The role of attitudes towards maternal employment in the relationship between job quality and fertility intentions.
In contrast to the United States, in some European countries, fertility has fallen below 1.5 births per woman. Family formation has been postponed to later ages, and childlessness rates have increased substantially (Johnson-Hanks et al., 2012). Among demographers there is a consensus that low fertility has emerged directly from fertility postponement and is mainly a consequence of changes in fertility timing (Kohler, Billari, & Ortega, 2002; Lutz, O'Neill, & Scherbov, 2004; Sobotka, 2004). Were postponement the only cause of fertility decline, such a decline would not persist, and a trend reversal would be expected. Indeed, most low-fertility European countries have recently experienced a reversal of fertility decline. Yet in all three German-speaking countries (Austria, Germany, and Switzerland), fertility has remained unchanged (Goldstein, Sobotka, & Jasilioniene, 2009; Sobotka & Zeman, 2011). Freijka & Sardon (2004, p. 376) estimated that women born in 1975 might reach completed fertility rates of 1.2-1.3 births by the end of their childbearing years in Austria, Germany, and Switzerland. Therefore, German-speaking countries are the only exception to the current reversal of fertility trends in Europe.
These developments make it crucial to understand intervening mechanisms other than demographic ones that can sustain such low total fertility rates. An emerging area of research focuses on the impact of housing conditions as mechanisms for determining fertility intentions. For example, Vignoli, Rinesi, & Mussino (2012) showed in a study using the Italian variant of the Generations and Gender Survey, a clear gradient between the fertility intentions of couples and the degree to which they feel secure about their housing situation. Home ownership represents one of the main sources of investment for family savings; it provides an indirect source of income (i.e. the imputed rent), it enables future and sustainable consumption (Dewilde & Raeymaeckers, 2008), and protects against risks of eviction (Mulder & Hooimeijer, 1999), thereby promoting the formation of childbearing intentions. Another important factor in the literature on fertility intentions are social network mechanisms. Social influence can help explain representations of parenthood and ideal family size (Bernardi, 2003), social learning mechanisms have been considered crucial to distinguish who forms childbearing intentions and puts them into practice, while finally social interaction is important to fertility because relationships and informal support networks can complement the institutional provision of childcare (Bernardi & Rossier, 2009; Bernardi & Klarner, 2014). Researchers also derive fertility differentials from the design of family and employment policies to facilitate the reconciliation of work and family (Hoem, 2005; Kaufman & Bernhardt; Olah & Bernhardt, 2008; Billingsley & Ferrarini, 2014).
Other scholars have emphasized the role of employment conditions on fertility (Blossfeld & Hofmeister, 2007; Blossfeld, Klijzing, Mills, & Kurz, 2005; Kreyenfeld, Andersson, & Pailhe, 2012; Sobotka, Skirbekk, & Philipov, 2011). Unemployment has been repeatedly related to low fertility, especially among men (Adsera, 2005; Pailhe & Solaz, 2012; Schmitt, 2012), and precarious work has been claimed to contribute to fertility postponement (Bernardi, Klaerner, & von der Lippe, 2008; Hanappi, Ryser, Bernardi, & Le Goff, 2012; Scherer, 2009; Steiber & Haas, 2009). Overall, findings suggest that the quality of a job matters in determining how easily parents can sustain the financial burden of a child and combine work and family, and therefore contributes to explaining childbearing intentions (Begall & Mills, 2011; Del Bono et al., 2014; Cazzola et al. 2016).
Far less is known about the effects of gender attitudes. Gender attitudes have been shown to influence work and care choices, as well as family planning (Rindfuss & Brewester, 1996; McQuillan et al., 2015). In contexts where women are disadvantaged in the work sphere (due to a large pay gap, less influential and often part-time positions, lower or uncertain income at retirement, and gender-biased parental leave) often such disadvantage translates in higher levels of childlessness and later transitions to motherhood. This is especially the case where the normative frame, expressed by individual attitudes towards care and maternal employment, does not sustain gender equality in the private sphere. The more unequal gender attitudes in the home the more likely families are under pressure, as women fulfill tasks in the labor market but still carry out most of the caregiving and domestic tasks, which puts great pressure on families to limit fertility (Neyer, Lappegard, & Vignoli, 2013; Olah, 2003). Especially attitudes towards maternal employment are decisive in how families distribute work and care burdens. The latest discourse about maternal employment has been characterized by a retraditionalization of gender attitudes in the private sphere. Wall (2013), in her media discourse analysis of Canada's leading family magazine, reports that there is a growing tendency to insist on "putting families first" rather than on improving conditions for work-family reconciliation for parents (mothers). If this is true two questions arise: On this basis are women and men who disapprove of maternal employment less likely to intend to have a child? And if so, do such attitudes change the relationship between job quality and fertility intention for them? To the best of our knowledge there is no previous study that has so far addressed this specific question.
In this article we therefore examined whether attitudes towards maternal employment change the relationship between job quality and the intention to have a child. First, we examined in what way job quality is associated with the intention to have a child and how this association differed for men and women; in a next step, we examined this association for first and subsequent child intentions (i.e., the childless versus parents) and again looked for gender differences. Subsequently, we examined the interaction between job quality and the gender-role attitude toward maternal employment and elaborated on how this attitude changes the effects of job quality on the childbearing intentions of men and women.
Our analysis is based on the Swiss Household Panel (2002-2011) and takes advantage of information on job stability and occupational prestige as two measures of job quality and the intention to have a child within 24 months. We thus account for the effects of occupational privileges--such as flexible work hours or the possibility of delegating work tasks to others if necessary--on the intention to have and raise children. We include interactions between individual gender-role attitudes, measuring how approval of maternal employment affects the link between job quality and fertility intentions. Whereas the concept of gender-role attitudes is relatively wide and contains gender roles in the public sphere (e.g., the belief that women are generally penalized), we focus on whether respondents approve of maternal employment because this attitude affects most how work and care choices are made upon the arrival of a child.
Given the reciprocal influences of partners on each other's fertility intentions, we control for partners' fertility intentions and a number of partners' characteristics (Cavalli & Rosina, 2011; Miller & Pasta, 1996; Vignoli, Drefahl, & De Santis, 2012). We make assumptions, based on previous empirical evidence, on the causal direction of the association between job quality and fertility intentions.
The Swiss case is particularly interesting as it is one of the few exceptions to the abovementioned fertility reversal. This country showcases a pattern whereby if women are disadvantaged at work and gender inegalitarian norms dominate, the outcome is low fertility and high rates of childlessness. According to SHP data concrete intentions to have children of 2.19 for men and 2.21 for women by far exceed actual fertility (i.e. the total fertility rate amounts to about 1.5) (for details see Philipov and Bernardi, 2011, p. 511). According to this panel data, 55 per cent of the respondents who declared they want a child within the next two years realized this intention, compared to for instance the Netherlands were this holds for 75 per cent of respondents (Philipov & Bernardi, 2011, p. 515). Most of the respondents who did not realize their intention declared an intention to have a child during the subsequent three years (Hanappi, 2014). Though such a finding is often observed in low fertility contexts and thus is not peculiar to Switzerland, the rather large gap between fertility intentions and their realization suggests that Switzerland's particularly low gender equality at work and in the family have made Switzerland the locus of a "third fertility compromise," whereby a hardly bearable compromise between working and having children are considered crucial for low fertility rates (Caldwell, 2008).
Many studies have aimed to identify factors that have been associated with a decline of micro- and macro-level fertility over the past quarter century. In the microeconomic literature employment is considered a key determinant in fertility differentials. Men's employment has been consistently described as a major factor associated with increasing fertility, whereas increasing female labor force participation has often been seen in relationship with fertility decline during that period (Engelhardt, Kogel, & Prskawetz, 2004). Female employment is seen to have two effects on fertility: first, women's employment might contribute to increasing household income and therefore improve the family's capacity to meet its budget, which might increase fertility by an income effect. Alternatively, a substitution effect might decrease fertility because a woman's temporary absence from the labor market due to childbearing might result in an earnings decline over the life course (Kohler & Kohler, 2002). Empirical studies seem to bear this out, having supported ambiguous employment effects on micro-level fertility (see Blossfeld & Hofmeister, 2007; Ozcan, Mayer, & Luedicke, 2010, for examples).
2.1. Job Quality, Gender Roles, and Fertility
Most microeconomic analyses examine the link between employment and fertility by focusing on employment status, types of contracts, and income (actual employment uncertainty). Empirical evidence is scarcer with respect to perceived job quality. The concept of job quality was most commonly conceived as job stability and, to some extent to occupational prestige (Hofmann & Hohmeyer, 2013). The higher the level of job stability and the higher the score of occupational prestige, the higher would be a person's job quality.
The gender literature on fertility intentions suggests that the direct relationship of perceived job quality on the intention to have a child is different for men and women (Neyer et al., 2013): in gender-unequal contexts women engage less in paid work and more in domestic work, so that the role job quality has for them in making childbearing decisions is a subordinate one. An approach that takes a similar perspective is the master status approach developed by Kruger & Levy (2001). They posited that despite some progress toward egalitarian values in the Western World, men and women are still preferentially assigned to either the work domain (in the case of men) or the family (in the case of women). Men therefore comply with their master status if they invest in work, whereas women need to accommodate their work with family needs to comply with norms of good mothering.
Perceived job instability. Perceived job instability is one component of job quality and has increasingly been studied to estimate the impact of the economic recession on fertility. One major reason has been that societal transformations, labor market reforms, and increasing internationalization have increased uncertainties in individuals' lives. Beck (1999) coined the term "the risk society" to denote the effects of these developments. Many southern, central, and eastern European countries have indeed recorded substantial fertility declines in moments of economic and political uncertainty. Researchers have examined perceived rather than actual job instability (e.g., job fears versus limited duration contracts) with the idea that individuals may perceive job instability independently from their actual situation (for instance, fearing that their plant may close down despite having a permanent contract). If this were the case, the perception of losing one's job might restrain people as much as being actually unemployed (Kasl, Gore, & Cobb, 1975). Assuming the predictions of microeconomic theory and gender role theory on the effects of perceived job instability on fertility intentions, we expected that:
Stable employment should be positively associated with fertility decisions, and thus fertility intentions. Job instability should be negatively associated with fertility intention. This association should be strong among men and weak among women (job instability hypothesis).
For men, we expect stable employment to strongly and positively affect their intention to have a child, because obtaining a stable income would support their breadwinner capacities and comply with existing norms of good fathering. For women, we expect a weaker positive impact of job stability on their fertility intentions because women are preferentially assigned the family and work is their subordinate domain.
Occupational prestige. Another component of job quality is occupational prestige. The literature on work-family conflict addresses the relationship between fertility decisions and occupational prestige--implying that occupational prestige matters for childbearing in different ways for men and women in gender-inegalitarian contexts: for men, prestigious jobs primarily imply higher income, while for women prestigious jobs are more about valuable resources to combine work and family (Voydanoff, 2005). For instance, people in high-prestige jobs might have the opportunity to work flexible hours and to delegate tasks and this might be especially important for mothers (Hochschild & Manchung, 1989). This enables them to better cope with tensions between work and family life; such "good" jobs, as Begall & Mills (2001) called them, have been found to be positively linked with the intent to have a child. Accordingly, in gender-inegalitarian contexts, and because of the direct relationship that occupational prestige has with the intention to have a child, we expected that:
Men in high-prestige more than low-prestige jobs will be prone to intend to have children because the widespread norms of good fathering define fathers as stable income providers and, therefore, are coherent with high-quality jobs. Women in high-prestige more than low-prestige jobs should also be prone to intend to have children; however, the widespread norms about good mothering define women as primary caregivers, and thus we expect the positive association between occupational prestige and childbearing intentions to be weaker among women compared to men (master status hypothesis I).
Gender roles. At the micro level, it is possible that the gender-role attitude toward maternal employment moderates the impact of job quality on fertility intentions. One explanation is that the gender-role attitude affects how men and women engage in working and caregiving within the couple (Buber-Ennser, 2015). Unequal gender roles make men more prone to engage in paid labor and women more prone to specialize in domestic work. Rindfuss and Brewster (1996) observed such specialization, which, according to them, resulted in a more traditional division of household labor. Gender roles also have repercussions on other couple decisions such as the decision to have a child (Olah & Bernhardt, 2008). Couples anticipate future work and care responsibilities arising from having children. In this way it is likely that the impact of job stability on child intentions is influenced by people's gender- role attitudes. In particular, their attitude toward maternal employment is central: If people are concerned about maternal employment, they implicitly assign caregiving to women, which affects their work and care choices. Given people assume a child suffers with a working mother, they may also believe men will need to earn most of the family income to sustain the household. This is coherent with the master status approach by Levy, Widmer, & Kellerhals (2002), which suggested the following substantial gender differences: Whereas a working father would meet societal expectations for him by merely being the provider of a stable income, mothers with a stable income would not necessarily meet expectations for them by providing most of the childcare. Even though in Switzerland many mothers have stable part-time jobs (Anxo, Fagan, Cebrian, & Moreno, 2007), their job stability is still subordinate to childrearing in order to reconcile work and family life. Consequently, do women concerned about maternal employment assign less importance to their own job stability in the formation of family plans. Simultaneously, do such attitudes increase the impact of job stability among men because they consider their ability to provide a stable income more important compared to those men with a gender-equal attitude.
Gender roles also influence the impact of occupational prestige on childbearing intentions. Occupational prestige is often linked with valuable resources to combine work and family (e.g., financial resources to outsource childcare, support from family). However, most women concerned about maternal employment may not base childbearing intentions on the availability of resources; rather, they rely on their partners. Men who disapprove of maternal employment consider themselves as the main income providers, and hence, gender roles should increase the association between occupational prestige and childbearing intentions. This means that the specific gender-role attitude toward maternal employment may bring gender differences to the forefront (Levy, Widmer, & Kellerhals, 2002; Olah, 2003) in the effect of occupational prestige on fertility intentions. We therefore expect that:
The attitude toward maternal employment increases the association between job quality (i.e., job stability and occupational prestige) and childbearing intention for men but weakens this association for women (Hank & Kreyenfeld, 2003; Rindfuss & Brewster, 1996) (master status hypothesis II).
2.2. Parenthood and Fertility
Research has shown variation among countries in the extent to which employment affects fertility decisions between mothers and fathers compared to childless women and men, respectively (Schmitt, 2012). Some have argued that this variation is closely related to the policy regime at work (Rindfuss, Guzzo, & Morgan, 2003), to the degree of gender equality (Neyer et al., 2013), and to common national labor-market participation patterns over the life course (Anxo, Fagan, Cebrian, & Moreno, 2007). (1) From a microeconomic perspective, low job quality will reduce the price of time required for family duties, in other words it will cost less to spend time with family because the time spent at work is of lesser value. Ignoring potential income effects, low job quality should have a positive association with fertility intentions due to the reduced opportunity-cost of the partner with a low-quality job; he or she can specialize in the homemaker role, and--in the case of parents--also in the caregiver role. However, there are practical (mostly financial) and normative limitations on the extent to which fathers can and are ready to take on a larger share of parental duties. Moreover, in countries such as Switzerland, where a male breadwinner model dominates, the negative social impacts of low father's job quality compared to low childless men's job quality may be stronger than opportunity costs arising from undertaking parental duties (e.g., forgone promotions, reduced income).
The job instability and master status I hypotheses shall translate as follows for men and women: in the case of men, there should be a stronger positive association between job quality and fertility intentions for fathers than childless men. Having a stable and high-status job is all the more important for fathers than men without children in a male breadwinner context, because each child increases the need for higher income. In the case of women, we expect to observe a weaker positive association among mothers when they are primary caregivers compared to childless women, because having children in the household increases the total housework and care load and then competes with often more demanding high-quality jobs.
Regarding parents, our master status II hypothesis draws on studies suggesting that parenthood renders actual care choices more important than attitudes toward gender roles. Two explanations are possible: first, men and women have to face practical limitations in their daily organization of work and care; and second, they face problems in putting their ideals about good parenting into practice; even if partners share similar ideas, each of them might have different expectations about how these are to be realized (Bernardi, Ryser, & Le Goff, 2013; Buhlmann, Elcheroth, & Tettamanti, 2009; Dribe & Stanfors, 2009). Given the practical limitations and gender-specific ideals about good parenting, we expect the following relationships between job quality and childbearing intentions: for men, whether they are fathers should not change the way in which gender-role attitudes affect the influence of job instability in relation to childbearing intentions; each additional child increases the need for higher income. For women, we expect that the effects of gender-role attitudes on the relationship between job quality and childbearing intentions should vanish for mothers while staying significant for childless women; particularly because mothers undertake less of the paid work. Another reason is that women who already have children are already in a partnership and their male partner often has a stable job so that women's own job stability or occupational prestige is less of an issue.
3. Employment and Fertility in Switzerland
The Swiss welfare state is characterized by a policy regime that is conservative in regard to family policies (Armingeon, 2001; Bonoli, 2008). The welfare provisions for families are weak, and public services for childcare are insufficient to meet the demand of the competitive and liberal Swiss labor market, in which women's labor force participation is high. This has made Switzerland (together with Austria and Germany) the locus of a "third fertility compromise," whereby a hardly bearable compromise between work and family has produced remarkably stable low fertility rates (Caldwell, 2008). In 2012, the total fertility rate in Switzerland was 1.53, ranking it below the EU-27 average. More than 20% of highly educated women in Switzerland remained childless, and 73% of mothers worked part time (FSO, 2013).
Switzerland has not yet introduced parental leave policies. Maternity leave regulations grant mothers the right to take time off from work to care for children for 98 days following birth. The replacement rate amounts to 80% of previous earnings and is given in the form of daily allowances. This maternity insurance, introduced in 2005, grants mothers 14 weeks (98 days), with additional rights for weeks 15 and 16, such as staying home without pay. (2) Because benefits are related to previous earnings, these policies represent a strong incentive to enter the labor market before becoming a mother. As there is no paternity leave for fathers at federal level, mothers and fathers are differentially engaged in parenting newborns. Moreover, primary school schedules and rigid public office hours cause organizational hurdles for dual- earner families and single parents. In many Swiss cantons, primary school hours vary widely from day to day and from child to child, and children are dismissed daily for two-hour lunch breaks (Charles et al., 2001). According to the findings of these authors, most couples experience lower costs for women's time at home than for men's time, which makes employment more salient for men and reinforces the traditional household division of labor.
In the fertility literature, the existence of income and employment effects is well-known (see for studies on other European countries e.g., Kreyenfeld, Andersson, & Pailhe, 2012; Schmitt, 2012; Sobotka, Skirbekk, & Philipov, 2011). However, the empirical evidence for such effects is surprisingly scarce for Switzerland. This applies in particular to income and employment conditions. In a recent study using the Swiss Household Panel Survey (2002-2009), Bernardi et al. (2013) found indirect evidence that completing education and training more than actual hours worked plays a role in women's intention to have a first child. One reason is that couples who already have one child are likely to have adopted a more traditional role set (i.e., the male partner works full time) after the transition to parenthood, in which women are at most secondary earners and the quality of their jobs matters less compared to their childless counterparts. We expected men's job quality to affect their fertility intentions not only in relation to the first child but also in relation to any additional child, given that the need for higher income grows with family size.
4.1. Data and Sample
We used data from the Swiss Household Panel (SHP) for 2002-2011, which includes information about fertility intentions since 2002. The SHP is based at the Swiss Centre of Expertise in the Social Sciences (FORS see http:// www.swisspanel.ch for a description of the data set). Three random samples were followed yearly, and all household members older than 14 years were interviewed separately. This computer-assisted telephone survey was based on a closed-ended questionnaire that was translated in the three main languages spoken in Switzerland (German, French, and Italian). The first sample had been followed since 1999, the second since 2004, and the third since 2013. We selected men and women who were living with a partner (married or cohabiting) and were in the age group 22 to 45 for women and 22 to 50 for men from 2002 to 2011 and for whom we also had their partner's interview data. Focusing on the pressures individuals experience to combine work and family, we also opted to examine only survey participants who were active in the labor market. Our sample contained 552 childless women, 588 childless men, and 923 women and 1119 men with at least one child. An analysis of employment trajectories among Swiss women showed most of them work before the arrival of a child, and it is the arrival of a second or subsequent child that causes many of them to drop out of the labor market. Those women that do so are characterized by modest social origin, a low level of education and household income in the Swiss Household Panel (SHP) (see Levy, Gauthier, & Widmer, 2006: pp. 481-482). These findings suggest that women in less prestigious jobs who become mothers are more likely to drop out of the labor market. For our results interpretation and conclusion this implies applying some caution.
We used at least one observation per respondent and a maximum of ten: The majority of respondents participated in between one and three waves; only 5.9% of respondents participated in all 10.3 Descriptive statistics of sociodemographic and employment variables are shown in Table 1.
4.2.1. Dependent Variable
Our dependent variable is the intention to have a child (or another child) in the 24 months following the interview. We used the answers yes versus no to the question "Do you intend to have a child in the next 24 months"? Operating between ideal fertility and actual behavior, fertility intentions have been shown to correlate positively with subsequent childbearing behavior at the individual level (Ajzen, 1991; Miller & Pasta, 1995). Short-term fertility intentions are frequently used in micro-level studies to examine the effects of external conditions on fertility behavior (W.B. Miller & Pasta, 1995; Westoff & Ryder, 1977). In the short run, individuals can anticipate the effects of their partnerships, housing statuses, and economic conditions on the realization of their fertility plans (Billari, Philipov, & Testa, 2009; Heaton, Jacobson, & Holland, 1999).
4.2.2. Explanatory Variables
Job quality measures. We focus in our analysis on two measures of job quality: job instability and occupational prestige. Job instability is measured for employed (and self-employed) respondents and captures both subjective and objective job instability. Subjective job instability combines information about perceived job insecurity by permanent employees and self-employed respondents and about perceived contractual instability by temporary employees. Objective job instability refers to contracts shorter than three years. With these two sets of indicators we built a dichotomous variable combining both subjective and objective measures, for which 0 indicates stable job conditions and 1 indicates unstable job conditions. Unstable job conditions were those in which either individual expressed insecurity about his or her job or the contract was shorter than three years.
Occupational prestige. Occupational prestige is used to reflect privileged, high-status jobs and is measured by Treiman's prestige scale, which is based on occupational prestige ratings using the International Standard Classification of Occupations (ISCO; Ganzeboom & Treiman, 1996). This scale models a prestige hierarchy whose scores range between 0 (lowest prestige) and 100 (highest prestige), and it is designed to be independent of national and cultural settings.
Gender-role attitude. This measure is based on a question from the International Social Survey Program (ISSP) Families and Changing Gender Roles II/III. Respondents answer on a scale from 0 (completely disagree) to 10 (completely agree) whether they believe that a "Child suffers with a working mother," which reflects approval of maternal employment. This item has been recoded in a dichotomous variable such that 0 = completely disagree and 1 = completely agree.
Control variables. We controlled for a set of sociodemographic variables. The first was age, which was centered on the grand mean. Second, we distinguished low, medium, and high level of education achieved. For the highest level of education achieved, we distinguish low level (incomplete compulsory school, compulsory school, elementary vocational training, domestic science course, 1-year school of commerce, or general training school), middle level (apprenticeship, technical or vocational school, fulltime vocational school, bachelor/maturity), and high level (vocational high school with a master's certificate or federal certificate, academic high school, or university). We included the annual net household income in its logged form (Kuhn, 2009; Lipps, 2010). To measure the individual contribution to the household income, we computed an indicator with 1 = individual income matches household income exactly (i.e., individual income is included in the nominator) with higher scores indicating a larger contribution to household income which reflects a practical aspect of gender role division. In the case of parents, we controlled for the age of the youngest child (coded as 1 = a child aged between 1 month and 3 years, 2 = a child aged between 4 and 5 years; 3 = a child aged between 6 and 12 years, 4 = a child aged 13 years or older). We controlled for whether participants were married or not, because although births outside marriage are relatively rare in Switzerland (Charton & Wanner, 2001; Le Goff et al., 2005), married couples may hold differing, potentially more unequal, gender attitudes (Buhlmann et al., 2009). We also controlled for partner's childbearing intention and partner's occupational status, distinguishing whether the partner is at home, in training, working full time or part time, or jobless.
4.3. Analytical Strategy
In order to estimate the associations between our explanatory variables and the intention to have a child, which is a binary dependent variable, we applied nested hierarchical two-level logit models. All covariates taken as independent variables were time-varying. We used data from multiple observations (DiPrete & Forristal, 1994) and adjusted for the dependency of observations. In this way each individual was allowed to contribute multiple observations (e.g., reports on job instability in several waves) that were considered more similar to one another than observations by different persons within the same wave (Raudenbush & Bryk, 2002). Our intra-individual measures were time-dependent and were retrieved from up to 10 time points nested in the individual respondent. We attained at least one declared fertility intention for each respondent in our 10-year annual panel, from 2002 to 2011. In most cases, the information on intentions covered multiple points in time. We used unit-specific models in which the lower level represented intra-individual measures (which vary across waves and are time dependent) and the higher level represented the individual man or woman. The estimated multilevel models enabled us to disentangle inter-individual and intra- individual variability (Hox, 2002; Singer & Willett, 2003). We estimated fixed effects for the intercept and the different covariates, as well as a random effect for the intercept. The hypothesis was that the intercept varies for each respondent, according to unknown characteristics, while there are no variations in the effect among different covariates. This hypothesis concerning a random effect on the intercept is commonly made when one is dealing with a case of multilevel logistic regressions. Models were estimated using HLM software, version 6, and the chosen method of estimation was restricted maximum likelihood. (4)
5.1. Descriptive Results
We start our analysis with a brief sample description and show fertility intentions for the following four groups in Table 1: childless women, childless men, mothers, and fathers.
There were differences in the composition of age and civil status for women and men in our sample. Women were on average younger than men, and respondents without children were younger than those who had children. For instance, childless women were on average 5.1 years younger than mothers, whereas the age difference was less pronounced between childless men and fathers (4.3 years). Most women and men without children were single or had never been married, whereas 90% of parents in our sample were married. On the contrary, differences in average individual net income between women and men were large. This difference was biggest among parents, where mothers earned on average one third less than fathers.
We next looked at the pattern of our explanatory variables: unlike gender, parenthood is what fuels gender differences in how job quality is distributed among men and women (see Table 1). The difference in job instability was greatest for fathers and childless men. More fathers than childless men reported having a stable job, and only about 11% of fathers mentioned having an unstable job at all. Conversely, job instability was similarly distributed among mothers and childless women. Most importantly, more mothers (17%) than fathers reported having an unstable job. This is not surprising because--in a male breadwinner context--parenthood is better compatible with men's than with women's paid work. This is also reflected by the difference in occupational prestige, where mothers score slightly lower than childless women. With respect to gender-role attitude, fewer mothers (41%) than childless women (46%) were concerned, but marginally more fathers (62%) than childless men (60%) had such concerns.
Descriptive statistics suggested that job quality and gender attitudes differ by parental status in first place and by gender in second place. In addition, Table 1 demonstrates that our SHP subsample is to some extent selective: The groups of highly educated men and women are overrepresented in our sample, because on the one hand men or women in a stable family and professional situation are overrepresented in the overall SHP sample. Men and women in precarious employment are underrepresented in the SHP sample. On the other hand, the arrival of a second or subsequent child causes many women to drop out of the labor market, so that there is by nature no data regarding their job quality. In general, we confirm findings by Levy, Gautier, & Widmer (2006) showing that those women are characterized by modest social origin, a low level of education and household income.
To obtain further evidence on the link between job quality measures, gender-role attitudes, and the formation of the intention to have a child, we now turn to the results of our multivariate analysis. This analysis directly tested whether job quality significantly related to fertility intentions for childless women and men, whether gender-role attitude had an impact on this relationship, and whether parenthood made a difference in this respect.
5.2. Multivariate Results
We ran four sets of estimations, first for childless women and men and then for mothers and fathers. All models were controlled for the sociodemographic factors mentioned above. We ran separate estimations for childless individuals and parents because the transition to parenthood is known to shape employment patterns for men and women as well as the couple's division of paid and unpaid work (Barber, 2001; Dommermuth, Klobas, & Lappegard, 2011; Hobcraft & Kiernan, 1995). Table 2 contains Models 1 and 2 for childless women and childless men. In Model 1 we estimate the effects of the indicators describing job quality measured by job instability (test of the job instability hypothesis) and occupational prestige (test of the master status hypothesis I) among childless respondents. In Model 2 we explore the interaction of gender-role attitudes on the relationship between job quality measures and fertility intentions for childless women and men (master status hypothesis II). Table 3 displays in Models 3 and 4 the test results of our three hypotheses for parents. In Model 3 we estimate the effects of job instability and occupational prestige, and Model 4 shows the interactions for parents.
We first turn to the results for Models 1 and 2 for childless women and men in Table 2. In each model, the intercepts show that the respondents were more likely to give a negative answer than to have the intention of having a child within 24 months. In addition, in all models the effect of age was significant and, as expected, showed that the intention to have a first child decreases with age (see e.g., Philipov, Speder, & Billari, 2006). Looking at covariates related to the couple situation, being married and having a partner who intends to have a child play a key role in increasing the chances that respondents intend to have a child. This is consistent with other studies on Switzerland, which have shown that individuals are less likely to give birth to a first child when unmarried (Charton & Wanner, 2001; Le Goff et al., 2005), and with an international study that shows the importance of couple's agreement in childbearing intentions (Berrington, 2004).
Model 1 indicates that a woman's intention to have a first child is affected by different job quality measures than would affect a man's intention. For women, as was expected in our job instability hypothesis, experiencing job instability makes them significantly less prone to intend to have a child (B = -0.853; [p.bar] < 0.01; odds ratio = 0.426), whereas women's occupational prestige does not seem to matter. For men our job instability hypothesis was not confirmed because there were no effects of job instability whatsoever. Results for men's and women's occupational prestige did not confirm our master status I hypothesis: Although high occupational prestige was positively associated with fertility intentions (which suggests that high-prestige jobs make the formation of a positive fertility intention more probable), the effects were not significant. In sum, Model 1 suggests that for childless women having a stable job (i.e., a stable income and a solid professional context) is conducive to entry into parenthood. Longer waiting time until occupational establishment presumably delays the timing of first births, which has been argued to potentially reduce the overall number of children these women actually intend to have (Bongaarts, 2002). Interestingly, Model 1 reveals that the female partner's job characteristics significantly affect a man's intention to become father. In particular, the negative association between having a female partner in training and men's intentions to have a first child (B = -0.982; [p.bar] < 0.1; odds ratio = 0.374) suggests the existence of social norms about sequencing of certain life course transitions--here, the start of family formation after having completed one's education (Bernardi, Klarner, & von der Lippe, 2008).
In Model 2 we tested whether childless women's and men's fertility intentions are also affected by factors other than job quality (job instability, occupational prestige). Here we tested our master status hypothesis II and introduced gender attitudes and the interactions between gender attitudes and job quality. The results provide evidence that first childbearing intentions are unrelated to a woman's gender-role attitude and do not act on the association of job quality and childbearing intentions. There is no difference between women who believe that a child suffers when the mother works and those who do not. In addition, interactions between gender-role attitudes and job quality measures have no effects. This implies at the same time that gender-role attitudes do not change the centrality of couple characteristics that we mentioned above (being married and the partner's intention to have a child). For men we observe a different pattern of associations. Model 2 shows that if childless men have a high-prestige job and think that the child would suffer with a working mother, they are less prone to intend to have a child in the near future (i.e., the interaction effect of occupational prestige and gender-role attitudes is significant and negative for this category).
We now proceed with Models 3 and 4 in Table 3 for parents, in which we assess whether job quality and gender-role attitudes affect mothers' and fathers' intentions differently. They allow us to evaluate whether these effects are similar to those we found for childless women and men or whether they differ according to life stage. Most importantly, across our models for parents we show that partner's intentions, own age, and the age of the youngest child are very important for mothers and crucial for fathers as well. As opposed to childless individuals, being married for parents does not significantly relate to intention to have another child; in other words, the transition to marriage has already occurred before the transition to the first child or marriage is not a precondition for having subsequent children. In order to explore the idea that parents need to combine work and different amounts of childcare, we controlled for the age of the youngest child. For both models, the results indicate that having at least one child under the age of 3 years in the home is strongly related to the intention to have a subsequent child: having children between 0 and 3 years more than having children between 4 and 5 years was positively related to the intention to have a subsequent child. This means that having children is possibly driven by age norms about spacing, or by institutional incentives to have children closer in age (i.e., paying less to have the second child in the same childcare facility). This would eventually mean that families economize total care over the reproductive life span, contradicting our expectation that current domestic workload would limit fertility in families with young children.
Model 3 indicates that a mother's intention to have a child is affected by different factors than affect a father's intention. For mothers, the results indicate no significant negative association between having an unstable job and intention to have another child. This result differs from the significant negative association we observed for childless women and confirms our expectation that job instability would be associated mainly with childless women's fertility intentions. For fathers we observed no effects whatsoever. Our job instability hypothesis is thus confirmed for women with and without children (but not for their male counterparts). Similar to the results in Model 1 for childless individuals, the results for mothers' and fathers' occupational prestige were insignificant and did not confirm our master status I hypothesis.
Model 4 tested our master status II hypothesis for mothers and fathers, and we introduced gender attitudes as well as the interactions between job quality measures and gender attitudes. The results demonstrated that for fathers, professional prestige was positively, but weakly, related to the intention to have another child (B = 0.021; [p.bar] < 0.1; odds ratio = 1.021). This reflects the idea that having a "good" job is all the more important for men, as their breadwinning responsibilities are higher. Moreover, the interaction effect between attitudes and job quality is significant: Fathers in high-prestige jobs who believed that a child suffers with a working mother were slightly less likely to intend to have another child (B = -0.026; p < 0.1; odds ratio = 0.974). One explanation is that intentions among these men are strongly linked with general characteristics of their partners; for instance, those men who have a highly qualified partner who works or wants to work have negative intentions. Results also show that men in unstable jobs who believed that a child suffers with a working mother were prone to intend to have another child (B = 1.140; [p.bar] < 0.05; odds ratio = 3.125). A plausible explanation for this result is that fathers who balance unstable jobs with family demands also assign high importance to both job and family. Therefore, these men are most likely to invest in paid labor while at the same time investing time at home and not compromising their family plans. For mothers, we found no significant interaction effects, and therefore concluded that gender roles do not act on the association of mother's job quality and their fertility intentions. In sum, the findings suggest that gender and life stage matter for how job quality measures and gender attitudes are associated with fertility intentions in the Swiss context.
The aim of our study was to provide empirical support for intervening mechanisms other than postponement for shrinking family sizes and the comparatively high childlessness rate in Switzerland. Using the Swiss Household Panel data (SHP 2002-2011) on job quality, gender-role attitudes, and fertility intentions, we estimated multilevel models in order to test whether, in addition to socio-demographic characteristics, the respondent's job quality, gender-role attitude, their interactions, and the partner's characteristics affect individuals' fertility intentions. We looked at men and women separately given the highly gendered specialization and expectations in terms of labor force participation and responsibility to the family.
As expected, intentions to have children were strongly age related. This is an indicator of the presence of a normative family life course in terms of both timing and sequencing of events. For both parents and non-parents, the intention to have a child is more likely to appear after some time spent in cohabitation, when marriage follows. A similar argument was developed by Sauvain-Dugerdil (2005) using the Swiss Fertility and Family Survey (FFS). In the 1990s, the youngest women in particular mentioned difficulties in reconciling their professional and family lives or unfavorable housing conditions as reasons not to intend a child.
Although there is a Europe-wide, universal policy of encouraging female labor-market participation as well as reducing the domestic workload, it has not yet resulted in major changes in social policy support for families in all countries (Rubery, Smith, Anxo, & Flood, 2001). Our analysis points to major shortcomings in the functioning of the Swiss employment system, despite the fact that it has heavily promoted equal opportunities for men and women since the 1980s. The compatibility of childrearing and employment, with its specific demands regarding schedule flexibility and autonomy in the organization of work, was not yet part of the political agenda back then. High-status jobs were designed as career tracks that offer quality training and promotions against continuous work commitment--in other words, a men's track. Conservative parties insisted on the mother's primary role for her child's development and were reluctant to create alternative childcare systems. This combination of elements contributed to reduce the attractiveness of children, reduce fertility, and increase the age of first motherhood.
This study provided partial evidence that men and women in developed countries having better employment are more likely to intend to have a child, other things being equal. However, we also showed with some degree of confidence that parents--especially mothers--are less sensitive to job quality when intending additional children. This confirms, to a certain extent, Caldwell's (Caldwell, 2008) hypothesis at the micro level. When couples are able to combine quality jobs with childbearing, in a context in which institutions facilitate a compromise between work and family (which is hardly the case in Switzerland), the effects on fertility intentions are positive. Both job stability and occupational prestige make it more likely for couples to intend to have children.
To integrate work into couples' fertility planning, job designs and policies must provide options for pursuing both career and family planning goals; it is crucial that parents perceive such options to be stable and realistic. That said, family relations and age continue to explain intentions to a large extent.
We limited our analysis to time-dependent fertility intentions, as these are claimed to closely relate to actual fertility behavior compared to declared family size ideals (Philipov & Bernardi, 2011; Philipov, et al., 2006). Provided that fertility intentions remain relatively stable over a short period of time but are sensitive to external factors, we plan to conduct further research on the association between intentions and actual fertility and to explore the potential reasons for the observed gap between intentions and realization. The first step is a research priority in family demography and sociology in many low fertility countries (Kohler, Billari, & Ortega, 2002). The second is to verify whether apart from couple's fecundity issues, the gap may be due to changes in intentions. Men and women may abandon or postpone their positive intentions or may switch from a negative to a positive intention. Examining whether these changes occur because of changes in attitudes, changes in employment conditions, or changes in family relations will allow us to better delimit the scope for gender and family policies in this area. Received 13 January 2016 * Received in revised form 16 February 2016
Accepted 17 February 2016 * Available online 28 February 2016
Department of Demography, University of California Berkeley; National Centre of Competence in Research LIVES, University of Lausanne
Swiss Centre of Expertise in Social Sciences (FORS), University of Lausanne
National Centre of Competence in Research LIVES, Institute of Social Sciences,
University of Lausanne
This project was supported by the Swiss National Centre of Competence in Research LIVES--Overcoming vulnerability: life course perspectives, which is financed by the Swiss National Science Foundation. Doris Hanappi's work has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under the grant agreement no. 320116 for the research project Families And Societies, and from the Austrian Academy of Science (OEAW) under the grant no. 11713 within the Austrian Program of Advanced Research and Technology.
We thank Irena Kotowska and the anonymous reviewers for their helpful comments on this paper, and Ian Copestake for his language editing service. Earlier versions of this paper have been presented at the 2015 Conference of the European Sociological Association (ESA), the 8th International Conference of the Swiss Household Panel (2015), the 7th Congress of the European Society on Family Relations (ESFR), and the 2015 International Conference of the Norwegian Research Council.
(1.) In many ways the German-speaking countries can be viewed as less gender- equal countries, whereas a country like Sweden shows, in comparison to other European countries, a very distinct labor market situation between mothers and fathers (Pettit & Hook, 2009). The most common effect of parenthood is that fathers invest more in paid labor, and mothers invest in unpaid care work. In the German speaking countries women are covered by maternity leave schemes and thus withdraw from the labor market for longer periods. The general trend, across countries, however, indicates that more women participate in paid work, which renders their concrete employment situation as important as that of men (Jacobs & Gerson, 2004).
(2.) We may note that the duration of the maternity leaves has been set too short to substantially affect gender roles. It legitimizes women's rights to be paid during the period of initial nurturing and caring for the child. Apparently, this reform reinforces gender roles since it has been amply justified as a health measure for the mother to recover after delivery and for the child to develop a better attachment to the mother (Valarino & Bernardi, 2010).
(3.) An analysis by Lipps (2006) based on the waves 2000 to 2005 showed that attrition is comparatively high in the SHP compared to other European panel surveys (e.g., British household panel), but it is not particularly selective with respect to sociodemographic or socioeconomic variables. Voorpostel (2010) added that the bias is small and similar to that found in other panel studies.
(4.) One may note that because we measured fertility intentions at the same time than job quality and gender attitudes, differences between the groups may be due to selection processes. Yet, most childless people here still have the opportunity to have a child, so they are not necessarily different from parents, apart from the way in which they time their fertility. Therefore this group is very heterogeneous and there is no way to disentangle the possible selectivity into this group with our model.
Adsera, A. (2005), "Vanishing Children: From High Unemployment to Low Fertility in Developed Countries," The American Economic Review 95(2): 189-193.
Ajzen, I. (1991), "The Theory of Planned Behavior," Organizational Behavior and Human Decision Processes 50(2): 179-211.
Anxo, D., Fagan, C., Cebrian, I., & Moreno, G. (2007), "Patterns of Labour Market Integration in Europe--A Life Course Perspective on Time Policies," SocioEconomic Review 5(2): 233-260.
Armingeon, K. (2001), "Institutionalising the Swiss Welfare State," West European Politics 24(2): 145-168.
Barber, J. S. (2001), "Ideational Influences on the Transition to Parenthood: Attitudes toward Childbearing and Competing Alternatives," Social Psychology Quarterly 64(2): 101-127.
Begall, K., & Mills, M. (2011), "The Impact of Subjective Work Control, Job Strain and Work-Family Conflict on Fertility Intentions," European Journal of Population 27(4): 433-456. doi: 10.1007/s10680-011-9244-z
Bernardi, L., Klaerner, A., & von der Lippe, H. (2008), "Job Insecurity and the Timing of Parenthood: A Comparison between Eastern and Western Germany," European Journal of Population 24(3): 287-313.
Bernardi, L., Ryser, V.-A., & Le Goff, J.-M. (2013), "Gender Role-Set, Family Orientations, and Women's Fertility Intentions in Switzerland," Swiss Journal of Sociology 39(1): 9-31.
Bernardi, L. (2003), "Channels of Social Influence on Reproduction," Population Research and Policy Review 22(5/6): 427-555.
Bernardi L., & Klarner, A. (2014), "Social Networks and Fertility," Demographic Research 30(22): 641-670.
Berrington, A. (2004), "Perpetual Postponers? Women's, Men's and Couples' Fertility Intentions and Subsequent Fertility Behaviour," Population Trends 117(Autumn): 9-19.
Billari, F. C., Philipov, D., & Testa, M. R. (2009), "Attitudes, Norms and Perceived Behavioural Control: Explaining Fertility Intentions in Bulgaria," European Journal of Population/Revue europeenne de Demographie 25(4): 439-465.
Billingsley, S., & Ferrarini, T. (2014), "Family Policy and Fertility Intentions in 21 European Countries," Journal of Marriage and Family 76(2): 428-445.
Blossfeld, H. P., & Hofmeister, H. (eds.) (2007), Globalization, Uncertainty and Women's Careers: An International Comparison. Cheltenham, UK, & Northampton, MA: Edward Elgar.
Bongaarts, J. (2002), "The End of the Fertility Transition in the Developed World," Population and Development Review 28(3): 419-443.
Bonoli, G. (2008), "The Impact of Social Policy on Fertility: Evidence from Switzerland," Journal of European Social Policy 18(1): 64-77.
Buber-Ennser, I. (2015), "Childbearing in Austria: Work and Family Roles," Journal of Research in Gender Studies 5(2): 121-146.
Buhlmann, F., Elcheroth, G., & Tettamanti, M. (2009), "The Division of Labour among European Couples: The Effects of Life Course and Welfare Policy on Value-Practice Configurations," European Sociological Review 26(1): 49-66.
Caldwell, J. C. (2008), "Three Fertility Compromises and Two Transitions," Population Research and Policy Review 27(4): 427-446.
Cavalli, L., & Rosina, A. (2011), "An Analysis of Reproductive Intentions of Italian Couples," Population Review 50(1): 21-39.
Cazzola, A., Pasquini, L., & Angeli, A. (2016), "The Relationship between Unemployment and Fertility in Italy: A Time-series Analysis," Demographic Research 34: 1-38.
Charles, M., Buchmann, M., Halebsky, S., Powers, J. M., & Smith, M. M. (2001), "The Context of Women's Market Careers," Work and Occupations 28(3): 371396.
Charton, L., & Wanner, P. (2001), "La premiere mise en couple en Suisse: choix du type d'union et devenir de la cohabitation hors mariage," Population (English edition) 56(4): 539-568.
Del Bono, E., Weber, A., & Winter-Ebmer, R. (2014), "Fertility and Economic Instability: The Role of Unemployment and Job Displacement," Journal of Population Economics 28(2): 463-478.
DeWilde, C., & Raeymaeckers, P. (2008), "The Trade-off between Home-ownership and Pensions: Individual and Institutional Determinants of Old-age Poverty," Ageing and Society 28(6): 805-830.
DiPrete, T. A., & Forristal, J. D. (1994), "Multilevel Models: Methods and Substance," Annual Review of Sociology 20: 331-357.
Dommermuth, L., Klobas, J., & Lappegard, T. (2011), "Now or Later? The Theory of Planned Behavior and Timing of Fertility Intentions," Advances in Life Course Research 16(1): 42-53.
Dribe, M., & Stanfors, M. (2009), "Does Parenthood Strengthen a Traditional Household Division of Labor? Evidence from Sweden," Journal of Marriage and Family 71(1): 33-45.
Engelhardt, H., Kogel, T., & Prskawetz, A. (2004), "Fertility and Women's Employment Reconsidered: A Macro-level Time-series Analysis for Developed Countries, 1960-2000," Population Studies 58(1): 109-120.
FSO (2013), "Labor Market Indicators for 2013," in Berset, M. (Labor Force Section) (eds.), Employment and Income. Neuchatel: FSO.
Goldstein, J., Sobotka, T., & Jasilioniene, A. (2009), "The End of "Lowest-low" Fertility?," Population and Development Review 35(4): 663-699.
Hank, K., & Kreyenfeld, M. (2003), "A Multilevel Analysis of Child Care and Women's Fertility Decisions in Western Germany," Journal of Marriage and Family 65(3): 584-596.
Hanappi, D., Ryser, V.-A., Le Goff, J.-M., & Bernardi, L. (2012), "Minding the Downside of Market Flexibilization. Does Precarious Work Affect Short-term Fertility Intentions and Their Realization," paper presented at the European Population Conference, University of Stockholm, June 13-16.
Hanappi, Doris (2014), "Linking Perceived Economic Uncertainty and Fertility Decisions. Evidence from Switzerland," paper presented at the Demography Brown Bag Seminar, October 29, University of California, Berkeley, CA.
Heaton, T. B., Jacobson, C. K., & Holland, K. (1999), "Persistence and Change in Decisions to Remain Childless," Journal of Marriage and Family 61(2): 531-539.
Hobcraft, J., & Kiernan, K. E. (1995), "Becoming a Parent in Europe," Plenary paper for the European Population Conference, in EAPS/IUSSP Proceedings of the European Population Conference, Milan, September, 27-65.
Hochschild, A. R., & Manchung, A. (1989), The Second Shift: Working Women and the Revolution at Home. New York: Viking.
Hofmann, B., & Hohmeyer, K. (2013), "Perceived Economic Uncertainty and Fertility: Evidence from a Labor Market Reform," Journal of Marriage and Family 75(2): 503-521.
Hox, J. J. (ed.) (2002), Multilevel Analysis. Techniques and Applications. Mahwah, NJ: Lawrence Erlbaum.
Jacobs, J. A., & Gerson, K. (2004), The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.
Johnson-Hanks, J., Bachrach, C., Morgan, S. P., & Kohler, H. P. (2012), Understanding Family Change and Variation: Structure, Conjuncture, and Action. New York: Springer.
Kasl, S. V., Gore, S., & Cobb, S. (1975), "The Experience of Losing a Job: Reported Changes in Health, Symptoms and Illness Behavior," Psychosomatic Medicine 37(2): 106-122.
Kohler, H. P., Billari, F. C., & Ortega, J. A. (2002), "The Emergence of Lowest Low Fertility in Europe during the 1990s," Population and Development Review 28(4): 641-680.
Kohler, H. P., & Kohler, I. (2002), "Fertility Decline in Russia in the Early and Mid1990s: The Role of Economic Uncertainty and Labour Market Crises," European Journal of Population 18(3): 233-262.
Kreyenfeld, M., Andersson, G., & Pailhe, A. (2012), "Economic Uncertainty and Family Dynamics in Europe: Introduction," Demographic Research 27(28): 835-852. doi: 10.4054/DemRes.2012.27.28
Kruger, H., & Levy, R. (2001), "Linking Life Courses, Work, and the Family: Theorising a Not So Visible Nexus between Women and Men," Canadian Journal of Sociology / Cahiers canadiens de sociologie 26(2): 145-166.
Kuhn, U. (2009), "Collection, Construction and Plausibility Checks of Income Data in the Swiss Household Panel," Swiss Household Panel Documentation. Vols. 112. Lausanne: FORS.
Le Goff, J.-M., Sauvain-Dugerdil, C., Rossier, C., & Coenen-Huther, J. (2005), Maternite et Parcours de Vie. Bern: Peter Lang.
Levy, R., Widmer, E., & Kellerhals, J. (2002), "Modern Family or Modernized Family Traditionalism? Master Status and the Gender Order in Switzerland," Electronic Journal of Sociology 6(4).
Levy, R., Gauthier, J-A., & Widmer, E. (2006), "Entre contraintes institutionnelles et domestiques: Les parcours de vie masculins et feminins en Suisse," Canadian Journal of Sociology/Cahiers canadiens de sociologie 31(4): 461-489.
Lipps, O. (2010), "Income Imputation in the Swiss Household Panel 1999-2007," FORS Working Paper Series, Lausanne, http://www2.unil.ch/fors/IMG/pdf/ FORS_WPS_2010-01_Lipps (accessed: Dec 8 2014)
Lipps, O. (2006), "Attrition in the Swiss Household Panel: Wave 2 through Wave 7," Working Paper 4, Swiss Household Panel, Neuchatel, 1-19.
McQuillan, J., Greil, A. L., Shreffler, K. M., & Bedrous, A. V. (2015), "The Importance of Motherhood and Fertility Intentions among US Women," Sociological Perspectives 58(1): 20-35.
Miller, W. B., & Pasta, D. J. (1995), "Behavioral Intentions: Which Ones Predict Fertility Behavior in Married Couples?," Journal of Applied Social Psychology, 25(6): 530-555.
Miller, W. B., & Pasta, D. J. (1996), "Couple Disagreement: Effects on the Formation and Implementation of Fertility Decisions," Personal Relationships 3(3): 307-336.
Mulder, C. H., & Hooimeijer, P. (1999), "Residential Relocations in the Life Course," in van Wissen, L. J. G. & Dykstra P. A. (eds.), Population Issues. An Interdisciplinary Focus. Kluwer Academic/Plenum Publishers: New York, 159-186.
Neyer, G., Lappegard, T., & Vignoli, D. (2013), "Gender Equality and Fertility: Which Equality Matters?," European Journal of Population/Revue Europeenne de Demographie 29(3): 245-272.
Olah, L. S. (2003), "Gendering Fertility: Second Births in Sweden and Hungary," Population Research and Policy Review 22(2): 171-200.
Olah, L. S., & Bernhardt, E. M. (2008), "Sweden: Combining Childbearing and Gender Equality," in Tomas Freijka (ed.), Demographic Research, Special Collection 7: Childbearing Trends and Policies in Europe ?9(article 28), 1105- 1144.
Ozcan, B., Mayer, K. U., & Luedicke, J. (2010), "The Impact of Unemployment on the Transition to Parenthood," Demographic Research 23(29): 807-846.
Pettit, B., & Hook, J. L. (2009), Gendered Tradeoffs: Family, Social Policy, and Economic Inequality in Twenty-one Countries. New York: Russell Sage Foundation.
Philipov, D., & Bernardi, L. (2011), "Concepts and Operationalisation of Reproductive Decisions Implementation in Austria, Germany and Switzerland," Comparative Population Studies--Zeitschriftfur Bevolkerungswissenschaft 36(2/3): 495-530.
Philipov, D., Speder, Z., & Billari, F. C. (2006), "Soon, Later, or Ever? The Impact of Anomie and Social Capital on Fertility Intentions in Bulgaria (2002) and Hungary (2001)," Population Studies 60(3): 289-308.
PRB (2014), 2014 World Population Data Sheet. Washington: Population Reference Bureau. http://www.prb.org/Publications/Datasheets/2014/2014-world- populationdata-sheet.aspx (accessed: Dec 8 2014).
Raudenbush, S. W., & Bryk, A. S. (2002), Hierarchical Linear Models: Applications and Data Analysis Methods. Vol. 1. Thousand Oaks, CA: Sage.
Rindfuss, R. R., & Brewster, K. L. (1996), "Childrearing and Fertility," Population and Development Review 22: 258-289.
Rindfuss, R. R., Guzzo, K. B., & Morgan, S. P. (2003), "The Changing Institutional Context of Low Fertility," Population Research and Policy Review 22(5): 411438.
Rossier, C., & Bernardi, L. (2009), "Social Interaction Effects on Fertility: Intentions and Behaviors," European Journal of Population/Revue Europeenne de Demographie 25(4): 467-485.
Rubery, J., Smith, M., Anxo, D., & Flood, L. (2001), "The Future European Labor Supply: The Critical Role of the Family," Feminist Economics 7(3): 33-69.
Schmitt, C. (2012). A Cross-National Perspective on Unemployment and First Birth. European Journal of Population 28(3): 303-335, doi: 10.1007/s10680-012-9262-5
Singer, J. D., & Willett, J. B. (2003), Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence. Oxford: Oxford University Press.
Sobotka, T., Skirbekk, V., & Philipov, D. (2011), "Economic Recession and Fertility in the Developed World," Population and Development Review 37(2): 267-306.
Sobotka, T., & Zeman, K. (2011), "Postponement and Recuperation in Cohort Fertility: Austria, Germany and Switzerland in a European Context," Comparative Population Studies--Zeitschrift fur Bevolkerungswissenschaft 36(2/3): 417-452. doi: 10.4232/10.CPoS-2011-10en.
Valarino, I., & Bernardi, L. (2010), "Fertility Discourse in Parental Leave Policies' Media Coverage: A Frame Analysis of French-speaking Swiss Press Articles from 1999 to 2009," Population Review 49(2): 47-69.
Vignoli, D., Drefahl, S., & De Santis, G. (2012), "Whose Job Instability Affects the Likelihood of Becoming a Parent in Italy? A Tale of Two Partners," Demographic Research 26(2): 41-62.
Vignoli, D., Rinesi, F., & Mussino, E. (2013), "A Home to Plan the First Child? Fertility Intentions and Housing Conditions in Italy," Population, Space and Place 19(1): 60-71.
Wall, G. (2013), "Putting Family First: Shifting Discourses of Motherhood and Childhood in Representations of Mothers' Employment and Child Care," Women's Studies International Forum 40: 162-171.
Westoff, C. F., & Ryder, N. B. (1977), "The Predictive Validity of Reproductive Intentions," Demography 14(4): 431-453.
Table 1 Descriptive Sample Statistics Childless Childless women men N 552 588 Dependent variable Intention to have a child within 24 months * Yes 29.3% 26.0% * No 59.2% 60.2% * Do not know / no answer 11.4 % 13.8% Sociodemographic factors Age groups * Mean age 31.2 years 34.05 years * 22-30 years of age 53.6% 39.6% * 31-39 years of age 29.2% 33.3% * 40-45 years of age 17.2% 16.5% * 46-50 years of age 10.5% Civil status * Single, never married 59.6% 54.4% * Married 36.1% 36.2% * Separated 0.7% 1.2% * Divorced 3.4% 7.8% * Widower/widow 0.2% 0.3% Level of education * Low education 8.3% 5.3% * Middle education 52.4% 45.2% * High education 39.3% 49.5% Being active on the labor market * Active on the labor market 98.4% 98.3% * Unemployed 1.6% 1.7% Yearly income: mean, CHF net * Yearly individual net 50138.45CHF 70621.50CHF * Yearly household net 123010.90CHF 128081.75CHF Number of children * One * Two * Three or plus Explanatory variables Job quality measures * Job stability 79.3% 80.6% * Job instability 18.3% 17.0% * Occupational prestige: mean 47.13 47.56 Gender-role attitude: child suffers with a working mother * Not at all 53.3% 38.9% * Yes 46.0% 59.7% Women with Men with at at least one least one N child 923 child 1119 Dependent variable Intention to have a child within 24 months * Yes 19.2% 21.6% * No 55.3% 54.6% * Do not know / no answer 25.6% 23.8% Sociodemographic factors Age groups * Mean age 36.3 years 38.39 years * 22-30 years of age 12.8% 7.5% * 31-39 years of age 58.3% 41.45% * 40-45 years of age 28.9% 31.7% * 46-50 years of age 19.3% Civil status * Single, never married 5.1% 6.0% * Married 89.7% 90.3% * Separated 0.9% 0.4% * Divorced 4.1% 3.2% * Widower/widow 0.2% 0.1% Level of education * Low education 11.6% 5.1% * Middle education 60.2% 44.8% * High education 28.2% 50.1% Being active on the labor market * Active on the labor market 98.4% 98.5% * Unemployed 1.6% 1.5% Yearly income: mean, CHF net * Yearly individual net 30718.80CHF 88519.45CHF * Yearly household net 113172.35CHF 114486.25CHF Number of children * One 37.7% 38.5% * Two 42.0% 42.1% * Three or plus 20.3% 19.4% Explanatory variables Job quality measures * Job stability 78.3% 87.7% * Job instability 17.0% 10.6% * Occupational prestige: mean 44.24 46.73 Gender-role attitude: child suffers with a working mother * Not at all 58.2% 37.4% * Yes 40.5% 61.8% Note: Data are from the Swiss Household Panel (2002-2011); authors' calculations. Table 2 Results of multilevel logit models for first-child intentions of childless women and men Model 1 Childless Childless N; Observations women men 552; 1254 588; 1338 Fixed effects Intercept -7.772 * -4.128 (0.0004) (0.016) Job quality measures Job instability -0.853 ** 0.065 (ref. stability) (0.426) (1.067) Occupational prestige 0.009 0.005 (1.009) (1.005) Gender-role attitude Child suffers with working mother Interactions Prof prest. * child suffers with working mother Instability * child suffers with working mother Socio-demographic factors Age -0.085 *** -0.052 *** (0.918) (0.949) Low level of education -0.553 -0.047 (ref. middle) (0.575) (0.954) High level of education 0.352 0.362 (ref. middle) (1.422) (1.436) Household income 0.390 0.073 (1.477) (1.076) Contribution to the hh income 0.374 0.354 (1.454) (1.424) Marital status 0.959 *** 0.892 *** (1=married) (2.609) (2.441) Partner characteristics Partner's child intention 3.467 *** 3.468 *** (32.035) (32.070) Partner at home 1.229 0.507 (ref. part time) (3.417) (1.661) Partner training -0.005 -0.982 (+) (ref. part time) (0.995) (0.374) Partner full time 0.102 0.366 (ref. part time) (1.107) (1.442) Partner jobless -0.043 -0.921 (ref. part time) (0.958) (0.398) Random effect Intercept 1.093 1.180 Log-Likelihood -1593.693 -1677.947 Model 2 Childless Childless N; Observations women men 552; 1254 588; 1338 Fixed effects Intercept -8.389 * -4.516 (0.0002) (0.011) Job quality measures Job instability -0.613 -0.250 (ref. stability) (0.542) (0.778) Occupational prestige 0.012 0.023 (+) (1.012) (1.024) Gender-role attitude Child suffers with working mother 0.630 1.462 * (1.877) (4.315) Interactions Prof prest. * child suffers with -0.006 -0.036 ** working mother (0.994) (0.964) Instability * child suffers with -0.684 0.589 working mother (0.505) (1.802) Socio-demographic factors Age -0.087 *** -0.053 ** (0.917) (0.948) Low level of education -0.593 -0.047 (ref. middle) (0.552) (0.954) High level of education 0.393 (+) 0.349 (ref. middle) (1.482) (1.418) Household income 0.416 0.035 (1.515) (1.035) Contribution to the hh income 0.447 0.466 (1.564) (1.594) Marital status 0.965 *** 0.882 *** (1=married) (2.626) (2.415) Partner characteristics Partner's child intention 3.483 *** 3.496 *** (32.567) (32.981) Partner at home 1.020 0.501 (ref. part time) (2.774) (1.650) Partner training -0.006 -0.939 (ref. part time) (0.994) (0.391) Partner full time 0.075 0.375 (ref. part time) (1.078) (1.456) Partner jobless -0.039 -0.941 (ref. part time) (0.961) (0.390) Random effect Intercept 1.075 1.184 Log-Likelihood -1600.099 --1679, 017 Note: Data are from the Swiss Household Panel (2002-2011); authors' calculations. Logit Models, coefficients, and odds ratio (round brackets). (+) p [less than or equal to] .1 ; * [p.bar] [less than or equal to] .05 ; ** [p.bar] < .01 ; *** p < .001. Mode of estimates: restricted PQL. Table 3 Results of multilevel logit models for subsequent childbearing intentions of mothers and fathers Model 3 Women with at least one Men with at child least one child N; Observations 923; 2771 1119; 3839 Fixed effects Intercept -3.892 -5.908 (+) (0.020) (0.002) Job quality measures Job instability 0.096 0.222 (ref. stability) (1.101) (1.249) Occupational prestige 0.009 0.007 (1.009) (1.007) Gender-role attitude Child suffers with working mother Interactions Prof prest. * child suffers with working mother Instability * child suffers with working mother Socio-demographic factors Age -0.075 * -0.107 *** (0.927) (0.899) Low level of education -0.372 0.991 * (ref. middle) (0.689) (2.693) High level of education 0.258 0.168 (ref. middle) (1.294) (1.183) Household income -0.146 0.161 (0.864) (1.175) Contribution to the hh income 0.562 -0.558 (1.754) (0.572) Marital status 0.233 -0.310 (1=married) (1.263) (0.733) Partner characteristics Partner's child intention 4 329 *** 4.358 *** (75.876) (78.071) Partner at home -0.217 0.012 (ref. part time) (0.805) (1.013) Partner training -0.005 1.323 (ref. part time) (0.995) (3.755) Partner full time 0.405 -0.114 (ref. part time) (1.499) (0.892) Partner jobless -0.547 0.686 (ref. part time) (0.578) (1.985) Age of the youngest child Aged between 0-3 1.407 *** 1.377 *** (ref. 6-12 years old) (4.085) (3.963) Aged between 4-5 0.941 ** 0.449 (+) (ref. 6-12 years old) (2.562) (1.566) 13 years old and older -0.602 -0.141 (ref. 6-12 years old) (0.547) (0.868) Random effect Intercept 1.182 1.121 Log-Likelihood -3100.743 -4404.816 Model 4 Women with at least one Men with at child least one child N; Observations 923; 2771 1119; 3839 Fixed effects Intercept -3.469 -6.399 (+) (0.031) (0.002) Job quality measures Job instability 0.307 -0.438 (ref. stability) (1.359) (0.645) Occupational prestige 0.002 0.021 (+) (1.002) (1.021) Gender-role attitude Child suffers with working mother -0.771 1.099 (0.462) (3.002) Interactions Prof prest. * child suffers with 0.021 -0.026 (+) working mother (1.021) (0.974) Instability * child suffers with -0.651 1.140 * working mother (0.522) (3.125) Socio-demographic factors Age -0.074 * -0.107 *** (0.928) (0.898) Low level of education -0.300 0.946 * (ref. middle) (0.741) (2.577) High level of education 0.292 0.163 (ref. middle) (1.340) (1.177) Household income -0.161 0.147 (0.851) (1.159) Contribution to the hh income 0.616 -0.546 (1.851) (0.579) Marital status 0.229 -0.305 (1=married) (1.257) (0.737) Partner characteristics Partner's child intention 4.362 *** 4.425 *** (78.390) (83.498) Partner at home -0.199 0.036 (ref. part time) (0.819) (1.036) Partner training 0.039 1.280 (ref. part time) (1.040) (3.598) Partner full time 0.408 -0.050 (ref. part time) (1.503) (0.951) Partner jobless -0.591 0.650 (ref. part time) (0.553) (1.917) Age of the youngest child Aged between 0-3 1.421 *** 1.362 *** (ref. 6-12 years old) (4.140) (3.903) Aged between 4-5 0.955 ** 0.427 (ref. 6-12 years old) (2.599) (1.533) 13 years old and older -0.590 -0.153 (ref. 6-12 years old) (0.554) (0.858) Random effect Intercept 1.185 1.134 Log-Likelihood -3093.488 -4387.659 Note: Data are from the Swiss Household Panel (2002-2011); authors' calculations. Logit Models, coefficients, and odds ratio (round brackets). (+) p [less than or equal to] .1 ; * p [less than or equal to] .05 ; ** p [less than or equal to] .01 ; *** p [less than or equal to] .001. Mode of estimates: restricted PQL.
|Printer friendly Cite/link Email Feedback|
|Author:||Hanappi, Doris; Ryser, Valerie-Anne; Bernardi, Laura|
|Publication:||Journal of Research in Gender Studies|
|Date:||Jan 1, 2016|
|Previous Article:||The road to Mandalay: orientalism, "Burma girls" and Western music.|
|Next Article:||Bullying experiences among sexual minority youths in England: the nature, prevalence and association with life satisfaction.|