are reported in Tables 5 and A5. Total (total sample): 23 EU countries; South (Southern European countries): Cyprus, Greece, Italy, Spain, and Portugal; Cont (Continental European countries): Austria, Belgium, France, Germany, Netherlands, and Luxembourg; North (Northern European countries): Denmark, Finland, Sweden, and Ireland; East (Eastern European countries): Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovak Republic. Despite being based on geographical proximity, the classification reflects also institutional similarities in historical-political developments. The diagrams (see Table A5 for the point estimates and statistical significance of the coefficients) confirm well-known facts in the geography of gender inequality in Europe; Southern EU countries exhibit the largest gaps in labour force participation, employment, permanent employment, self-employment, and hourly wage. As a result of low female employment rates and of a relatively less intensive diffusion of part-time contracts, gender gaps in hours worked and full-time employment are instead aligned to the average levels. Conversely, in Northern EU countries gender differences are of a lower magnitude in basically all labour marker Total South Cont North East -0.18 -0.16 -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 Labour force Total South Cont North East -0.2 -0.18 -0.16 -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 Employment Total South Cont North East -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 Hours Total South Cont North East -0.26 -0.24 -0.22 -0.2 -0.18 -0.16 -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 0 Full-time Total South Cont North East -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04 Permanent Total South Cont North East -0.12 -0.11 -0.1 -0.09 -0.08 -0.07 -0.06 -0.05 -0.04 Self-employment Total South Cont North East -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 Hourly wage Total South Cont North East -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 0.1 Hourly earnings PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 35 domains, and continental EU countries sit in an intermediate position. The highest gaps emerging in terms of full-time employment, permanent employment and hours worked are consistent with the relatively low gender gaps in employment facilitated by the extensive use of flexible contractual arrangements. Eastern EU countries stand in different relative positions depending on the specific labour market indicator considered. They have average gender gaps in employment, labour force participation and labour remunerations; however, they exhibit low gender disparities in hours worked, full-time and permanent employment and self-employment. Figure 6 (and Table A5 in the Appendix) provides a picture of the association between parenthood and labour market gender gaps in the four country groups. The two diagrams in the top panel of Figure 6 indicate that parenthood exacerbates gender gaps in labour force participation and employment in all macro-groups; however, the magnitude of the gender gap and the jump due to the presence of children is more pronounced in Southern Europe. Similarly, the wage gap is in all groups higher in the presence of children and, in the case of the Continental and the Northern EU countries, parenthood emerges as the sole driver of wage inequality. Results for the gender gap in hours worked and full-time employment highlight that the flexibility associated with such contracts is more intensively used in continental Europe countries to reconcile participation in the labour market and asymmetric childcare workloads charged on women. Lastly, parenthood seems not to be blamed, in any region of the EU, for the existing gender disparities in permanent and self-employment. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 36 Figure 6. Gender labour market gaps and parenthood in macro-groups Source: Own elaborations on EU-SILC data Notes: The bars are the coefficients of the female dummy of equations 31.-3.3, obtained using weighted OLS (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Pooled sample of demographic groups for EU countries from 2006 to 2018. The coefficients (with statistical significance level) and the standard errors of the female dummy are reported in Table A5. Total (total sample): 23 EU countries; South (Southern European countries): Cyprus, Greece, Italy, Spain, and Portugal; Cont (Continental European countries): Austria, Belgium, France, Germany, Netherlands, and Luxembourg; North (Northern European countries): Denmark, Finland, Sweden, and Ireland; East (Eastern European countries): Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovak Republic. -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Total South Cont North Ea st Labour force no child one child two child + -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Total South Cont North Ea st Employment no child one child two child + -18 -16 -14 -12 -10 -8 -6 -4 -2 0 Total South Cont North Ea st Hours no child one child two child + -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Total South Cont North Ea st Full-time no child one child two child + -0.05 -0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04 Total South Cont North Ea st Permanent no child one child two child + -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 0 Total South Cont North Ea st Self-employment no child one child two child + -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 Total South Cont North Ea st Hourly wage no child one child two child + -0.45 -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 0.05 Total South Cont North Ea st Hourly earnings no child one child two child + PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 37 3.3. Gender gaps and parenthood across household types In this section, we describe the heterogeneity in labour market gender gaps associated with parenthood in households with different characteristics. Figure 7 and Table A6 report the results of the gender dummy coefficients estimated using equations 3.1-3.3 for subsamples of demographic groups with a low/high level of the following household characteristics: household labour force participation rate, elderly dependency rate, single-parent household, disposable equivalised income, and gender of the breadwinner. The threshold for the allocation of the demographic groups into low/high subsample is based on their position on the country-year distribution (below and above the median) of the specific variable. The first piece of information we can draw from Figure 7 is that in households with low labour force participation, the gender gap in activity rates and employment is higher (top-left panels). This probably indicates that the allocation into low/high subsamples is driven by the labour market position of female household components and, particularly, of mothers. Interestingly, once the analysis is restricted to employed only, there are no significant differences in gender gaps in hours worked: this suggests that the presence of household members not in the labour market does not help alleviate the motherhood penalty in labour supply at the intensive margin. However, it exacerbates the gap in full-time employment in the presence of two children or more. Conversely, gender gaps in permanent employment and hourly wage are clearly related to the household activity rate (see columns 5 and 7 of Table A6). In low labour force participation households, the gender gap is not statistically different from zero; higher household participation rates, on the contrary, are associated with larger gender differences in hourly wages, especially for parents. This suggests that, although the presence of non-active household members does not enable higher female and mothers’ employment or labour supply, it helps attain more equal outcomes (in stable employment and remunerations) within employment. The second household characteristic considered (elderly dependency ratio) helps shed light on the way elderly household components affect parenthood gender gaps. Outcomes (second row of Figure 7 and panel 2 of Table A6) suggest that the two samples do not differ significantly in terms of gender gaps in labour supply at the extensive margin and employment. Conversely, the presence of elderly household members exacerbates the motherhood gap in hours worked, full-time and selfemployment; this suggests that they impose an additional workload disproportionately burdened on women. However, employed mothers in households with more elderly people achieve levels of permanent employment and pay comparable to their male counterparts. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 38 Figure 7. Gender labour market gaps and parenthood by household type LF par'cipa'on rate Elderly dependency ra'o Single parents Equivalised income Male breadwinner Source: Own elaborations on EU-SILC data Notes: See Figure 5. The coefficients (with statistical significance level) and the standard errors are reported in Table A6. -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Labour force no child one child two child + -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Employment no child one child two child + -14 -12 -10 -8 -6 -4 -2 0 Low High Hours no child one child two child + -0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Full-time no child one child two child + -0.08 -0.07 -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 Low High Self-employment no child one child two child + -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Labour force no child one child two child + -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Employment no child one child two child + -16 -14 -12 -10 -8 -6 -4 -2 0 Low High Hours no child one child two child + -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Full-time no child one child two child + -0.09 -0.08 -0.07 -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 Low High Self-employment no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Labour force no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Employment no child one child two child + -14 -12 -10 -8 -6 -4 -2 0 Low High Hours no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Full-time no child one child two child + -0.08 -0.07 -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 Low High Self-employment no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Labour force no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Employment no child one child two child + -14 -12 -10 -8 -6 -4 -2 0 Low High Hours no child one child two child + -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Full-time no child one child two child + -0.09 -0.08 -0.07 -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 Low High Self-employment no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Labour force no child one child two child + -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Employment no child one child two child + -16 -14 -12 -10 -8 -6 -4 -2 0 Low High Hours no child one child two child + -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0 Low High Full-time no child one child two child + -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 0 Low High Self-employment no child one child two child + PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 39 Demographic groups with a more intensive presence of single parents exhibit smaller gender gaps in labour force participation and employment, indicating that the economic consequence of the absence of a partner imposes a higher labour market attachment for women and mothers. However, single parenting exacerbates the gender gap in basically all other outcomes (particularly full-time employment, permanent employment, and wages). This suggests that, despite being obliged for economic reasons to participate in the labour market, childcare tasks impose tight constraints on single mothers and probably force them to accept lower-quality jobs. The analysis of gender gaps in subsamples of low/high-income households, despite being the economic conditions endogenous to the labour market fate and performance of female members, offer interesting descriptive insights. While gender gaps in the absence of children are substantially equal across subsamples, gender disparities associated with parenthood are systematically higher for low-income households. This suggests that in such contexts, possibly for social and cultural reasons associated with economic conditions, the child-related workload and constraints within the household are even more disproportionally placed on women. Lastly, we look at the role of the gender of the household breadwinner, as a proxy of the asymmetry in power related to unbalanced economic positions. As expected, a male breadwinner is associated with higher gender disparities, both in the presence and in the absence of children, in employment and labour supply. Surprisingly, gender inequality in job characteristics (especially permanent employment and wages) is instead lower, indicating that the relatively fewer women who enter the labour market manage to attain positions closer to their male counterparts. 4. Policies, reforms, and gender disparities in parenthood penalty across the EU 4.1. Data and methods In this section, we use data from various sources to assemble a dataset of variables that describe institutional and policy settings related to family and parenthood, with the aim of assessing their impact on parenthood penalty in a set of labour market outcomes. The empirical literature has devoted extensive attention to highlight if and to what extent the size of the child penalty depends on the architecture of parental leave and childcare systems and on the model to which the division of labour within the family is inspired (see Waldfogel, 1998a, 1998b and 2001; Haan and Wrohlich, 2009). Parental leave policies positively impact women’s employment continuity and careers only when they guarantee job security (Hegewisch and Gornick, 2011) and when the leave is paid (De PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 40 Henau et al., 2007). Its length should also be appropriate: an excessive duration keeps mothers out of employment for too long (Pettit and Hook, 2005; Jaumotte, 2003); in contrast, if it is too short, leave increases the risk of women dropping out of the labour market altogether (Keck and Saraceno, 2013). Cross-country comparisons show that paid maternity and family leave provisions of up to one year increase the likelihood of employment shortly after childbirth and have either positive or zero impacts on women’s mediumand long-run employment and earnings (Rossin-Slater, 2018). Longer paid leave entitlements can negatively affect women’s wages in the long term (Blau and Kahn, 2013) and for all skill levels (Olivetti and Petrongolo, 2017). The impact of parental leave provisions is also found to depend crucially on the availability of complementary measures, particularly formal childcare and tax/benefit systems (OECD, 2007), especially for full-time employment (Pettit and Hook, 2009). Its importance is lower where part-time jobs are more widely available (Steiber and Haas, 2012; Havnes and Mogstad, 2009). The availability of places and opening hours of kindergartens (see Jaumotte, 2003), as well as positive attitudes towards formal childcare (Hegewisch and Gornick, 2011), also play a crucial role. Asymmetries in parental leave and childcare provisions across genders still permeate virtually all societies and depends on a number of factors (Valentova et al., 2022). Even when fathers have leave opportunities like those of mothers, as in northern Europe, the gender gap in the take-up rate remains remarkable (see Thorsdottir, 2013, and Hegewisch and Gornick, 2011). Mandatory paternity leave is instead found to reduce gender imbalances in household tasks, with persistent effects after the leave period (Patnaik, 2019). Better availability of childcare facilities is only partially able to reduce the asymmetry; this translates into higher difficulties for mothers to re-enter employment and into higher part-time rates (Paull, 2008), when this is an option. Availability and fiscal incentives for part-time work may indeed represent better chances to return to employment (see Jaumotte, 2003) and the main channel through which the child penalty for mothers materializes (see Budig and England, 2001; Gangl and Ziefle, 2009; Davies and Pierre, 2005). Obviously, due to data limitations, we cannot account for the complexity of the institutional and policy environment just sketched out. However, this limitation is offset by the large geographical coverage of our analysis and by the relatively high number of policy/reform variables we were able to assemble, which are suitable to describing the heterogeneity of policies across the EU and their most important evolution over time. To this aim, we use a combination of data sources. The first one is the Labour Market Reform database (LABREF), provided by the DG for Employment, Social Affairs and Inclusion of the European Commission. LABREF is an open-access descriptive database that records labour market and welfare policy measures introduced by the EU Member States. It has become one of the standard references in the employment field, providing information PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 41 on adopted reform measures and their key design characteristics. To date, it provides information on the reform measures passed in the EU between 2000 and 2018. Figure 8. Number of countries with a policy change, by year (cumulative) Source: Own elaborations on LABREF (Labour Market Reform) data Notes: For the variables’ definition, see Table A7 in the Appendix For our 23 EU countries and over the period 2006-2018, we extracted all policy measures implemented in the policy fields “Family-related working-time organisation” and “Family-related benefits”. This resulted in a total of 111 reforms implemented, which we classified into the following groups based on the detailed description available in the dataset: (i) Expanding access to childcare (Child_care); (ii) Expanding parental leave (Par_leave); (iii) Facilitating work-life balance 0 2 4 6 8 10 12 14 16 18 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Child_care 0 2 4 6 8 10 12 14 16 18 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Par_leave 0 1 2 3 4 5 6 7 8 9 10 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Work_family_bal 0 2 4 6 8 10 12 14 16 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Gend_bal_par 0 2 4 6 8 10 12 14 16 18 20 22 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Child_support PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 48 extension of leave provisions for father enables mothers to re-enter the labour market (Thorsdottir, 2013; Patnaik, 2019). Panels 5 and 6 of Table 9 refer to the effects of aggregate public spending on family benefits and on early-stage education and care; results indicate that only the second type of public spending alleviates the parenthood gender gap, as the coefficient of the interaction terms is significantly higher for demographic groups of parents compared to the benchmark of non-parent demographic groups. The replications of the estimates of equations 4.1-4.3 with the employment rate (rather than participation rate) as the dependent variable (see Tables 10 and 11) generally confirm the outcomes of the beneficial effect of measure improving access to childcare facilities, work-family reconciliation, and gender balanced parenting. They also confirm that child support tax/benefits are not effective in reducing the motherhood penalty. Interestingly, the aggregate metric for the extensions of parental leave (panel 2 of Table 10) shows that the weak effect emerged for labour force participation disappears for employment. However, the detail on parental leave policies by gender helps clarifying the overall picture (Panels 1-4 of Table 11). Once again, a clear dichotomy exists between the effects of enhancing maternity leave and paternity leave measures which exacerbate and reduce the motherhood penalty in employment, respectively. The effects of aggregate measures of public spending (panels 5 and 6 of Table 10) on the gender parenthood gap in employment are similar to labour force participation. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 49 Table 10. Effects of reforms on the gender gap in employment (0 before the reform, 1 after the reform) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.114*** -0.071*** -0.230*** -0.271*** (0.007) (0.007) (0.008) (0.009) child_care 0.001 0.000 -0.005 -0.009 (0.005) (0.006) (0.008) (0.010) female * child_care 0.012* 0.009 0.021*** 0.020** (0.007) (0.005) (0.008) (0.010) (2) female -0.122*** -0.087*** -0.206*** -0.250*** (0.005) (0.004) (0.010) (0.012) par_leave -0.021*** -0.018*** -0.017** -0.026*** (0.004) (0.004) (0.008) (0.009) female * par_leave 0.023*** 0.022*** 0.015 0.021 (0.005) (0.004) (0.010) (0.014) (3) female -0.110*** -0.080*** -0.203*** -0.246*** (0.006) (0.005) (0.010) (0.010) work_family_bal -0.013** -0.009* -0.029*** -0.032*** (0.006) (0.005) (0.010) (0.011) female * work_family_bal 0.029*** 0.025*** 0.040*** 0.054*** (0.007) (0.007) (0.010) (0.013) (4) female -0.108*** -0.076*** -0.201*** -0.238*** (0.005) (0.005) (0.011) (0.012) gen_bal_par -0.009* -0.005 -0.020*** -0.022** (0.005) (0.004) (0.008) (0.009) female * gen_bal_par 0.023*** 0.018*** 0.023** 0.041** (0.006) (0.005) (0.012) (0.016) (5) female -0.104*** -0.073*** -0.169*** -0.208*** (0.005) (0.005) (0.009) (0.011) child_support -0.015*** -0.015*** -0.008 -0.005 (0.004) (0.004) (0.007) (0.009) female * child_support 0.026*** 0.029*** 0.005 0.002 (0.006) (0.005) (0.009) (0.013) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 50 Table 11. Effects of reforms on the gender gap in employment (continuous and ordered variables) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.064*** -0.033*** -0.130*** -0.147*** (0.004) (0.005) (0.010) (0.012) length_maternity 0.001*** 0.001*** 0.002*** 0.002*** (0.000) (0.000) (0.000) (0.001) female * length_maternity -0.001*** -0.001*** -0.003*** -0.004*** (0.000) (0.000) (0.001) (0.001) (2) female -0.103*** -0.062*** -0.200*** -0.235*** (0.004) (0.003) (0.008) (0.008) length_paternity -0.005*** -0.004*** -0.005*** -0.004*** (0.001) (0.001) (0.001) (0.001) female * length_paternity 0.006*** 0.003** 0.011*** 0.013*** (0.001) (0.001) (0.001) (0.002) (3) female -0.068*** -0.049*** -0.103*** -0.104*** (0.005) (0.005) (0.008) (0.007) paid_maternity -0.003 -0.006 0.008 0.028** (0.005) (0.005) (0.010) (0.012) female * paid_maternity -0.008*** -0.003 -0.028*** -0.039*** (0.002) (0.002) (0.003) (0.004) (4) female -0.108*** -0.065*** -0.221*** -0.266*** (0.006) (0.006) (0.010) (0.012) paid_paternity 0.004** 0.005*** -0.001 -0.010*** (0.002) (0.002) (0.002) (0.003) Female * paid_paternity 0.007*** 0.004* 0.018*** 0.025*** (0.002) (0.002) (0.003) (0.004) (5) female -0.144*** -0.110*** -0.246*** -0.301*** (0.009) (0.007) (0.015) (0.019) ps_family_ben -0.011*** -0.009*** -0.030*** -0.032*** (0.003) (0.003) (0.006) (0.006) female * ps_family_ben 0.021*** 0.021*** 0.028*** 0.034*** (0.003) (0.003) (0.005) (0.007) (6) female -0.131*** -0.088*** -0.246*** -0.321*** (0.005) (0.005) (0.010) (0.010) ps_early_ed_care -0.028*** -0.021** -0.052*** -0.049*** (0.010) (0.010) (0.013) (0.015) female* ps_early_ed_care 0.063*** 0.048*** 0.101*** 0.153*** (0.005) (0.006) (0.008) (0.009) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. Tables 12 and 13 illustrate the effects of policy reform variables on gender parenthood gaps in intensive labour supply margin (hours worked per week of those in employment). PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 51 Table 12. Effects of reforms on the gender gap in hours worked (0 before the reform, 1 after the reform) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -4.698*** -4.342*** -8.175*** -10.498*** (0.681) (0.743) (1.330) (1.858) child_care 0.310 0.465 0.398 -1.491* (0.562) (0.630) (0.816) (0.789) female * child_care -0.850 -0.657 -1.593* -0.549 (0.675) (0.720) (0.951) (1.110) (2) female -6.637*** -5.759*** -11.784*** -12.394*** (1.261) (1.191) (2.150) (2.301) par_leave -1.509* -1.404* -0.940 -2.676** (0.773) (0.773) (1.119) (1.030) female * par_leave 1.005 0.651 2.362** 2.322** (0.969) (1.051) (1.080) (0.910) (3) female -3.274*** -3.017*** -7.094*** -6.990*** (0.912) (0.972) (2.231) (1.613) work_family_bal 0.147 -0.078 -0.053 1.969 (0.865) (0.931) (1.281) (1.326) female * work_family_bal -2.447*** -2.598*** -1.708 -0.349 (0.861) (0.956) (1.287) (1.319) (4) female -5.713*** -5.308*** -7.653*** -10.696*** (0.525) (0.582) (1.133) (1.046) gen_bal_par -1.216* -0.912 -1.750* -2.833*** (0.682) (0.738) (1.033) (0.973) female * gen_bal_par 0.866 0.659 0.625 3.093*** (0.685) (0.769) (0.882) (0.978) (5) female -6.396*** -5.877*** -9.326*** -11.425*** (0.696) (0.748) (1.226) (1.400) child_support -0.611 -0.419 -0.490 -1.823** (0.624) (0.669) (0.907) (0.861) female * child_support 0.958 0.760 0.797 2.277** (0.753) (0.817) (0.940) (1.038) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. The first piece of information emerging from the summary of results is that while for labour force participation and employment many reforms impacted the magnitude of the parenthood gender gap, this is not the case for hours worked. Only the extension of parental leave clearly helps decreasing the gender gap in hours worked (panel 3 of Table 10). Once again, in combination with the more detailed indicators of maternity and paternity leave (panels 1-4 of Table 13), we conclude that the beneficial effect is limited to the extension of length and payment for fathers. However, contrary to what we observed for employment and participation, the extension of maternity leave measures is neutral, not exacerbating, the motherhood gap. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 52 Table 13. Effects of reforms on the gender gap in hours worked (continuous and ordered variables) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -6.186*** -5.767*** -10.605*** -8.290*** (0.934) (1.019) (1.257) (1.491) length_maternity 0.018 0.013 -0.044 0.076* (0.046) (0.059) (0.039) (0.044) female * length_maternity -0.018 -0.006 0.033 -0.126** (0.048) (0.053) (0.054) (0.054) (2) female -7.629*** -6.739*** -12.314*** -12.836*** (0.470) (0.508) (0.913) (1.095) length_paternity -0.155 -0.105 -0.787*** -0.165 (0.117) (0.139) (0.192) (0.155) female * length_paternity 0.434*** 0.332*** 0.769*** 0.871*** (0.073) (0.084) (0.105) (0.140) (3) female -6.176*** -5.947*** -8.039*** -9.183*** (1.031) (1.153) (1.323) (1.489) paid_maternity -2.836* -3.323* 2.607 -3.089* (1.647) (1.969) (2.579) (1.605) female * paid_maternity -0.112 0.030 -0.722* -0.426 (0.335) (0.369) (0.374) (0.388) (4) female -7.965*** -6.715*** -12.694*** -15.601*** (0.825) (0.885) (1.233) (1.399) paid_paternity -0.689** -0.599* -0.896** -1.076*** (0.293) (0.328) (0.399) (0.395) Female * paid_paternity 0.641** 0.375 1.055*** 2.062*** (0.313) (0.346) (0.365) (0.411) (5) female -4.597*** -4.283*** -8.161*** -9.349*** (1.019) (1.123) (1.534) (1.902) ps_family_ben 1.467** 1.613** 1.131 0.453 (0.616) (0.682) (0.718) (0.750) female * ps_family_ben -0.674* -0.543 -0.823* -0.582 (0.357) (0.387) (0.472) (0.571) (6) female -6.844*** -6.167*** -11.657*** -13.798*** (0.711) (0.762) (1.376) (1.548) ps_early_ed_care 2.948** 2.633* 5.419** 2.359 (1.321) (1.548) (2.553) (2.194) female* ps_early_ed_care 0.762 0.782 1.407 2.820*** (0.746) (0.814) (1.061) (1.043) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. As regards, the other reforms, gender balanced parenting and child support measures are effective in reducing the gender gap in hours worked only for mothers with two children and more. While for more gender balanced parenting measures the outcome is difficult to explain, in the case of child PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 53 support measures, a possible explanation is that only by cumulating the benefits (or tax rebates) for more children allows reaching the economic capacity to afford additional childcare services (baby-sitting, extension of hours in childcare facilities) that enable mothers to participate more intensively in employment. 4.3. Policies and motherhood penalty in employment characteristics In this section we summarize the outcomes of the estimation of equations 4.1-4.3 using as a dependent variable various metrics of labour market outcomes. All results are summarized in the set of Tables (A8–A17) placed in the Appendix. When the full-time employment rate is used as the labour outcome variable, the analysis of the effects of policies and reforms on the parenthood penalty (Table A8 and A9) supplies evidence that is, not surprisingly, largely overlapping with the one for hours worked. The only measures able to attenuate the gender gap associated to the presence of children are related to the extension of parental leave (panel 2 of Table A8); once again, the effect is limited to paternity leave length and generosity (Panels 2 and 4 of Table A9). The extension of the length of maternity leave is, differently from the case of hours worked, also inequality-reducing. However, the magnitude of the effect is smaller compared to paternity leave and independent on the presence of children or not (Panel 1 of Table A9, columns 2, 3, and 4). More generous child support measures also emerge as beneficial for closing the gender gap between mothers and fathers in full-time employment (panel 5 in Table A8). A last result worth to be highlighted is the negative sign of the coefficients of the interaction terms of the public spending variables (Panels 5 and 6 of Table A9). They indicate, not unsurprisingly, that higher spending in family benefits and early education and care services increase the gender gap in all subsamples (hence, irrespective of the presence of children). One tentative explanation, to be scrutinized in future research efforts, is that more generous family benefits increase household disposable income and, consequently, decrease labour market involvement of second earners (usually female household components) who are more willing to accept part-time positions. When we look at the parenthood gender gap in permanent employment (Tables A10 and A11), we substantially find no significant outcomes (the only exception being the detrimental effect of an increase in the generosity of maternity leave). This is not totally surprising, as we already highlighted how this is the labour market outcome in which the gender gap related to parenthood is generally low. The analysis of the effects of policy/reforms on self-employment gender gap provides richer results, substantially concentrated on parental leave measures. In general, reforms expanding parental PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 54 leave are found to decrease the gender gap. However, this happens irrespective of the presence of children (see the size of the coefficient of the interaction terms in panel 2 in Table A12). A closer look at the detail of the parental leave measures (panels 1-4 of Table A13) suggests that the result is certainly driven by paternity leave policies. However, it is interesting to note that also better paid maternity leave contributes closing the gender gap in self-employment. The interpretation of this evidence is not straightforward, as it is not possible here to distinguish which type of selfemployment is described by the data (professionals, large, small, or micro entrepreneurship). However, this is an interesting research pattern for future research, as self-employment has been identified as one of the most resilient in labour participation (Ferrín, 2023). A last, interesting set of results regards gender wage and earnings inequality. According to our estimates, with very few exceptions (work-family reconciliation policies and length of maternity leave for the subsample of two children and more) none of the policies or reforms considered has an impact on the (large) parenthood gender gaps documented in the previous section (see Tables A14-A17). This suggests that, although being important in supporting higher levels of labour market participation, employment and certain desirable job characteristics, such policies are not able to affect the labour market spheres in which the remuneration of labour is decided. Particularly, we refer to the gender asymmetry in bargaining power of employees vis-à-vis employers who implement statistical gender discrimination practices in wage setting. Possibly, institutional dimensions not considered here and more directly related to wage setting (centralisation, coordination, minimum wage regulation, social dialogue practices, etc.) have a better capacity to affect this side of the parenthood gender gap. 5. Summary and concluding remarks In this paper, we analyse the interactions between childcare, parenthood, household characteristics, labour market outcomes and institutions through the lens of gender inequality. Our contribution to the existing knowledge lies in the first place in the comparative approach of the analysis, which covers many EU countries. This is a distinctive feature of our work, as most of the studies looking at the association between parenthood and labour market outcomes are carried out for single country or for a small set of them, due to the demanding nature of (longitudinal) data required for the analysis. Our approach to overcome such constraints is to assemble a pseudo-panel dataset, in which the units of analysis are not the individuals but the demographic groups (defined by gender, age, and education) for a large set of countries and a relatively long period of time. A second distinctive feature of our work is the analysis of how a large set of policies related to family and childcare moderate the relationship between parenthood and labour market outcomes in the EU. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 55 The report is organised into three sections. In the first one, we have used HETUS data to provide a cross-country descriptive analysis of the disparities in time allocation to paid work, housework and childcare within the household. Employing micro-level data from ten European countries (year 2010), we analyse the scale and cross-country variation in within-couple gender gaps in time allocation in relation to a broad range of spousal and household characteristics and of country-level indicators of gender equality. Our outcomes suggest the existence of systematic specialization patterns, with wives spending less time on employment and more time on housework and childcare. Wife’s relative worktime converges to 0.5 in dual-earner couples but declines with larger household sizes. Both in absolute and relative terms, worktime decreases remarkably in the presence of children; female spouses restrict on average their daily working time by one hour in the presence of one child and by almost two hours when there are two or more children in the household. Conversely, the husband’s working time remains virtually unchanged. This evidence descriptively confirms the existence of a remarkable asymmetry in the labour market parenthood penalty across genders. Couples in which the husband is older or more educated than his wife also achieve notably less gender equality in worktime allocation. However, all types of couples appear far from within-couple gender equality in housework and childcare, as the wife’s relative housework remains around or significantly over 60% and close to 70%, respectively, for all household types. The country-specific analysis confirms that significant gender asymmetries exist in all countries in our sample, but their magnitude differs. Specifically, as regards worktime, Finland appears the most gender-equal country in terms of time allocated in employment; in Estonia the gender disparity in relative worktime turns statistically insignificant upon controlling for a full set of individual and household characteristics. As for housework and childcare, in all sample countries the time wives invest in housework exceeds the time invested by their husbands enormously, even when demographic, employment and household characteristics are controlled for, and relative worktime is more balanced. One notable exception, with reference to childcare, is Finland, where the time allocation if gender balanced. Lastly, our analysis reveals that larger country-level gender asymmetries in worktime and housework are correlated with various metrics of gender equality in the labour market and in society. The association between asymmetries in childcare and such macroeconomic indicators is more nuanced. In the second part of the analysis, we focus on gender asymmetries in labour market performance related to parenthood. To this purpose, we have employed microdata from the European Union Survey on Income and Living Conditions (EU-SILC) to assemble a longitudinal dataset at the demographic group (gender, age, education) level for twenty-three EU countries over the period 2006-2018. We first provide a measure of the asymmetries across genders in parenthood penalty/premium in various labour market outcomes: labour force participation, employment, hours PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 56 worked per week, job characteristics. We then map this gender disparity across groups of countries and household characteristics that describe different household models. Our results for the whole sample of EU-countries, consistently with a large body of empirical evidence, indicate that parenthood implies a labour market participation and employment premium for fathers and, conversely, a penalty for mothers. This means that the presence of children exacerbates labour market gender inequalities and accounts for a significant share of the observed disparities. This applies to labour supply (at both the extensive and the intensive margins), employment, full-time employment, permanent employment, hourly wage, and hourly earnings. The analysis of the parenthood penalty for sub-groups of countries confirms well-known facts in the geography of gender inequality in Europe. Southern EU countries exhibit the largest gaps in labour force participation, employment, permanent employment, self-employment, and hourly wage. Conversely, in Northern EU countries gender differences are of a lower magnitude in basically all labour marker domains, and continental EU countries sit in an intermediate position. Eastern EU countries stand in different relative positions depending on the specific labour market indicator considered: they have average gender gaps in employment, labour force participation and labour remunerations. However, they exhibit low gender disparities in hours worked, full-time, permanent employment and self-employment. As regards gender inequality and parenthood penalty across household types, our evidence reveals that households with low labour force participation exhibit higher gender gaps in activity rates and employment in the presence of children. Conversely, higher household participation rates are associated with larger gender differences in permanent employment and wages, especially between parents. The presence of elderly household members is found to exacerbate the motherhood gap in hours worked, full-time and self-employment; however, employed mothers in households with more elderly people achieve levels of permanent employment and pay comparable to their male counterparts. Single parenting is associated to smaller gender gaps in labour force participation and employment, indicating that the economic consequence of the absence of a partner imposes a higher labour market attachment for women and mothers. At the same time, it exacerbates the gender gap in basically all other outcomes. This suggests that, despite being obliged for economic reasons to participate in the labour market, childcare tasks impose tight constraints on single mothers and probably force them to accept lower-quality jobs. As for the analysis for subsamples of low/high-income households, we find that gender disparities associated with parenthood are systematically higher for low-income households. This suggests that in such contexts, possibly for social and cultural reasons associated with economic conditions, the childrelated workload and constraints within the household are even more disproportionally placed on women. Lastly, household in which the breadwinner is a man exhibit higher gender disparities, both PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 57 in the presence and in the absence of children, in employment and labour supply; gender inequality in job characteristics (especially permanent employment and wages) is instead lower. In the third part of the report, we analyse how gender asymmetries in the labour market effects of parenthood are moderated (or not) by an array of family-related public policies. To this aim, we assemble a country-level policy/reform dataset composed of: (i) binary indicators (pre-and postintroduction) of specific measures related to access to childcare, parental leave, work-family reconciliation, gender-balanced parenting and child tax/benefits; (ii) ordered or continuous variables that describe the length and generosity of maternity and paternity leave, public spending on family benefits and on early education and childcare. Our evidence suggests that most policies included in our analysis have in general a better capacity to reduce gender gaps in parenthood penalty in labour supply and employment, rather than in job characteristics. This is particularly the case for measures aimed at increasing access to childcare facilities, favouring work-family reconciliation, and promoting gender-balanced parenting. As regards parental leave policies, a clear dichotomy emerges between the moderating effects of paternity and maternity leave policies. Only extensions of the length and generosity of paternity leave is able to reduce the parenthood gender gap in labour force participation, employment and hours worked; conversely, longer and better paid parental leave for mothers exacerbate the gendered effects of parenthood (in the case of labour force participation and employment) or play a neutral role (on labour supply at the intensive margin). The impact of family policies/reforms on parents’ gender gap in full time employment is, not surprisingly, like the case of hours worked. However, as a distinctive feature, higher spending in family benefits and early education and care services increase the full-time employment gender gap in all subsamples (hence, irrespective of the presence of children). One tentative explanation, to be scrutinized in future research efforts, is that more generous family benefits increase household disposable income and, consequently, decrease labour market involvement of second earners (usually female household components) who are more willing to accept part-time positions. The analysis of the effects of policy/reforms on self-employment gender gap reveals that reforms expanding parental leave are found to decrease the gender gap. However, this happens irrespective of the presence of children, and it is mainly driven by paternity leave policies. However, it is interesting to note that also better paid maternity leave contributes closing the gender gap in selfemployment, which addresses towards a further research effort on the effects of different types of self-employment. Lastly, we find no significant impact of policies and reforms on the parents’ gender gap in labour remunerations. This suggests that the measures considered here are not able to affect the gender asymmetries in wage bargaining, which are probably more sensitive to institutional features more directly related to wage setting. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 64 Appendix Table A1. Tobit regression results for relative worktime, by country (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Country BE DE EE EL FI FR LU PL RO UK Gender gap -0.046*** -0.056*** -0.022** -0.057*** -0.002 -0.032*** -0.093*** -0.035*** -0.045*** -0.033*** (0.009) (0.007) (0.010) (0.008) (0.010) (0.004) (0.010) (0.004) (0.003) (0.011) Age (base: 20-24) 25-29 -0.011 0.007 0.026 -0.015 0.012 -0.008 -0.009 -0.025* -0.003 0.022 (0.037) (0.037) (0.041) (0.062) (0.066) (0.023) (0.051) (0.015) (0.013) (0.109) 30-34 0.025 0.026 0.054 -0.075 0.014 -0.017 -0.044 -0.020 0.015 0.010 (0.038) (0.038) (0.040) (0.052) (0.066) (0.023) (0.050) (0.015) (0.013) (0.108) 35-39 0.016 0.048 -0.007 -0.050 0.057 -0.018 -0.019 -0.022 0.019 -0.011 (0.037) (0.037) (0.040) (0.052) (0.065) (0.023) (0.050) (0.015) (0.013) (0.109) 40-44 0.012 0.040 0.004 -0.047 0.063 -0.013 0.023 -0.027* 0.014 0.024 (0.037) (0.037) (0.040) (0.052) (0.065) (0.023) (0.051) (0.015) (0.013) (0.108) 45-49 -0.012 0.037 0.013 -0.053 0.056 -0.017 -0.001 -0.021 0.012 0.003 (0.037) (0.037) (0.040) (0.052) (0.065) (0.023) (0.051) (0.015) (0.013) (0.108) 50-54 0.007 0.046 0.006 -0.050 0.043 -0.017 0.003 -0.015 0.008 -0.003 (0.037) (0.037) (0.040) (0.052) (0.065) (0.023) (0.051) (0.015) (0.013) (0.109) 55-59 0.001 0.049 0.001 -0.045 0.034 -0.019 -0.043 -0.021 0.010 0.007 (0.037) (0.037) (0.041) (0.054) (0.065) (0.023) (0.054) (0.016) (0.013) (0.108) 60-64 0.027 0.013 -0.001 -0.041 0.040 -0.028 0.115 -0.048** -0.023 -0.001 (0.049) (0.038) (0.046) (0.117) (0.065) (0.027) (0.092) (0.020) (0.019) (0.110) Migration status (base: foreignborn) Born in country -0.001 -0.006 -0.007 0.012 0.073* -0.002 0.023** -0.003 0.008 -0.014 (0.016) (0.013) (0.012) (0.014) (0.040) (0.010) (0.010) (0.062) (0.033) (0.018) Household size (base: 2 persons) 3 persons 0.002 0.010 -0.006 -0.016 -0.005 -0.000 0.009 0.002 -0.002 0.005 (0.013) (0.008) (0.015) (0.018) (0.017) (0.006) (0.016) (0.006) (0.004) (0.015) 4 persons 0.003 0.009 -0.002 -0.022 -0.013 0.003 0.003 0.002 -0.002 0.003 (0.014) (0.011) (0.017) (0.021) (0.024) (0.008) (0.017) (0.007) (0.004) (0.018) 5 and more persons 0.013 0.011 0.003 -0.023 -0.008 0.002 0.011 0.004 -0.003 0.009 (0.018) (0.015) (0.024) (0.030) (0.033) (0.010) (0.023) (0.009) (0.005) (0.026) Education (base: low) Education medium 0.009 0.001 0.030 -0.043*** -0.027* -0.003 -0.021* 0.013 0.011* 0.011 (0.010) (0.019) (0.025) (0.013) (0.015) (0.005) (0.012) (0.010) (0.006) (0.030) Education high 0.009 0.015 0.003 -0.031** -0.018 0.013** 0.006 0.012 0.001 0.010 (0.011) (0.019) (0.026) (0.014) (0.016) (0.007) (0.013) (0.011) (0.007) (0.030) Number of children aged 0 to 6 (base: 0) 1 child -0.014 0.004 -0.013 0.009 0.009 -0.004 0.008 0.004 0.002 -0.001 PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 65 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Country BE DE EE EL FI FR LU PL RO UK (0.015) (0.009) (0.016) (0.015) (0.017) (0.006) (0.014) (0.005) (0.004) (0.016) 2 and more children -0.017 0.008 -0.019 0.031 0.015 0.000 0.015 0.009 0.002 0.008 (0.019) (0.015) (0.035) (0.025) (0.028) (0.011) (0.020) (0.009) (0.008) (0.022) Number of children aged 7 to 17 (base: 0) 1 child -0.004 0.002 0.000 0.009 -0.000 -0.003 -0.010 -0.001 -0.004 -0.005 (0.012) (0.009) (0.013) (0.015) (0.017) (0.006) (0.012) (0.005) (0.003) (0.015) 2 and more children -0.011 0.005 0.006 0.020 -0.000 -0.001 -0.014 -0.001 -0.003 -0.004 (0.015) (0.012) (0.020) (0.019) (0.025) (0.009) (0.018) (0.008) (0.005) (0.021) Household net income band (base: <P20) P20 to P40 0.014 0.022 0.000 -0.014 0.061 0.022 -0.007 -0.010 0.002 -0.002 (0.048) (0.031) (0.017) (0.075) (0.062) (0.023) (0.026) (0.021) (0.012) (0.039) P40 to P60 0.010 0.004 0.006 -0.010 0.034 0.003 -0.015 -0.013 0.001 -0.011 (0.046) (0.013) (0.018) (0.059) (0.033) (0.007) (0.025) (0.020) (0.008) (0.037) P60 to P80 -0.001 0.007 0.005 -0.001 0.006 0.003 -0.012 -0.013 -0.002 -0.015 (0.046) (0.006) (0.017) (0.058) (0.011) (0.004) (0.025) (0.020) (0.007) (0.037) >P80 -0.003 0.000 0.005 -0.003 0.000 0.000 -0.025 -0.014 -0.001 -0.013 (0.046) (.) (0.021) (0.059) (.) (.) (0.025) (0.021) (0.007) (0.036) Employment type (base: part-time) Full time 0.027*** 0.061*** 0.037** 0.020 0.058*** 0.022*** 0.032*** 0.030*** 0.060*** 0.029** (0.009) (0.007) (0.018) (0.015) (0.020) (0.005) (0.012) (0.009) (0.013) (0.012) Industry (base: Other community, social & personal service) Agriculture, fishing, mining & quarrying, utility supply 0.039 -0.030* 0.044 -0.021 -0.003 -0.015 -0.017 -0.020 -0.010 -0.055 (0.033) (0.018) (0.031) (0.038) (0.033) (0.015) (0.033) (0.014) (0.009) (0.039) Manufacturing and construction 0.070*** -0.009 0.012 -0.011 0.003 0.000 0.008 -0.007 -0.004 -0.023 (0.025) (0.013) (0.028) (0.025) (0.023) (0.010) (0.020) (0.012) (0.008) (0.030) Wholesale and retail trade 0.062** -0.019 0.035 0.013 0.022 0.015 0.043* 0.001 0.000 -0.043 (0.026) (0.015) (0.029) (0.025) (0.024) (0.010) (0.022) (0.013) (0.008) (0.031) Hotels and restaurants, transport, storage and communication 0.022 0.000 0.051* 0.038 0.005 -0.011 0.044** -0.013 -0.006 -0.038 (0.026) (0.014) (0.029) (0.025) (0.025) (0.010) (0.021) (0.013) (0.008) (0.030) Financial intermediation; real estate, renting and business activities 0.050* -0.011 0.064** 0.038 0.032 -0.001 0.033 -0.008 -0.001 -0.035 (0.026) (0.014) (0.029) (0.027) (0.024) (0.010) (0.021) (0.013) (0.009) (0.031) Education, health and social work 0.054** -0.027** 0.027 0.028 0.011 -0.011 0.009 -0.013 -0.011 -0.012 (0.025) (0.013) (0.028) (0.024) (0.022) (0.010) (0.021) (0.012) (0.008) (0.029) Public administration, defense, social security, extra-territorial bodies 0.036 -0.022 0.045 -0.019 0.021 -0.001 0.015 -0.017 0.006 -0.013 (0.026) (0.013) (0.031) (0.025) (0.025) (0.010) (0.021) (0.013) (0.008) (0.032) N 790 1632 632 420 600 2992 426 3132 3752 694 Pseudo R-sq -0.095 -0.192 -0.067 -0.135 -0.038 -0.048 -0.265 -0.029 -0.061 -0.043 Notes: Tobit regression estimates based on HETUS wave 2010 data. Dependent variable is relative housework censored at 0 and 1. All models additionally control for year, month and day of a week fixed effects. The estimates account for combined individual response and day weight. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 66 Table A2. Tobit regression results for relative housework, by country (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Country BE DE EE EL FI FR LU PL RO UK Gender gap 0.319*** 0.231*** 0.318*** 0.610*** 0.134*** 0.262*** 0.404*** 0.302*** 0.496*** 0.269*** (0.024) (0.016) (0.027) (0.026) (0.023) (0.012) (0.032) (0.010) (0.009) (0.026) Age (base: 20-24) 25-29 -0.166* 0.090 -0.161 0.066 -0.046 0.022 0.290* -0.009 -0.063 -0.175 (0.096) (0.088) (0.111) (0.193) (0.157) (0.069) (0.158) (0.039) (0.044) (0.273) 30-34 -0.137 0.073 -0.024 0.143 0.034 0.055 0.422*** -0.003 -0.099** -0.098 (0.099) (0.089) (0.109) (0.163) (0.157) (0.068) (0.156) (0.038) (0.043) (0.272) 35-39 -0.138 0.064 0.046 0.037 -0.035 0.036 0.368** -0.000 -0.087** -0.046 (0.097) (0.087) (0.108) (0.162) (0.156) (0.069) (0.155) (0.039) (0.043) (0.272) 40-44 -0.145 0.051 0.045 0.068 -0.049 -0.011 0.244 -0.005 -0.096** -0.156 (0.097) (0.087) (0.108) (0.161) (0.155) (0.068) (0.157) (0.039) (0.043) (0.271) 45-49 -0.081 0.065 0.057 0.045 -0.010 0.024 0.342** 0.017 -0.074* -0.115 (0.098) (0.087) (0.109) (0.162) (0.156) (0.068) (0.158) (0.039) (0.044) (0.271) 50-54 -0.085 0.056 -0.026 0.084 -0.008 0.048 0.339** -0.021 -0.084* -0.024 (0.096) (0.087) (0.108) (0.162) (0.155) (0.068) (0.158) (0.040) (0.044) (0.272) 55-59 -0.127 0.003 0.029 0.124 -0.017 0.019 0.464*** -0.004 -0.098** -0.075 (0.098) (0.088) (0.110) (0.166) (0.156) (0.069) (0.166) (0.041) (0.046) (0.271) 60-64 -0.103 0.082 -0.047 -0.107 -0.006 0.049 0.025 0.012 -0.047 -0.083 (0.129) (0.089) (0.124) (0.363) (0.156) (0.081) (0.285) (0.053) (0.064) (0.275) Migration status (base: foreignborn) Born in country 0.050 0.050* 0.000 0.006 -0.195** -0.005 -0.035 -0.020 0.021 -0.012 (0.041) (0.030) (0.033) (0.044) (0.096) (0.030) (0.032) (0.163) (0.114) (0.044) Household size (base: 2 persons) 3 persons -0.007 -0.024 -0.010 0.015 0.006 0.006 -0.046 -0.010 0.001 -0.006 (0.034) (0.019) (0.040) (0.055) (0.041) (0.018) (0.050) (0.016) (0.013) (0.037) 4 persons -0.026 -0.032 -0.030 0.004 0.021 0.002 -0.041 -0.017 0.003 -0.001 (0.037) (0.026) (0.047) (0.064) (0.057) (0.024) (0.052) (0.019) (0.015) (0.045) 5 and more persons -0.034 -0.028 -0.032 0.003 0.026 0.001 -0.062 -0.023 0.006 -0.026 (0.048) (0.035) (0.065) (0.092) (0.078) (0.031) (0.072) (0.023) (0.019) (0.065) Education (base: low) Education medium 0.032 -0.054 -0.019 0.048 0.017 0.015 0.045 -0.048* -0.002 -0.026 (0.027) (0.045) (0.068) (0.041) (0.036) (0.015) (0.036) (0.026) (0.021) (0.074) Education high 0.013 -0.086* 0.077 0.003 0.016 -0.043** -0.018 -0.060** 0.008 -0.018 (0.030) (0.045) (0.070) (0.045) (0.037) (0.020) (0.042) (0.028) (0.022) (0.074) Number of children aged 0 to 6 (base: 0) 1 child 0.035 -0.006 0.032 -0.003 -0.006 0.002 0.005 0.010 -0.002 0.016 (0.038) (0.022) (0.043) (0.047) (0.040) (0.018) (0.042) (0.014) (0.013) (0.039) 2 and more children 0.053 -0.012 0.022 -0.023 -0.020 0.007 0.013 0.003 -0.008 0.017 (0.051) (0.035) (0.095) (0.076) (0.067) (0.032) (0.063) (0.024) (0.027) (0.055) Number of children aged 7 to 17 (base: 0) PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 67 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Country BE DE EE EL FI FR LU PL RO UK 1 child 0.017 -0.008 -0.019 0.005 -0.012 0.010 0.028 -0.001 0.003 0.017 (0.032) (0.021) (0.036) (0.045) (0.041) (0.018) (0.038) (0.013) (0.011) (0.037) 2 and more children 0.035 -0.010 -0.026 -0.001 -0.011 0.017 0.057 0.006 0.000 0.039 (0.040) (0.029) (0.054) (0.060) (0.060) (0.026) (0.056) (0.020) (0.018) (0.053) Household net income band (base: <P20) P20 to P40 0.005 -0.036 -0.015 0.013 0.043 -0.040 0.037 0.017 0.001 0.034 (0.126) (0.073) (0.046) (0.234) (0.149) (0.069) (0.079) (0.055) (0.042) (0.097) P40 to P60 -0.014 -0.003 -0.027 -0.005 -0.035 -0.008 0.041 0.014 -0.003 0.024 (0.122) (0.030) (0.050) (0.184) (0.080) (0.021) (0.076) (0.054) (0.027) (0.093) P60 to P80 -0.006 -0.013 -0.027 -0.013 -0.000 -0.020 0.025 0.022 0.003 0.024 (0.121) (0.014) (0.046) (0.181) (0.026) (0.012) (0.078) (0.054) (0.024) (0.092) >P80 -0.009 0.000 -0.035 0.013 0.000 0.000 0.057 0.022 0.002 0.018 (0.122) (.) (0.056) (0.184) (.) (.) (0.078) (0.054) (0.023) (0.090) Employment type (base: part-time) Full time -0.020 -0.130*** -0.032 -0.039 0.005 -0.032** -0.051 -0.032 -0.065 -0.028 (0.024) (0.017) (0.050) (0.045) (0.049) (0.015) (0.037) (0.023) (0.044) (0.031) Industry (base: Other community, social & personal service) Agriculture, fishing, mining & quarrying, utility supply 0.007 0.024 -0.064 0.092 -0.054 0.066 -0.099 0.040 -0.025 0.256*** (0.088) (0.042) (0.085) (0.119) (0.079) (0.046) (0.103) (0.035) (0.029) (0.099) Manufacturing and construction -0.061 0.019 -0.026 0.081 -0.027 0.027 -0.077 -0.009 -0.038 0.020 (0.066) (0.031) (0.076) (0.077) (0.055) (0.029) (0.061) (0.032) (0.026) (0.075) Wholesale and retail trade -0.107 0.028 -0.045 0.020 -0.026 -0.051* -0.026 -0.038 -0.061** -0.054 (0.068) (0.036) (0.079) (0.078) (0.058) (0.031) (0.069) (0.033) (0.027) (0.076) Hotels and restaurants, transport, storage and communication -0.047 -0.008 -0.096 -0.018 -0.057 0.021 -0.143** -0.000 -0.050* -0.005 (0.068) (0.034) (0.079) (0.077) (0.059) (0.031) (0.065) (0.033) (0.028) (0.076) Financial intermediation; real estate, renting and business activities -0.083 0.054 -0.124 -0.067 0.003 0.003 -0.097 0.002 -0.017 0.013 (0.068) (0.033) (0.080) (0.083) (0.058) (0.031) (0.064) (0.033) (0.029) (0.076) Education, health and social work -0.070 0.054* -0.089 -0.036 0.000 0.048* -0.073 0.021 -0.015 0.004 (0.065) (0.031) (0.076) (0.076) (0.053) (0.029) (0.064) (0.033) (0.028) (0.073) Public administration, defense, social security, extra-territorial bodies -0.052 0.054* -0.042 0.038 -0.109* 0.019 -0.049 0.002 -0.070** 0.025 (0.068) (0.031) (0.084) (0.078) (0.060) (0.031) (0.065) (0.034) (0.029) (0.080) N 790 1632 632 420 600 2992 426 3132 3752 694 Pseudo R-sq 0.541 0.761 0.323 0.700 0.446 0.189 0.809 0.475 0.643 0.361 Notes: Tobit regression estimates based on HETUS wave 2010 data. Dependent variable is relative housework censored at 0 and 1. All models additionally control for year, month and day of a week fixed effects. The estimates account for combined individual response and day weight. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 68 Table A3. Sample countries macro-level indicators of gender equality over years when HETUS wave 2010 was conducted Country Year Gender gap in employment, `pp Gender gap in part-time employment, pp Children under 3 y.o. in childcare, % Females in top management positions, % Females in top governmental positions, % Unadjusted gender wage gap, % BE 2013 10,20 31,60 46,00 16,70 39,70 7,50 DE 2013 9,50 43,50 28,00 21,50 35,70 22,10 EE 2009 10,40 6,20 25,00 6,40 21,80 27,60 EL 2013 19,50 7,80 14,00 8,40 21,00 15,00 FI 2009 2,20 10,20 27,00 23,60 40,00 20,80 FR 2010 7,90 24,90 43,00 12,30 20,40 15,60 LU 2014 12,90 30,90 49,00 11,70 28,30 5,40 PL 2013 13,70 6,00 5,00 12,30 22,30 7,10 RO 2012 17,20 -0,80 15,00 7,80 9,90 6,90 UK 2014 9,80 12,00 28,90 24,20 23,70 20,90 Sources: Macro-level indicators are available at https://ec.europa.eu/eurostat/web/main/data/database Notes: The indicators are defined as (i) male-female gap in employment; (ii) female-male gap in part-time employment; (iii) male-female unadjusted wage gap; (iv) the percentage of children (under 3 years old) cared for by formal arrangements other than by the family; (v) share of female board members and executives in the largest publicly listed companies; (vi) the proportion of women in national parliaments and national governments PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 69 Table A4. Children and labour market outcomes: variables and definitions Variable Definition Individual characteristics (Defining the demographic group - dg) female Gender (0/1) prim_edu Primary Education (0/1) sec_educ Secondary Education (0/1) ter_educ Tertiary Education (0/1) age20_29 20 to 29 years-old (0/1) age30_39 30 to 39 years-old (0/1) age40_49 40 to 49 years-old (0/1) age50_59 50 to 59 years-old (0/1) age60_ 60 years-old and more (0/1) no_child No children (0/1) two_child_more One child (0/1) two_child_more Two children or more (0/1) Labour market outcomes lab force In the labour force (% of the dg population) employed In employment (% of the dg population) hours Number of hours worked per week (dg average of employed individuals) self Self-employed (% of the dg total employment) Full_time Employed on a full-time contract (% of the dg total employment) Permanent Employed on a permanent contract (% of the dg total employment) hwage Real hourly wage (average of real hourly wage of employees in the dg) hearnings Real hourly earnings from dependent work or self-employment (dg average of real hourly earnings) Other individual characteristics health Self-perceived general health (dg average) (ranging from 1 – very good to 5 very bad) migrant Born outside the country of residence (% of the dg population) partner_house Married or in a consensual union (% of the dg population) respond Person responding the household questionnaire (% of the dg population) Household characteristics rel_disp_eq_income Disposable equivalised household income relative to the country/year average (dg average) nhousehold Household size (dg average) household_d Household dependency ratio (n of 0-14 y.o. + n of 65+/n of 15-64) (dg average) elderly_d Elderly dependency ratio (n of 65+/n of 15-64) (dg average) n_care_hh Total months spent in domestic/caregiving tasks in the household (dg average) labour force participation rate Share of adult household components in the labour force (dg average) single_parent Households in which children are associated to only one parent (%in the dg) breadwinner_m Household with a male breadwinner (%in the dg) Country-level controls ur Unemployment rate lr_pc_gdp Real per capita GDP (log) s_emp_sec Share of employment in the secondary sector s_emp_ter Share of employment in the tertiary sector Notes: Elaborations from EU-SILC data and Eurostat data (country-level indicators) PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 70 Table A5. Gender gap in labour market outcomes and parenthood (macro groups) total no_child one_child two_child_more (1) Labour force South -0.144*** -0.120*** -0.232*** -0.274*** (0.008) (0.006) (0.014) (0.014) Cont -0.098*** -0.070*** -0.140*** -0.196*** (0.004) (0.004) (0.008) (0.010) North -0.068*** -0.051*** -0.100*** -0.125*** (0.005) (0.006) (0.009) (0.011) East -0.104*** -0.101*** -0.184*** -0.136*** (0.004) (0.004) (0.014) (0.024) (2) Employment South -0.142*** -0.104*** -0.256*** -0.297*** (0.008) (0.007) (0.014) (0.013) Cont -0.091*** -0.057*** -0.151*** -0.206*** (0.005) (0.005) (0.008) (0.011) North -0.046*** -0.024*** -0.112*** -0.133*** (0.003) (0.005) (0.010) (0.011) East -0.084*** -0.082*** -0.217*** -0.190*** (0.004) (0.004) (0.013) (0.019) (3) Hours South -7.178*** -6.617*** -10.450*** -10.885*** (0.776) (0.855) (1.668) (1.639) Cont -9.765*** -8.672*** -15.657*** -16.915*** (0.722) (0.801) (1.327) (1.733) North -5.127*** -4.002*** -5.191*** -6.646*** (0.743) (0.913) (1.831) (1.711) East -3.720*** -4.215*** -4.009*** -4.109** (0.507) (0.692) (1.207) (1.866) (4) Full-time South -0.080*** -0.066*** -0.165*** -0.200*** (0.009) (0.007) (0.019) (0.021) Cont -0.200*** -0.170*** -0.311*** -0.406*** (0.013) (0.009) (0.033) (0.026) North -0.100*** -0.080*** -0.122*** -0.170*** (0.016) (0.017) (0.019) (0.032) East -0.028*** -0.032*** -0.031*** -0.045*** (0.003) (0.004) (0.005) (0.008) (5) Permanent South -0.030*** -0.033*** -0.004 -0.009 (0.005) (0.006) (0.008) (0.008) Cont -0.036*** -0.040*** -0.029*** -0.041*** (0.005) (0.006) (0.007) (0.006) North -0.012*** -0.011*** -0.006 0.008 (0.003) (0.004) (0.008) (0.006) East 0.026*** 0.033*** -0.007 -0.002 (0.005) (0.007) (0.006) (0.011) (6) Self-employment South -0.098*** -0.101*** -0.070*** -0.081*** (0.005) (0.006) (0.009) (0.006) Cont -0.061*** -0.058*** -0.062*** -0.072*** (0.003) (0.004) (0.005) (0.005) PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 71 North -0.081*** -0.066*** -0.060*** -0.058*** (0.005) (0.008) (0.010) (0.010) East -0.051*** -0.048*** -0.074*** -0.022** (0.003) (0.004) (0.007) (0.009) (7) Hourly wage South -0.151** -0.160** -0.233** -0.336** (0.059) (0.070) (0.115) (0.132) Cont -0.035 0.005 -0.055 -0.314*** (0.047) (0.059) (0.090) (0.106) North -0.050 -0.021 -0.129 -0.284** (0.055) (0.073) (0.122) (0.120) East -0.076*** -0.095*** -0.304*** -0.384* (0.029) (0.031) (0.091) (0.198) (8) Hourly earnings South -0.194*** -0.193*** -0.102 -0.215** (0.041) (0.049) (0.101) (0.106) Cont -0.196*** -0.153* -0.129 -0.413*** (0.071) (0.086) (0.079) (0.086) North -0.016 0.004 -0.069 -0.106 (0.044) (0.056) (0.090) (0.111) East -0.091*** -0.121*** -0.265*** -0.166 (0.030) (0.033) (0.082) (0.105) Source: Own elaborations on EU-SILC data Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. South (Southern European countries): Cyprus, Greece, Italy, Spain, and Portugal; Cont (Continental European countries): Austria, Belgium, France, Germany, Netherlands, and Luxembourg; North (Northern European countries): Denmark, Finland, Sweden, and Ireland; East (Eastern European countries): Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovak Republic. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 72 Table A6. Gender gap in labour market outcomes and household characteristics (1) Labour foce (2) Employment (3) Hours (4) Full-time (5) Permanent (6) Self-empl (7) Hourly wage (8) Hourly earn (1) LF part rate Low No child -0.097*** -0.075*** -5.737*** -0.101*** -0.001 -0.062*** -0.035 -0.049* One_child -0.175*** -0.189*** -9.758*** -0.174*** -0.008 -0.073*** -0.081 -0.104* Two child + -0.357*** -0.341*** -12.267*** -0.355*** 0.015 -0.062*** -0.360 -0.227* High No child -0.034*** -0.026*** -5.224*** -0.081*** -0.003 -0.055*** -0.069** -0.064*** One_child -0.143*** -0.170*** -10.766*** -0.197*** -0.030*** -0.055*** -0.278*** -0.233*** Two child + -0.180*** -0.201*** -10.958*** -0.226*** -0.026*** -0.066*** -0.202*** -0.179*** (2) Elder dep rate Low No child -0.111*** -0.098*** -3.150*** -0.032*** -0.004 -0.046*** 0.002 0.062 One_child -0.167*** -0.189*** -9.960*** -0.192*** -0.018*** -0.065*** -0.153*** -0.140*** Two child + -0.201*** -0.216*** -10.634*** -0.245*** -0.019*** -0.068*** -0.256*** -0.215*** High No child -0.077*** -0.055*** -5.687*** -0.093*** -0.001 -0.062*** -0.049** -0.083*** One_child -0.159*** -0.163*** -11.588*** -0.168*** -0.017* -0.082*** -0.070 -0.032 Two child + -0.199*** -0.217*** -14.123*** -0.294*** -0.018 -0.077*** -0.031 -0.039 (3) Single parent Low No child -0.065*** -0.045*** -5.736*** -0.092*** 0.002 -0.061*** -0.056** -0.090*** One_child -0.240*** -0.242*** -6.882** -0.153*** 0.013 -0.043* -0.425** -0.216 Two child + -0.245*** -0.259*** -12.087*** -0.116*** 0.001 -0.062*** -0.238 -0.080 High No child -0.128*** -0.102*** -4.308*** -0.064*** -0.007 -0.053*** -0.054 -0.052 One_child -0.162*** -0.180*** -10.775*** -0.203*** -0.014*** -0.070*** -0.148*** -0.159*** Two child + -0.182*** -0.196*** -11.685*** -0.270*** -0.021*** -0.070*** -0.278*** -0.264*** (4) Equiv income Low No child -0.092*** -0.062*** -5.759*** -0.098*** -0.000 -0.047*** -0.015 0.022 One_child -0.205*** -0.233*** -12.081*** -0.226*** -0.006 -0.082*** -0.234*** -0.207*** Two child + -0.256*** -0.280*** -13.180*** -0.306*** -0.013** -0.080*** -0.290*** -0.255*** High No child -0.075*** -0.059*** -5.668*** -0.074*** 0.001 -0.066*** -0.095*** -0.132*** One_child -0.119*** -0.134*** -8.375*** -0.146*** -0.020*** -0.051*** -0.124** -0.113** PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 73 Two child + -0.149*** -0.161*** -9.035*** -0.195*** -0.026*** -0.059*** -0.155** -0.141** (5) Breadw male Low No child -0.082*** -0.064*** -5.407*** -0.074*** -0.009** -0.070*** -0.085*** -0.118*** One_child -0.148*** -0.162*** -9.690*** -0.171*** -0.016*** -0.070*** -0.206*** -0.177*** Two child + -0.177*** -0.193*** -11.959*** -0.275*** -0.022*** -0.075*** -0.266*** -0.209*** High No child -0.180*** -0.178*** -14.095*** -0.143*** 0.046* -0.118*** -0.157 -0.092 One_child -0.237*** -0.265*** -15.137*** -0.327*** -0.020** -0.067*** 0.013 -0.039 Two child + -0.248*** -0.265*** -9.793*** -0.099 -0.011 -0.045*** -0.211** -0.145* (6) Breadw female Low No child -0.087*** -0.072*** -7.588*** -0.161*** -0.005 -0.083*** -0.112** -0.195*** One_child -0.204*** -0.230*** -12.483*** -0.219*** -0.025*** -0.070*** -0.175*** -0.163*** Two child + -0.229*** -0.247*** -10.987*** -0.245*** -0.029*** -0.063*** -0.252*** -0.211*** High No child -0.085*** -0.062*** -3.792*** -0.029*** 0.009 -0.048*** -0.097*** -0.055* One_child -0.137*** -0.145*** -7.435*** -0.122*** -0.003 -0.064*** -0.157** -0.112* Two child + -0.166*** -0.173*** -13.681*** -0.299*** 0.001 -0.081*** -0.221* -0.156 Source: Own elaborations on EU-SILC data Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. South (Southern European countries): Cyprus, Greece, Italy, Spain, and Portugal; Cont (Continental European countries): Austria, Belgium, France, Germany, Netherlands, and Luxembourg; North (Northern European countries): Denmark, Finland, Sweden, and Ireland; East (Eastern European countries): Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, and Slovak Republic. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 80 Table A13. Effects of reforms on the gender gap in self-employment (continuous and ordered variables) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.044*** -0.037*** -0.054*** -0.067*** (0.007) (0.008) (0.006) (0.006) length_maternity 0.001** 0.001* 0.001* 0.001** (0.000) (0.000) (0.000) (0.000) female * length_maternity -0.001*** -0.001*** -0.001*** -0.001* (0.000) (0.000) (0.000) (0.000) (2) female -0.070*** -0.066*** -0.075*** -0.084*** (0.003) (0.004) (0.004) (0.004) length_paternity -0.001 -0.001 -0.002** -0.001 (0.001) (0.001) (0.001) (0.001) female * length_paternity 0.003*** 0.002*** 0.002*** 0.003*** (0.001) (0.001) (0.001) (0.001) (3) female -0.089*** -0.088*** -0.087*** -0.097*** (0.007) (0.008) (0.006) (0.006) paid_maternity -0.001 0.001 -0.004 -0.002 (0.005) (0.006) (0.009) (0.004) female * paid_maternity 0.009*** 0.010*** 0.006*** 0.008*** (0.002) (0.003) (0.002) (0.002) (4) female -0.074*** -0.070*** -0.083*** -0.089*** (0.005) (0.006) (0.005) (0.005) paid_paternity 0.004* 0.005* 0.002 0.002 (0.002) (0.003) (0.003) (0.003) Female * paid_paternity 0.005*** 0.005** 0.006*** 0.005*** (0.002) (0.002) (0.002) (0.002) (5) female -0.053*** -0.047*** -0.076*** -0.067*** (0.007) (0.008) (0.007) (0.008) ps_family_ben 0.009** 0.010** -0.001 0.006* (0.004) (0.004) (0.003) (0.004) female * ps_family_ben -0.003 -0.005* 0.002 -0.001 (0.002) (0.003) (0.002) (0.002) (6) female -0.058*** -0.052*** -0.071*** -0.078*** (0.005) (0.006) (0.006) (0.006) ps_early_ed_care 0.023* 0.023* 0.008 0.020* (0.012) (0.013) (0.012) (0.011) female* ps_early_ed_care -0.007 -0.010* 0.003 0.008* (0.005) (0.006) (0.005) (0.005) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 81 Table A14. Effects of reforms on the gender gap in hourly wage (0 before the reform, 1 after the reform) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.112*** -0.099*** -0.111* -0.232** (0.026) (0.032) (0.058) (0.095) child_care -0.000 0.009 -0.039 -0.002 (0.034) (0.042) (0.041) (0.044) female * Child_care 0.009 0.025 -0.083** 0.014 (0.028) (0.036) (0.041) (0.047) (2) female -0.105*** -0.081** -0.206*** -0.275*** (0.026) (0.032) (0.058) (0.072) par_leave 0.022 0.006 0.125*** 0.024 (0.036) (0.044) (0.048) (0.048) female * par_leave 0.012 0.023 -0.034 0.003 (0.031) (0.040) (0.043) (0.050) (3) female -0.081** -0.078* -0.249*** -0.311*** (0.035) (0.042) (0.080) (0.108) work_family_bal -0.085* -0.082 -0.011 -0.181*** (0.049) (0.057) (0.056) (0.060) female * work_family_bal 0.043 0.046 0.016 0.123** (0.041) (0.051) (0.053) (0.056) (4) female -0.120*** -0.112*** -0.220*** -0.294*** (0.024) (0.028) (0.059) (0.081) gen_bal_par -0.069** -0.085** 0.006 -0.005 (0.033) (0.038) (0.047) (0.046) female * gen_bal_par 0.032 0.046 0.016 -0.045 (0.032) (0.040) (0.041) (0.052) (5) female -0.103*** -0.094*** -0.187*** -0.242*** (0.025) (0.031) (0.053) (0.073) child_support -0.115*** -0.131*** -0.031 -0.089** (0.030) (0.035) (0.042) (0.044) female * child_support 0.030 0.043 0.005 -0.029 (0.026) (0.034) (0.036) (0.046) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 82 Table A15. Effects of reforms on the gender gap in hourly wage (continuous and ordered variables) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.115*** -0.075 -0.210*** -0.392*** (0.040) (0.053) (0.066) (0.066) length_maternity -0.005** -0.005** -0.007*** -0.004 (0.002) (0.002) (0.002) (0.004) female * length_maternity 0.002 0.001 0.003 0.007*** (0.002) (0.002) (0.003) (0.002) (2) female -0.090*** -0.070** -0.170*** -0.313*** (0.024) (0.028) (0.056) (0.063) length_paternity 0.007 0.009 -0.005 0.004 (0.009) (0.009) (0.010) (0.013) female * length_paternity 0.004 0.004 0.003 0.012* (0.004) (0.005) (0.006) (0.006) (3) female -0.033 0.006 -0.143 -0.263*** (0.039) (0.047) (0.091) (0.061) paid_maternity 0.036 0.040 0.062 0.006 (0.049) (0.056) (0.071) (0.085) female * paid_maternity -0.016 -0.022 -0.006 -0.001 (0.012) (0.014) (0.026) (0.015) (4) female -0.144*** -0.141*** -0.209*** -0.225*** (0.031) (0.037) (0.060) (0.067) paid_paternity 0.006 0.009 0.015 -0.016 (0.019) (0.022) (0.019) (0.020) Female * paid_paternity 0.027** 0.036** 0.018 -0.014 (0.011) (0.014) (0.014) (0.017) (5) female -0.078* -0.057 -0.091 -0.337*** (0.042) (0.053) (0.074) (0.098) ps_family_ben 0.075** 0.087** 0.029 0.026 (0.033) (0.037) (0.039) (0.048) female * ps_family_ben 0.002 0.003 -0.022 0.033 (0.016) (0.020) (0.022) (0.027) (6) female -0.114*** -0.103*** -0.178*** -0.262*** (0.029) (0.034) (0.066) (0.072) ps_early_ed_care 0.152 0.144 0.028 0.308** (0.093) (0.103) (0.132) (0.135) female* ps_early_ed_care 0.064** 0.087** 0.038 0.009 (0.032) (0.040) (0.054) (0.047) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 83 Table A16. Effects of reforms on the gender gap in hourly earnings (0 before the reform, 1 after the reform) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.088*** -0.059** -0.137*** -0.160*** (0.025) (0.030) (0.051) (0.052) child_care 0.004 0.022 -0.061* -0.012 (0.030) (0.035) (0.035) (0.033) female * Child_care -0.044 -0.042 -0.059 0.007 (0.030) (0.036) (0.036) (0.035) (2) female -0.111*** -0.092** -0.157*** -0.221*** (0.031) (0.036) (0.046) (0.057) par_leave 0.055** 0.050 0.104*** 0.029 (0.027) (0.032) (0.038) (0.034) female * par_leave -0.027 -0.024 -0.063* 0.040 (0.033) (0.040) (0.038) (0.041) (3) female -0.057* -0.045 -0.155** -0.247** (0.031) (0.038) (0.067) (0.097) work_family_bal -0.022 -0.016 -0.034 -0.107* (0.041) (0.047) (0.046) (0.056) female * work_family_bal -0.011 -0.019 -0.004 0.121** (0.037) (0.043) (0.050) (0.047) (4) female -0.132*** -0.130*** -0.158*** -0.179*** (0.023) (0.027) (0.049) (0.065) gen_bal_par -0.068** -0.083** -0.012 -0.007 (0.029) (0.033) (0.039) (0.039) female * gen_bal_par 0.020 0.022 0.024 0.022 (0.030) (0.035) (0.037) (0.041) (5) female -0.076** -0.060* -0.126*** -0.152** (0.029) (0.035) (0.047) (0.059) child_support -0.036 -0.039 -0.003 -0.041 (0.029) (0.034) (0.037) (0.034) female * child_support -0.008 -0.002 -0.028 -0.017 (0.031) (0.038) (0.034) (0.038) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 84 Table A17. Effects of reforms on the gender gap in hourly earnings (continuous and ordered variables) (1) total (2) no_child (3) one_child (4) two_child_more (1) female -0.134*** -0.105** -0.173*** -0.295*** (0.038) (0.047) (0.058) (0.050) lengthmaternity -0.004** -0.004** -0.004** -0.003 (0.002) (0.002) (0.002) (0.002) c.female#c.lengthmaternity 0.001 0.001 0.001 0.003 (0.002) (0.002) (0.002) (0.002) (2) female -0.157*** -0.148*** -0.174*** -0.278*** (0.026) (0.029) (0.045) (0.048) lengthpaternity 0.005 0.006 0.001 0.010 (0.008) (0.008) (0.009) (0.010) c.female#c.lengthpaternity 0.017*** 0.020*** 0.004 0.009 (0.005) (0.006) (0.005) (0.006) (4) female 0.051 0.113*** -0.085 -0.242*** (0.036) (0.041) (0.089) (0.046) paidmaternity 0.096** 0.130*** 0.079 -0.079 (0.046) (0.049) (0.071) (0.049) c.female#c.paidmaternity -0.055*** -0.071*** -0.025 -0.001 (0.012) (0.013) (0.027) (0.012) (5) female -0.184*** -0.186*** -0.186*** -0.243*** (0.041) (0.048) (0.052) (0.051) paidpaternity -0.021 -0.023 0.001 -0.003 (0.019) (0.022) (0.017) (0.015) c.female#c.paidpaternity 0.032** 0.041** 0.010 0.000 (0.014) (0.016) (0.014) (0.013) (7) female -0.160*** -0.139*** -0.160*** -0.311*** (0.039) (0.049) (0.061) (0.067) psfamilybenefits 0.016 0.014 0.037 0.013 (0.031) (0.033) (0.034) (0.038) c.female#c.psfamilybenefits 0.027* 0.027 0.006 0.034* (0.015) (0.018) (0.020) (0.020) (8) female -0.197*** -0.195*** -0.216*** -0.252*** (0.029) (0.033) (0.056) (0.054) psccandee 0.089 0.081 0.057 0.074 (0.084) (0.091) (0.123) (0.130) c.female#c.psccandee 0.141*** 0.171*** 0.084* 0.034 (0.030) (0.037) (0.049) (0.039) Source: Own elaborations on EU-SILC data. Notes: Weighted OLS estimates (weights: population share of the demographic group in the country/year); Robust SE clustered at country/year level; all regressions include time and country fixed effects. Complete estimates are available upon request. PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 85 List of Tables Table 1. Worktime, housework and childcare, by demographic and household characteristics ............. 14 Table 2. Descriptive labour market outcomes by gender, education and age (23 EU-countries, 2006-2018) 25 Table 3. Table 3. Descriptive gender gaps in labour market outcomes for parents (23 EU-countries, 20062018) 26 Table 4. Baseline estimates: gender gap in labour force participation and parenthood .......................... 30 Table 5. Baseline estimates: gender gap in employment and parenthood ............................................. 32 Table 6. Gender gap in other labour market outcomes and parenthood (EU 23 countries) .................... 33 Table 7. Parental leave length and generosity, public spending on family benefits and early education and care (average EU 22 countries) ...................................................................................................................... 44 Table 8. Reforms on the gender gap in labour force participation (0 before the reform, 1 after the reform) 46 Table 9. Table 9. Effects of reforms on the gender gap in labour force participation (continuous and ordered variables) ........................................................................................................................................... 47 Table 10. Effects of reforms on the gender gap in employment (0 before the reform, 1 after the reform) 49 Table 11. Effects of reforms on the gender gap in employment (continuous and ordered variables) ... 50 Table 12. Table 12. Effects of reforms on the gender gap in hours worked (0 before the reform, 1 after the reform) 51 Table 13. Table 13. Effects of reforms on the gender gap in hours worked (continuous and ordered variables) 52 Table A1. Tobit regression results for relative worktime, by country ......................................................... 64 Table A2. Tobit regression results for relative housework, by country ...................................................... 66 Table A3. Sample countries macro-level indicators of gender equality over years when HETUS wave 2010 was conducted ....................................................................................................................................... 68 Table A4. Children and labour market outcomes: variables and definitions .............................................. 69 Table A5. Gender gap in labour market outcomes and parenthood (macro groups) ................................ 70 Table A6. Gender gap in labour market outcomes and household characteristics ................................... 72 Table A7. Policy reforms variables: definition, source and number of countries ....................................... 74 Table A8. Effects of reforms on the gender gap in full-time employment (0 before the reform, 1 after the reform) 75 PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 86 Table A9. Effects of reforms on the gender gap in full-time employment (continuous and ordered variables) 76 Table A10. Effects of reforms on the gender gap in permanent employment (0 before the reform, 1 after the reform) 77 Table A11. Effects of reforms on the gender gap in permanent employment (continuous and ordered variables) 78 Table A12. Effects of reforms on the gender gap in self-employment (0 before the reform, 1 after the reform) 79 Table A13. Effects of reforms on the gender gap in self-employment (continuous and ordered variables) 80 Table A14. Effects of reforms on the gender gap in hourly wage (0 before the reform, 1 after the reform) 81 Table A15. Effects of reforms on the gender gap in hourly wage (continuous and ordered variables) ... 82 Table A16. Effects of reforms on the gender gap in hourly earnings (0 before the reform, 1 after the reform) 83 Table A17. Effects of reforms on the gender gap in hourly earnings (continuous and ordered variables) 84 PARENTHOOD, LABOUR MARKET OUTCOMES AND POLICIES www.projectwelar.eu Page ! 87 List of Figures Figure 1. Gender gaps in relative worktime, housework and childcare, by country ................... 17 Figure 2. Country-level correlation of wife’s relative worktime and gender equality indicators . 19 Figure 3. Country-level correlation of wife’s relative time spent on housework, including childcare, and gender equality indicators ....................................................................................... 20 Figure 4. Figure 4. Country-level correlation of wife’s relative time spent on childcare, and gender equality indicators ............................................................................................................... 21 Figure 5. Gender labour market gaps in macro-groups ............................................................. 34 Figure 6. Gender labour market gaps and parenthood in macro-groups .................................. 36 Figure 7. Gender labour market gaps and parenthood by household type ............................... 38 Figure 8. Number of countries with a policy change, by year (cumulative) ................................ 41 Figure 9. Number and timing of reforms, by country .................................................................. 42 WeLaR is Horizon Europe research project examining the impact of digitalisation, globalisation, climate change and demographic shifts on labour markets and welfare states in Europe. It aims to improve the understanding of the individual and combined effects of these trends and to develop policy proposals fostering economic growth that is distributed fairly across society and generates opportunities for all.