1 Introduction

Does gender equality matter for fertility? This question has led to a fervent debate among demographers who seek explanations for the differentials in low fertility levels in modern societies (Puur et al. 2008; Westoff and Higgins 2009; Goldscheider et al. 2010; Toulemon 2011; Philipov 2011; Oláh 2011; Neyer 2011). Low fertility (TFR below 1.5) and low intentions to have children are often seen as the result of gender inequality in areas of life that have been recognized as essential for childbearing in contemporary societies: employment, economic resources, and household and care work. However, studies on the effects of gender equality on fertility intentions and fertility behavior render inconclusive results (Puur et al. 2008; Westoff and Higgins 2009; Goldscheider et al. 2010). We argue in this paper that this may be partly attributed to issues of how demographers conceptualize and perceive gender equality in their studies.

First, most studies focus on analyzing the impact of gender equality on the childbearing intentions and on the childbearing behavior of women. Less attention is paid to its effect on men’s childbearing behavior or on their plan for children. Since gender equality affects both women’s and men’s lives, the omission of men may give a distorted picture of the association between gender equality and fertility (see also Watkins 1993). Second, micro-level studies on the relationship between gender equality and fertility often concentrate on the impact of the gender division of household work on fertility intentions or fertility behavior. Limiting the investigation to gender equality in the private sphere provides only a partial explanation for the linkage between gender equality and fertility. Gender is a structuring element of all relationships in societies (Scott 1986). We therefore need to examine the influence of gender equality on fertility intentions and fertility behavior in different areas of life, such as in employment, in household work, in care, and so forth. We also need to explore the impact of gender equality on fertility intentions and behavior in different life-course constellations, such as across parities. Third, most demographic studies use the gender distribution of a specific indicator (such as the gender distribution of household work) as the measure for the degree of gender equality. Such an approach collapses different aspects inherent in the concept of gender equality into one dimension. Four, the gender distribution describes the degree of gender difference, but it does not necessarily bring out gender inequality, that is, the unequal gender distribution of power, agency, justice or fairness ingrained in gender differences. To understand the linkage between gender equality and fertility, we need concepts that capture such aspects of gender equality (Greenhalgh 1990; Watkins 1993; Presser 1997). For micro-level studies this implies that we need to regard gender differences not merely from the perspective of “performance”, that is, as differences in doing, but we also need to regard them as “power”, that is, as constituting unequal relationships, unequal life chances or unequal opportunities to act. We thus need an approach that allows us to distinguish between “gender as performance” and “gender as power”, and capture the dimensions of power, agency, justice, and fairness in the gender differences in doing (see also Ferree 2010).

In this paper we outline such an approach, drawing on the gender-equity approach proposed by Fraser (1994, 2008), and McDonald (2000a, b) and on feminist research on concepts and meanings of gender equality (Scott 1988; Orloff 1993; Phillips 2004, 2006). We exemplify our approach concentrating on specific dimensions of gender equality in employment, financial resources, and family work. These three areas of life are considered essential in structuring gender relationships and in shaping fertility decisions and fertility behavior. We investigate the impact of gender equality in these areas on short-term childbearing intentions, and we do this separately for women and for men as well as for different parities. We concentrate on the link between gender equality and fertility intentions, but our approach is also applicable to fertility behavior. Fertility intentions have been generally regarded as a fairly suitable predictor of actual behavior at the individual level (Westoff and Ryder 1977; Rindfuss et al. 1988; Schoen et al. 1999), provided one distinguishes intentions by parity, considers age and partnership dynamics, includes a time frame for the realization of the intention, and assumes that the conditions at the time of interview, in particular a person’s or couple’s economic conditions, persist (Thomson 1997; Schoen et al. 1999; Quesnel-Vallée and Morgan 2003; Billari et al. 2009; Régnier-Loilier and Vignoli 2011; Balbo and Mills 2011).Footnote 1 By looking at the link between various dimensions of gender equality and fertility intentions we get a good grip at which dimensions of gender equality ease or hamper considerations to have a(nother) child for women and for men.

Our paper proceeds as follows: We first give an overview of recent studies of fertility intentions and of the relationship between gender equality in employment, financial resources, and family work on the one hand and fertility intentions on the other hand. We include research on fertility wherever necessary to complement the still rare research on gender equality and fertility intentions across our areas of interest. We proceed with theoretical reflections on the measurement of gender equality in fertility studies, followed by a suggestion for how gender equality could be conceptualized in such a way that it captures the gender-equality relevant meaning of employment, financial resources, and family work. We then present the results of our empirical analysis based on an application of these notions. We conclude with some reflections on our results, on issues of indicators of gender equality, and on the implications of our approach and findings for research.

2 Gender Equality and Fertility: Does Equality Matter?

A number of studies related to Western Europe point to the importance of gender equality for fertility intentions and fertility behavior, but whether gender equality promotes them (as many wish) is contested. The results of empirical analyses vary considerably depending on which indicators of gender equality are included, whether women or men are studied, which parity transition, and which country is considered in the analysis (for contrasting examples, see: Mills et al. 2008; Philipov 2008; Puur et al. 2008; Sobotka and Testa 2008; Neyer and Rieck 2009; Westoff and Higgins 2009; Goldscheider et al. 2010; Mills 2010; Miettinen et al. 2011). This applies in particular to the many studies on gender equality and fertility intentions that use gender role attitudes, that is the individual identification with cultural stereotypes of “doing gender”, as indicators of gender equality (Kaufman 2000; Philipov 2008; Puur et al. 2008; Westoff and Higgins 2009; Goldscheider et al. 2010; Miettinen et al. 2011). It also applies to the still rare research that explores the effects of gender equality regarding employment, financial resources, and family work—and thus the effect of (un)equal gender relationships in these areas—on fertility intentions. Some studies show that being in employment increases women’s intentions to have a child in the next few years or at some unspecified time in the future. Yet, this may only apply to childless women, to women in specific countries or to full-time employed women (Philipov 2009a; Neyer and Rieck 2009; Begall and Mills 2011; Testa 2012; Modena and Sabatini 2012; Vignoli et al. 2013). Others find no such effects or their results indicate that employed women tend to have lower fertility intentions than non-employed women, even if the same countries or parities are studied (Sobotka and Testa 2008; Pailhé 2009; Balbo and Mills 2011; Iacovou and Tavares 2011; Schmitt 2012; Matysiak and Vignoli 2008 for a meta-analysis). The findings for men are less contradictory than those for women: Irrespective of parity or country, men in employment display higher fertility intentions or their employment status has no effect on their childbearing intentions (Sobotka and Testa 2008; Schmitt 2012; Neyer and Rieck 2009; Philipov 2009a; Balbo and Mills 2011; Iacovou and Tavares 2011; Modena and Sabatini 2012; Testa 2012). Men also tend to want a child, if their partner is not employed (Sobotka and Testa 2008; Berninger et al. 2011), but this may vary by parity (Neyer and Rieck 2009).

Just as for employment, the economic resources available to a person have no uniform effect on fertility intentions either. Mills (2010) concludes from her study on the relationship between gender indexes and fertility intentions across European countries that economic security supports fertility intentions for women as well as for men. Modena and Sabatini (2012) corroborate this result. They find for Italy that poor economic well-being decreases fertility intentions substantially. By contrast, Sobotka and Testa (2008) for Europe, Berninger et al. (2011) for Germany, Iacavou and Tavares (2011) for Great Britain, Fiori et al. (2013) for Italy find no effect, weak effects or inconsistent effects of financial resources or economic (in)security on childbearing intentions of women and of men. These varying effects are confirmed by studies on the association between economic resources and childbearing. Limited financial resources may affect the childbearing behavior of some socio-economic groups, but not of others (Kohler and Kohler 2002; Aassve et al. 2006; Kreyenfeld 2010; Vignoli et al. 2012).

Greater gender equality in the division of household work and care is generally assumed to increase fertility intentions and childbearing, but research does not confirm this consistently (Oláh 2003; Duvander and Andersson 2006; Brodmann et al. 2007; Esping-Andersen et al. 2007; Mills et al. 2008; Duvander et al. 2010; Lappegård 2010; Fiori 2011). The results tend to depend on the country studied, on the burden of work put on women through employment or through the number of children, and on the share of fathers’ involvement (Cooke 2004; Torr and Short 2004; Esping-Andersen et al. 2007; Mills et al. 2008; Craig and Siminski 2010, 2011). There are also indications that the effect may vary depending on whether the father engages in childcare or in general household work, and whether the woman is satisfied with the division of family work, be it equal or not (Oláh 2003; Cooke 2004; Rosina and Testa 2009; Craig and Siminski 2010, 2011). We have found no study on fertility intentions that disentangles household work, care, and satisfaction with them, nor have we found studies that examine the effects of these on men’s fertility intentions.

The studies that we have reviewed draw contrasting pictures of the relationship between gender equality and fertility intentions: Some research provides evidence that greater gender equality in employment, economic resources, and family work increases fertility intentions, other research reports converse results or at best finds no impact of gender equality on fertility intentions or fertility behavior. Part of the variation may be explained by differences in national contexts, by the parity studied, and by the coding of the variables included in the analysis. However, contrasting results for the same countries indicate that neither an overall positive nor an overall negative association between gender equality and fertility can generally and unconditionally be confirmed.

3 Conceptualizing Gender Equality in Employment, Financial Resources, and Family Work

The lack of a uniform fertility effect of gender equality in employment, financial resources, care, and household work shows that one cannot just assume that there is a monotonic relationship between gender equality and fertility. The empirical results show that less gender equality does not always imply lower fertility intentions or less childbearing and more gender equality does not necessarily lead to higher fertility intentions or more childbearing. This challenges the very concept of gender equality, which forms the key commonality in empirical demographic studies, namely, that gender equality means “sameness of distribution”. The results rather underline the need to find a different notion of gender equality. Assuming that the same distribution for women and for men represents gender equality may be misleading. It may brush over gender differences that exist despite the same distribution and that influence fertility intentions and fertility behavior. It ignores that “the same” may mean something else for childless women than for childless men, for mothers than for fathers, and that it may have different implications for either of them. It does not allow us to distinguish between differences that do not imply inequality and differences that do imply inequality, that is, differences that represent an unequal distribution of power, resources, agency, justice, and fairness. Such indistinctiveness and such discords between sameness of distribution and gender equality cast doubt on the notion that the relationship between gender equality and fertility intentions or fertility behavior can be adequately measured if gender equality is conceived as sameness of distribution only.

Fraser (1994) proposes to replace the uni-dimensional, sameness-based concept of gender equality by the concept of gender equity, that is, by an understanding of gender equality that stresses fairness and social justice. She argues that using gender equity as the baseline concept for gender equality would force us to define the criteria of what is fair and just from a gender perspective. Such an approach would better capture the complexity of gender equality and would overcome the shortcomings of sameness of distribution as the principal measure of gender equality. It would allow for gender differences, but capture gender inequality. McDonald (2000a, b) applies Fraser’s concept to fertility research at the macro-level. He points out that in order to understand the linkage between gender equity and fertility we need to distinguish between gender equity in the institution of the family and gender equity in individual-oriented institutions such as employment and education. The development towards low and lowest-low fertility is associated with different (and incongruent) developments of gender equity in the family and gender equity in individual-oriented institutions (McDonald 2000b). The very low fertility levels in many advanced Western societies are the result of cleavages between high gender equity in individual-oriented social institutions and low gender equity in family-oriented institutions (such as familial childcare; McDonald 2000b). His work shows that even if we use the broader concept of gender equity, there is no uni-directional relationship between the development and the status of gender equality and fertility development.

Putting the emphasis on gender equity implies, however, that gender differences and even gender inequalities that are perceived as fair and just may not necessarily hamper childbearing intentions and reduce fertility. For example, in Becker’s (1981) prototype family, in which the strict gender segregation of employment and care is—by definition—fair, just, and to the benefit of all, fertility intentions and fertility behavior might not be negatively affected, even if from the perspective of resources and power, this family form is a manifestation of gender inequality, both at the level of the individual-oriented institution (employment) and at the level of the family. Similarly, Hakim’s (2000) “home-oriented women”, who are defined as preferring care and household work to paid employment, might not limit their fertility intentions or fertility behavior because—by definition—they regard staying home as a social right to free choice and their choice of a gendered division of care and work as fair and just. Both Becker and Hakim view gender inequality as a matter of rational choice and of individual choice-based gender performance; their concept of gender equity obscures the distinction between gender difference and gender inequality, and it conceals aspects of power, resources, and agency. The fuzziness of the concept of gender equity may make it difficult to distinguish between the fertility effects of gender equality and the fertility effects of gender inequality that is perceived as fair and just. This may be one of the reasons why even studies that state that they investigate the effect of gender equity on fertility rely in essence on concepts of “sameness of distribution”.

McDonald (2000b) points out that the possibilities to directly investigate the effects of gender equity on childbearing intentions and fertility outcome in quantitative research are limited. Questions on the perception of fairness and justice, particularly with respect to individual-level institutions (such as employment) are rarely included in demographic questionnaires. Moreover, even when they are present, such questions would only grasp the individual perception of fairness but would not provide a basis for measuring gender equality across countries with different gender regimes. The individual perception of gender equity in a society may be mediated by the specificities of gender equality in this society. For example, in some countries, where access to employment is strongly gendered, such difference may be perceived as fair and just, while in countries with a more gender-equal access to the labor market, gender differences in employment may be regarded as manifestations of gender inequality (for a discussion see Korpi et al. 2013).

We therefore need an understanding of gender equality that overcomes the limitations of a pure measure of sameness, but which also goes beyond the confines of the concept of gender equity. Such an understanding should allow for gender differences, but should also identify gender inequalities (see also Bridges 2003). It should include an understanding of gender equity, that is, it should capture fairness and justice from the perspective of both women and men. Such a concept of gender equality would reflect the complexity and the multi-dimensional aspects of gender equality, and it would capture gender imbalances of power, agency, life chances, justice, and fairness without eliminating gender differences. To demonstrate the potential of such an approach, we look at specific aspects of gender equality that are regarded as essential to achieve gender equality. We have chosen three dimensions of gender equality that apply to both women and men across different societies, and which address different areas of life. These dimensions also capture gender equality in employment, financial resources, household work and care: “forming and maintaining a household”, “having agency and the capability to choose”, and “having fairness in the gender distribution of family work and care.” We briefly outline these three dimensions and their representation through employment, financial resources, household work and care.

3.1 To Form and Maintain a Household and a Family

The possibility to form and maintain one’s own household and family has been recognized as one of the most essential features of gender equality (Hobson 1990; Orloff 1993). It means that one has the option to live independently and to support oneself and one’s children. In all European societies, the primary source of means to maintain oneself and one’s family is employment. It usually provides the monetary basis that enables a person to set up a household, ensures her own and her family’s livelihood and grants economic independence and welfare protection over the life course. In most countries, this can only be achieved through full-time employment or through employment that secures an income on the level of full-time employment. Having full-time employment may thus be regarded as a proxy for a person’s capacity to form and maintain an autonomous household, to assure her independent social protection, and to maintain her bargaining power in a partnership. This usually distinguishes full-time employment from part-time work. Working part-time mostly implies less income, lower social-security benefits, a reduced capacity to sustain a household, and in couples with an unequal amount of paid work, it implies a reduced bargaining power in the relationship (Bittman et al. 2003).

3.2 Agency and the Capability to Choose

Agency and the capability to choose are also regarded as central aspects of gender equality (Sen 1992; Phillips 1999, 2004; Korpi 2000; Robeyns 2003; Ferree 2010; Hobson 2011). They depend largely on the resources that are available to a person and that enable her to act. Usually, a person’s resources are seen as an indicator of her standard of living, her social status in society and—as a consequence of her social position—as an indicator of her well-being. From a gender perspective, however, a person’s resources can also be regarded as a capability (Sen 1992), as a feature that enhances a person’s agency, that is, the scope of alternatives available to a person to choose activities that she values and which further her well-being (Sen 1992; Korpi 2000).

In modern societies, economic resources have become crucial to agency. As agency resources, economic resources are not simply a sign of possessions, of poverty, of wealth, or of economic status. They are an indicator of the power to act, of the range of choices a person has, of the capacity to participate in social activities, of the potential to decide one’s own life-course directions, and of the possibilities to do the things she values. This includes the option to have (a)nother child (Hobson and Oláh 2006; Hobson and Fahlén 2009; Hobson 2011). Yet, in order to grasp agency, the range of choices, and the power to act, it is not sufficient to rely solely on a quantitative measure of economic resources, such as on the level of income. The range of choices and the capability to act do not necessarily depend on whether a person is rich or poor by objective standards; they rather depend on whether she perceives her economic situation as constrained or not.

3.3 Fairness in the Gender Distribution of Family Work and Care

The distribution of household and care work within a partnership is considered a prime indicator of gender equity. However, using the distribution of household and care work as a sign of gender equity mixes up the actual distribution of work with the perceived fairness of the distribution. A gendered division of household and care work may be the outcome of a joint decision by a couple, in which case one would expect each partner to regard the division as fair even if it is unequal. Moreover, whether the distribution of family work is conceived as fair may differ between women and men, since the gains and losses of family engagement may also be distributed unequally between women and men. We therefore need to distinguish between the gender distribution of household chores and of care work on the one hand and the perceived fairness on the other hand, if we want to assess the effect of gender equality/gender equity in the family on childbearing.

The three dimensions we have outlined focus on the gender-relevant meaning of employment, economic resources, and the division of household work and familial care. They incorporate the notion of gender equity, but focus on gender equality, in that they allow for gender differences but aim at capturing gender inequality. They should thus allow us to overcome the limitations of purely distributional sameness-based measures of gender equality in fertility research.

4 The Three Dimensions of Gender Equality and Fertility Intentions: Findings from the Generations and Gender Surveys (GGS)

We illustrate the potential of our conceptual approach by investigating the relationship between these dimensions of gender equality and fertility intentions. To this end we make use of data from the first wave of GGS of Eastern and Western European countries.Footnote 2 The GGS was specifically designed to facilitate research on the relationship between gender aspects, family dynamics, and fertility intentions. Unlike many other surveys, the GGS provides data for both women and men. Women and men were asked separately about their childbearing intentions; they are thus seen as agents of reproduction in their own right. This allows us to study the impact of gender equality on fertility intentions of women and of men alike. We use the first wave of the GGS of Austria, Bulgaria, France, Germany, Georgia, Norway, Romania, and Russia, as well as the Hungarian Survey “Turning Point of the Life Course”, the Italian Multipurpose Household Survey on “Family and Social Subjects”, and the “Netherlands Kinship Panel Study”. The latter three surveys incorporate large parts of the GGS and are part of the Generations and Gender Programme. With the exception of the Italian data, all national datasets were harmonized and made available by the project “GGP-Design Studies for Research Infrastructure” funded through the 7th Framework Programme (FP7) of the EU (Grant 212749); the data for Italy were provided by ISTAT (the Italian National Institute of Statistics). The fieldwork of the first wave of the GGS was carried out in the various countries between 2001 (Hungary) and 2008/2009 (Austria), with most fieldwork taking place in 2004/2005. All GGSs comprise information on women and on men aged 18–79. For our study on childbearing intentions, we limited the sample to non-pregnant women aged 18–42 and to men aged 18–49 at the time of the interview. We chose these age ranges because the decision to have a child beyond these ages may be less influenced by economic, private, and gender-equality considerations than the decision to have a child at a socially accepted childbearing age.Footnote 3 Moreover, very few of the interviewed women and men outside these ages intended to have a(nother) child. Since we are interested in the impact of gender equality, we furthermore limited our sample to women and men who lived in a (married or cohabiting) partnership at the time of the interview.Footnote 4

Our investigation focuses on the intention to have a child within the next 3 years (as reported at the interview date). By limiting the question about childbearing intention to a foreseeable time period we overcome some of the problems normally associated with the surveying of intentions. Answers to questions about an individual’s fertility intention in general, such as “how many children do you intend to (ever) have”, are likely to capture a social norm, that is, the number of children individuals think they should have rather than what they believe they will have. Such general questions therefore tend to lead to answers which confound intentions and social norms. Questions on intentions that cover an overseeable time period and that therefore are “in close temporal proximity to the prospective behavior” (Ajzen and Fishbein 1973, p. 49) are generally considered to be the more suitable predictors of actual behavior (Philipov 2009b). They offer the possibility to draw inferences from a person’s current status about conditions that are crucial in her decision process to have a(nother) child. Positive short-term fertility intentions have proved to be a valid predictor of subsequent outcomes, although they tend to partly overestimate fertility realizations (Spéder and Kapitány 2009; Régnier-Loilier and Vignoli 2011).

We study women’s and men’s childbearing intentions separately, because parenthood has different consequences for women than for men, in particular with respect to employment and to household/care work (Sanchez and Thomson 1997; Mencarini and Tanturri 2004; Misra et al. 2007). Employment and financial aspects as well as the gender division of household and care work may therefore have different effects on the childbearing intentions of women and of men. We also analyze the intentions to have a first child, a second child, or three or more children separately, because the relevant issues may play out differently for the various parities. For women, the birth of the first child, more so than the birth of subsequent children, often marks a critical juncture for gender equality (Sanchez and Thomson 1997; Neyer and Rieck 2009). Correspondingly, the decision to have three or more children often means going beyond the widely acknowledged norm of two children and may therefore be motivated by other economic, social, and gender-equality aspects than the decision to have a second child (Hoem et al. 2001).

We use logistic regressions with the intention to have or not to have a(nother) child within the next 3 years as the dependent outcome.Footnote 5 We estimate the influence of employment, of the respondent’s perception of her family’s financial situation, of the division of and the satisfaction with care and with household work on women’s and men’s intention to have a first, a second, or a third or subsequent child separately. If one of these key dimensions was not collected in a country-specific survey, we excluded the corresponding country from the respective model. We present the results of the impact of each dimension on fertility intentions separately, while the results of the full model (i.e., all dimensions in one model) are reported in the Appendix (Tables 6, 7).

At the onset of the analysis, we estimated a first model including only country as an explanatory variable in order to appreciate the gradient of fertility intentions across countries.Footnote 6 In subsequent models, we adjusted the estimates by intra-group correlation (at the level of the country) instead of adding the country as a control variable. This approach specifies that the standard errors allow for (country) intra-group correlation, relaxing the usual requirement that the observations are independent. That is, the observations are independent across clusters (i.e., countries) but not within clusters. In this way we acknowledge that there may be greater similarities in fertility intentions and their determining gender-equality factors among respondents living in one country than among respondents living in different countries. In essence, this means that we recognize differences in country contexts in our models.

We control for the respondent’s age, her union status, her educational attainment, and the country she lives in, all at the time of the interview. The respondent’s age is coded as below age 30 or above age 30 (up to the specified maximal age for women and for men). Following the ISCED classification of educational levels we grouped the respondent’s educational attainment into the three standard levels: basic education, secondary and upper secondary education, and post-secondary and tertiary education. The respondent’s union status differentiates between cohabiting and married couples. We also include the partner’s employment status, coded as employed or not employed,Footnote 7 and in models of childbearing intentions of parents we include the age of the youngest child, coded as below age 3 or age 3 and above. In all analyses of childbearing intentions of parents with two or more children, we also control for the number of children these parents have. The dataset for our study comprises 3,622 childless women, 6,552 mothers with one child, and 12,386 mothers with two or more children, as well as 3,674 childless men, 6,106 fathers with one child, and 12,290 fathers with two or more children. For simplicity, in what follows, we only illustrate the results of the key variables we have selected as markers of the three dimensions of gender equality. We do not report the results of the control variables because they are very much in line with previous research. Discussing their influence on fertility intentions would also take us too far from the main aim of the paper.

4.1 Country Idiosyncrasies

To give an overview of childbearing intentions across countries, we present predicted probabilities as well as odds ratios. Childless women and childless men in Eastern Europe are much more inclined to intend to have a child within the next three years than childless women and men in Western Europe (with the partial exception of Italy) (see Table 1). The higher intention rates in Eastern European countries correspond to the pattern of universal childbearing in these countries: Almost all women and men in these countries become parents; they do so at a comparatively young age and soon after forming a union (Kesseli 2007; Frejka et al. 2008). The particularly high childbearing intentions of childless women and men in Georgia are attributable to the very close connection between partnership formation and childbearing in this country: Once married, everyone tends to want to have a child as soon as possible (Olds and Westoff 2004; Baduraschvili et al. 2008).Footnote 8

Table 1 Odds ratio and predicted probabilities of women’s and men’s intentions to have a(nother) child in the next 3 years across selected European countries
Full size table

Georgian women and men are also more inclined to want a second or subsequent child within the next 3 years than mothers and fathers in other European countries (see Table 1). Apart from this, the clear East–West difference found for fertility intentions among childless women and men largely vanishes both for mothers and for fathers. We find no discernible country pattern of second-birth intentions. Third- and higher-order childbearing intentions are somewhat more common among mothers in Western Europe than in Eastern Europe, but this does not hold for fathers (Table 1). When we interpret the results we have to keep in mind that we look at the intentions to have a child within the next 3 years; intentions to have a child in the more distant future are not captured. Nevertheless, the results indicate that a child is still something that most women and men (in a union) are inclined to have, in particular in Eastern Europe. The findings also reflect the general tendency towards small families (of two children at most); in almost all countries, only a minority of the respondents expressed intentions to have a third or subsequent child in the next 3 years.

4.2 Employment: The Possibility to Form and Maintain a Household

As we have highlighted, we use employment as an indicator of whether and to what extent a person can afford to form and maintain a household independently of the support of a partner or of others. We therefore differentiate between full-time, part-time, and no employment.

Our results (Table 2) show that the impact of employment on childbearing intentions varies considerably by gender and by parity. First, for both childless women and for childless men being in full-time employment is a pre-condition for considering parenthood in the near future. The importance of own employment also holds if we include the partner’s employment status in our analysis. In fact, including the partner’s employment status does not much change the effect and the gradient of the respondent’s own employment status on her or his childbearing intentions.

Table 2 Intentions to have a(nother) child in the next three years by women's/men's employment status and by women's/men's and their partners' employment status
Full size table

The rather gender-equal pattern as to the importance of one’s own employment for childless women’s and men’s intention to become parents in the near future vanishes once women and men have one child. Full-time employment loses its positive and significant predictive power for mothers’ intentions to have a second child in the next 3 years. However, if a mother’s partner is employed, her intentions to have another child are noticeably higher than if the partner is not employed.

This contrasts markedly with the childbearing intentions of fathers who have one child: Fathers who are employed are more prone to consider having a second child in the next 3 years than fathers who are not employed, while their partners’ employment has no decisive bearing on their intentions to have a second child in the near future.

There is also a gendered pattern of childbearing intentions among parents who have two or more children. Among mothers with two or more children, the employment status of their partners is still crucial in directing fertility intentions, irrespective of the woman’s own employment status. For fathers, the employment status has no impact on their childbearing intentions.

If, as suggested, full-time employment may be regarded as an indicator of the possibility to form and maintain one’s own household and to retain one’s bargaining power vis-à-vis a partner, then the results confirm that being able to support themselves (and their child) is an essential factor for (childless) women as well as for (childless) men in their consideration to become a parent in the near future. By contrast, parenthood clearly exerts a gendering effect on women’s and men’s intentions to have another child in the near future. Moreover, mothers and fathers own employment and that of their partner work in different directions. Mothers’ own employment does not affect their intentions of further childbearing, while fathers’ own employment is positively related to their intentions to have a child. For mothers, their partners’ employment increases their intention to have another child in the next 3 years; for fathers, the partners’ employment does not affect their intention to consider another child in the near future.

4.3 Financial Situation: Agency Possibilities and Capabilities to Choose

We regard the financial situation of women and men and whether they judge it as easy or tight as an indicator of women’s and men’s agency (Korpi 2000), that is, of their possibilities to participate in social life and to engage in activities that they value (Sen 1992), one of which may be having children. We consider a tight economic situation to constrain childbearing intentions, since having a child increases expenses for parents and—mostly for women—hampers their possibilities to pursue (let alone to increase) employment or other money-earning activities. To test the impact of the financial situation on the intentions to have a(nother) child in the next 3 years, we use the answer to the question whether it is difficult for the respondent to make ends meetFootnote 9 (Table 3). Our outcomes suggest that there is a slightly depressing, partly u-shaped influence of economic difficulties on short-term fertility intentions beyond the first child. This indicates that at each parity the economic capabilities are judged differently with regard to (further) childbearing. However, there are hardly any gender differences nor are the results consistently significant (Table 3; Tables 6, 7 in the Appendix). This lack of gender differences and of significance is surprising. Since, contrary to men, women usually deal with expenses covering basic, everyday household needs (such as food, body care, cleaning material, children’s diapers, toys, clothes, and so forth), one would expect that difficulties making ends meet exert a different influence on women’s childbearing intentions than on men’s. Given the impact and the gendered effects of employment and of unemployment on childbearing intentions, one would also expect to find some gender differences and some stronger influence of economic pressure on childbearing intentions. The absence of such differences suggests that one’s own and/or the partner’s (un)employment exert a different influence on childbearing intentions than the overall economic situation of the household does. Single-country analyses (not shown) confirm that the employment status and the economic situation of the household may influence childbearing intentions differently (see also Fiori et al. 2013). One of the reasons may be that (un)employment captures an objectively given situation, while “making ends meet” captures the respondent’s perception of her household’s economic situation, but does not disclose the objective basis for the tight economic situation. Our country-specific analyses indicate that in Western European countries, economic difficulties are not necessarily linked to unemployment; they can be caused, for example, by (long-term) investments that strain the financial situation temporarily but pay off in the long run, such as the purchase of a house and the loan or mortgage taken for it. In Eastern European countries, there seems to be a closer link between unemployment and economic difficulties, due to the high unemployment rates in these countries and the greater risks of long-term unemployment. Our single-country analyses reveal that, as a consequence, in most Eastern European countries, mothers and fathers who face economic difficulties are very reluctant to intend to have another child in the next 3 years, while we observe no fertility-depressing impact of economic difficulties on mothers and fathers in Western European countries.Footnote 10 Overall, the results suggest that looking at the perceived economic situation is not sufficient to assess the link between agency and fertility intentions. We need to distinguish whether a tight economic situation is brought about by volition (such as the purchase of housing) or whether the economic difficulties are brought about by something else than one’s own choice (such as unemployment) and are thus a sign of economic vulnerability.

Table 3 Intentions to have a(nother) child in the next three years by economic situation
Full size table

4.4 Division of Household and Care Work: Gender Equity in Family Work

Men’s participation in household and care work is recognized as the core indicator of gender equity in the family and it is also regarded as an important factor shaping fertility intentions and fertility outcomes. For our study, we constructed an indexFootnote 11 of men’s contribution to daily or recurrent household chores, namely preparing meals, doing the dishes, shopping for food, and doing the vacuum cleaning. We built a similar index of men’s involvement in childrearing tasks, that is, dressing the children, putting them to bed, playing with them, and staying at home with them when they are sick. The measure for the division of household work and for childcare tasks ranges from no involvement by the man (value 0) to doing an equal share (value 1) of family work. We regard the division of household work as balanced if the score on the index exceeds 0.75. We grouped the division of childcare work between the partners similarly. To control for the potential consent of both partners to the current division of household work and of childrearing tasks, we also consider whether the respondent is satisfied with the division of household work or childcare tasks. Even though being satisfied with the sharing of work may not correspond fully to a respondent’s perception whether the division of work is fair and just, we regard satisfaction as a proxy for gender equity, assuming that a respondent would not be satisfied with the division of work if she perceived it as truly unjust and unfair. Considering the actual sharing of work as well as the satisfaction with it allows us to look at the impact on fertility intentions of gender (in)equality in the division of work on the one hand and of gender equity, that is, the perception of this division as fair and just on the other hand.

Since household work and childcare involve very different tasks, we present the results for each of them separately. Table 4 reveals the importance of distinguishing between the actual sharing of household work and the satisfaction with it. For the childless, neither the division of household work nor their satisfaction with it influences their childbearing intentions significantly. But among parents there are clear gender differences in the impact of sharing and of contentment with it. Mothers who get support in household work from their partner are more inclined to intend to have another child in the next 3 years than mothers who do not get support. For fathers, the actual division of household work does not matter. It is their satisfaction with it that affects their childbearing intentions: The more satisfied they are with the division of household work, the more they tend to consider having another child.

Table 4 Intentions to have a(nother) child in the next three years by gender division of housework and by gender division of housework and satisfaction with it
Full size table

This underlines the gendered impact of gender equality on childbearing intentions: For mothers, greater equality in the sharing of household tasks is associated with a higher inclination to have another child. For fathers, the actual division of work has no significant impact on their further childbearing intentions. Since women often reduce their employment when they become mothers and consequently do more household work than men do, we interpret the influence of fathers’ satisfaction with the division of household work on their childbearing intentions as a consequence of the “secondary gain”, which men have from becoming fathers.

A more balanced division of care work between the parents and satisfaction with it tend to encourage further childbearing intentions of mothers as well as of fathers (Table 5). Again, the results indicate that it is important to consider the actual gender division of care work as well as the satisfaction with it. The intensity and the significance of the results change once we control for parents’ satisfaction with the division of care work. One-child parents who are satisfied with the sharing of childcare are more inclined to consider having a second child in the next 3 years than those who are not satisfied. Surprisingly, for one-child fathers the actual sharing of childcare influences their childbearing intentions positively, while for one-child mothers there is no significant effect of the actual division of care work on their intentions to have a second child (once we control for their satisfaction with their childcare arrangement with their partner). Relief from care work, however, matters significantly for mothers of two children.

Table 5 Intentions to have a(nother) child in the next three years by gender division of childcare work and gender division of childcare work and satisfaction with it
Full size table

These results illustrate how complex the relationship is between the gender division of work in the family and childbearing intentions. To grasp the potential impact of the gender division of family work on fertility, we need to distinguish between the distribution of household work and of childcare work; we also need to account for the perceived fairness of the distribution—in our case measured through satisfaction; we need to look at women and men separately, and we need to distinguish by parity. The distinction between actual work and the perceived fairness/satisfaction of the division is particularly relevant, since, given the gendered pattern of family work, the division of work and the support received may mean something different for women and for men. In summary, to understand the impact of the gender division of work in the family on childbearing intentions, it is essential to apply a gender- and parity-specific approach and to distinguish between effective gender distribution (gender equality) and gender equity (perceived fairness) in family work.

5 Discussion and Conclusion: Which Equality Matters for Fertility Intentions?

We have taken the inconclusive results of previous research on the impact of employment, economic resources, and the division of family work on fertility intentions as our starting point to explore the relationships between gender equality and fertility intentions. We have argued that the assumptions about this relationship are often simplistically based on “sameness of distribution”, that is, on a uniform and uni-directional understanding of gender equality. We maintain that in order to capture the complexity of gender equality one needs concepts that allow for gender differences but expose gender inequality. Such an approach incorporates the notion of gender equity suggested by Fraser (1994) and McDonald (2000a, b), and connects it to recent debates on “which equalities matter” for gender equality (Phillips 1999, 2004, 2006). In our application of this approach we have proposed and tested three dimensions of gender equality and their impact on childbearing intentions: the capacity to form and maintain a household (range of employment), the capabilities and potentials for agency (economic resources), and gender equity in family work (gender division of household work and of care and the satisfaction with them).

First, we find that the capacity to maintain a household by means of one’s own full-time employment is essential for childless women’s and men’s intentions to have a child in the next 3 years. Once they have become parents, the positive effect of such employment on childbearing intentions disappears for women, while it remains positive for men. Second, difficulties to make ends meet tend to lower women’s and men’s childbearing intentions when they have one child. Third, a more gender-balanced division of household and of care work tends to support childbearing intentions of both women and men. But the results vary by type of work. We find indications that for women’s intentions their partners’ engagement in household work matters, while for men their satisfaction with the division of household work (be it equal or not) matters more than the actual sharing.

Overall, our results cannot be reconciled with any notion of a simple, uniform, and uni-directional relationship between gender equality and fertility intentions. They emphasize the need for a multi-dimensional approach as outlined in this paper. Without evaluating the impact of employment on childbearing intentions from the perspective of “having the capacity to maintain one’s household and family”, we could have regarded the shift in full-time employed women’s childbearing intentions after a first child simply as a matter of gender differences, for example, as differences in preferences between women and men or as a matter of individual choice. Viewed from the perspective of “maintaining one’s own household”, our results challenge an interpretation that these preferences or choices are based on equal opportunities. The results rather prompt the question why mothers still seem to be confronted with having to choose between either maintaining their employment (and thus their capacity to maintain their household) or opting for a(nother) child, while fathers do not. Within the gender-equity framework proposed by Fraser (1994) and by McDonald (2000a, b), such a “choice” is neither fair nor just. Within a framework which furthermore aims at exposing gender inequalities, these results lead us to look for the factors and circumstances that produce inequality in choices, preferences, and in the results—in our case, in childbearing intentions. This shifts the attention from individual-level gender differences, such as preferences or choices, to gender issues in the labor market and in society, and thus to contextual—and politically changeable—aspects of gender inequality.

As regards agency and the capability to choose, our results indicate some impact of economic aspects on childbearing intentions for both mothers and fathers. We can interpret their tendency to lower their intentions to have a second child in the next 3 years if they have difficulties to make ends meet as an expression of “agency poverty” (Korpi 2000), that is, as a limitation of their agency, which also affects their childbearing intentions. There are no gender differences as to the influence of (economic) agency poverty on childbearing intentions. “Making ends meet” also displays the weakest link to childbearing intentions for women and for men of all parities. The results contrast with the gender-differential impact which employment or non-employment have on these intentions. For research on the links between gender equality and fertility, this suggests that we need to distinguish between the economic capabilities or limitations incurred through one’s own employment situation, through the partner’s employment, and through the perceived joint financial situation of the household. We also need information on the causes of economic difficulties. Agency seems to be influenced by whether the economic difficulties have been brought about by one’s own decision (and thus one’s own agency) or by circumstances that one cannot (fully) control.

Turning to the influence of gender equality in household work and childcare tasks on childbearing intentions, we have shown that it is not sufficient to regard family work as one indistinguishable bulk of work. It seems that men’s engagement in household work and in care are judged differently with respect to childbearing intentions. We have also shown that it does not suffice to only look at the division of work and care among the partners. We also need to consider satisfaction with the division of household work and care, that is, to consider gender equity in each type of family work. The differential results of the actual sharing and of the satisfaction with it for women and for men and by parity underline the importance of distinguishing between gender differences and gender inequality. Only if gender differences are perceived as gender inequalities may we expect depressing effects on fertility.

Our explorations thus raise a number of issues regarding research on gender equality and fertility. First, we have illustrated the importance of conceptualizing gender equality in a way that recognizes gender equity and allows for the distinction between gender differences and gender inequality. We need concepts that grasp the manifestations of inequality but acknowledge the existence of gender differences. Only such an approach will open up space for detecting which aspects of gender equality matter for fertility decisions.

One may contest our choice of gender equality dimensions and their operationalization via employment, the perceived economic situation, and the division of household work and care. In particular, one may argue that these dimensions and their representations are interrelated. Having the possibility to form and maintain a household as well as to divide family work and care fairly and satisfactorily may be seen as part of having agency and the capabilities to choose. Likewise, employment and economic resources are to some extent mutually dependent, since employment contributes to one’s financial resources, while having sufficient economic resources may offer the opportunity to opt out of employment. Either of the two may in turn affect the division of household work and care. We are aware of the fact that our three dimensions of gender equality may not be sufficient to grasp the multi-dimensionality of gender equality; neither do we regard the operationalization as the last word in the matter. Our intention has been to challenge the currently prevalent conceptualization of gender equality in demographic research and to provide a simple example in order to demonstrate the need for more theoretical and empirical work. As for theoretical work, this should blend in with long-standing research in the social sciences to determine which dimensions are relevant for gender equality in a comparative perspective, at the individual and at the societal level (see, e.g., Fraser 1994, 2008; Phillips 1999, 2006; Robeyns 2003) and to understand how the various dimensions of gender equality are interrelated (Ferree 2010). This includes the need for more nuanced concepts of institutional contexts to understand how various dimensions of institutional contexts relate to various aspects of gender (in)equality and to fertility (see also Korpi et al. 2013). As for empirical work, there remains a need for comparative surveys and survey questions designed to capture such multi-dimensional conceptions of gender equality.

Second, we have demonstrated the need to investigate childbearing decisions of women, of men, and for each parity separately, since gender issues play out differently for each of them. To assess which gender issues play which role in which decision-making process, we need the differential perspectives of women and of men, of mothers, and of fathers. Our study illustrates the various ruptures in gender equality brought about by parenthood at different parities and the need to investigate those in more detail. Third, even though we explore the relationship between gender equality and short-term fertility intentions, our results show that there is no single-dimensional answer to the question of which equality matters for fertility. Compared to the general assumption in demography that the gap between gender equality in the employment sphere and gender inequality in the family sphere keeps fertility at low levels, our results reveal that the relationship between gender equality, employment, family work, and fertility is much more complex. There exist various concurrent gender inequality issues within employment (not least related to parenthood) as well as within the family. Our results highlight the need to consider the plurality of inequalities and to identify their substantive elements, not only in employment and in the family, but also in other gender-equality and fertility-relevant areas of life. To look for inequalities in resources, in capabilities, in agency, and in the perception of fairness provides a useful tool to locate the essential dimensions of inequality and to understand which gender (in)equalities matter for childbearing decisions.