Objective: This study investigated the association between individuals' economic situation and divorce among the population aged 60+ in Sweden, with a focus on the role of gender and potential changes across cohorts.

Background: Previous research on divorce has mainly considered individuals of working age or all ages combined, although late-life divorce is increasing in several Western countries. Economic considerations regarding divorce may differ for older members of the population, who often have a more restricted economic situation and fewer possibilities to respond to the consequences of a dissolution.

Method: Using Swedish population registers, this study investigated late-life divorce among cohorts born 1930–1956. Discrete-time event-history analysis was employed to study the relationship between income (recent and accumulated individual income, and spouses combined income levels) and divorce across gender and cohorts.

Results: For women, the results showed a shifting pattern from a positive to a slightly negative gradient of the two individual income measures for divorce. Men had an increasingly negative income gradient in divorce across cohorts. The results for combined income levels for couples corroborate these patterns. Late-life divorce has become increasingly linked to low income over cohorts.

Conclusion: The novel findings for older individuals mirror previous findings on trends in the general population, although those studies used other socioeconomic measures. As the association between income and divorce becomes increasingly negative among older women and men, and as the divorce rate increases, there is a growing need to understand different aspects of couple dynamics in later life.

Developments over time in the prevalence of marriage and cohabitation formation has long received much interest, but less is known about more recent developments for different population subgroups in European countries. This applies as well to Sweden, a country considered a forerunner in family-demographic change. In contrast, much attention has been paid to the falling birth rates during the 2010s, and explanations that focus on the role of increasing uncertainties. In the Swedish case, the fertility decline has been documented across all main socio-demographic subgroups. The objective of this study is to examine whether the same situation holds for first marriage and cohabitation formation during the 2010s and the exceptional years of the Covid-19 pandemic. Based on Swedish population registers, including with new cohabitation data, we present annual indices of first marriage formation (1991–2022) and cohabitation formation (2012–2022) across a number of socio-demographic strata. We demonstrate a continuous decline in first marriage formation since the early 2010s with an additional sharp dip during the pandemic and a post-pandemic recovery. In contrast, there was a remarkable stability in cohabitation formation during 2012–2022. Although socio-demographic groups differ in their overall levels of marriage and cohabitation formation, the recent trends are strikingly similar across groups. Cohabiting couples, across population subgroups, have become less inclined to transition their union status to a more committed level, as manifested by marriage or parenthood. This occurred in spite of a positive economic climate in the 2010s and stable family policies, indicating that other forces are at play.

BACKGROUND

The downward fertility trend in Western countries during the 2010s is puzzling, not least in the Nordic region.

OBJECTIVE

In order to better understand its driving forces, we examine whether the decline is driven by differential behavior or compositional changes across sociodemographic population subgroups, for the empirical case of Sweden.

METHODS

Event-history techniques are applied to register data of the Swedish-born population to provide an in-depth analysis of the sociodemographic profile of the fertility decline.

RESULTS

The decline is confined to first births, with no apparent difference between individuals living in different types of municipalities or between those with fully Swedish and non-Swedish backgrounds. The first-birth decline is notable across labor market activity groups, but is somewhat more pronounced among those with weaker labor market positions. However, the shares of men and women who were active in the labor market and who had high earnings increased. The findings are strikingly similar for men and women.

CONCLUSIONS

For the most part the factors driving the Swedish fertility decline do not appear to be structural. Other forces, perhaps global, may underlie the general tendency to increasingly forego or postpone having children. The polarization in childbearing across labor market positions is an area for future research.

CONTRIBUTION

The study provides new insights into the conundrum of Nordic fertility decline during the 2010s.

Objective: This study examines the roles of commitment and gender equality for marriage formation in a highly individualized and gender-egalitarian context.

Background: Marriage is commonly perceived as a more committed form of union than cohabitation. Individualization perspectives propose that this discourages marriage formation, whereas gender perspectives suggest that this is only the case for couples living inegalitarian lives.

Method: The roles of marriage attitudes and gender equality for marriage formation are studied among 1085 cohabiting men and women born in Sweden in 1968–1980 using the 2003 wave of the Young Adult Panel Study, which is based on a stratified random sample. The authors examine how cohabitants perceive: (1) the level of commitment in cohabitation versus marriage and (2) the division of housework in their current relationship, and link these factors to population register data showing the individuals' likelihood of marrying in 2004–2007.

Results: Cohabitants are more likely to marry if they believe that marriage indicates relationship seriousness, but less likely if they see a marriage as more difficult to leave than cohabitation. Gender equality, measured as satisfaction with the division of housework, appears to be positively related to marriage formation. Whether perceiving marriage as a particularly committed form of union is more positive for marriage formation among gender-equal couples remains unclear, as the direction of the findings varies and statistical power is low.

Conclusion: Even in a highly individualized context, cohabitants seek certain forms of commitment through marriage. Gender equality plays a role, but needs further investigation.

The last half a century has witnessed swift changes in patterns of family formation and family dissolution in developed countries, followed by increased attention to these changes and to the causes and consequences of such change. In the wake of these developments, the field of family demography has developed in new directions. In the European context, Sweden and the other Nordic countries have frequently been referred to as forerunners in many aspects of family change, and these countries have gained much attention in family-demographic research. In the current contribution, we give an overview of the current state of Swedish family-demographic affairs.

This study investigates how the childbearing intentions of women and men in couples affect actual childbearing over the following years with the aim to explore whether women's or men's intentions may be more important. The study is set in Sweden, a country known for ranking high in terms of gender equality and a country with relatively high fertility. We use the Young Adult Panel Study (YAPS), which gives information about both partners' long-term childbearing intentions in 2009, and follow these couples for five years with register data on childbearing. In 30 percent of the couples, both partners intended to have a child, and out of these about three quarters have a child. The results show that, in general, both partners need to intend to have a child for the couple to do so but that women's intentions tend to have more influence over the decision to have a second or third child. This phenomenon is interpreted as decision-making in relation to the cost and utility of children for women and men.

Even though ideals in favor of gender equality in the private sphere are wide spread, discrepancies between ideals and actual behavior are common. Such discrepancies and potential dissatisfaction with gender unequal behavior within a couple are expected to influence partnership dynamics negatively. This study examines how discrepancies between the perceived ideal sharing of parental leave and the actual division of leave, as well as satisfaction with the division are associated with (a) relationship satisfaction, (b) continued childbearing, and (c) union dissolution, using Swedish panel data. The findings cannot confirm an effect of discrepancies on partnership dynamics. However, men who wish they had used a larger share of the parental leave have lower relationship satisfaction, lower continued childbearing, and higher probability of union dissolution. Women are seemingly not affected by their (dis)satisfaction with the division. The findings may reflect a changing father role related to the policy setting and norms in Sweden.

Sweden stands out as a forerunner in the development of gender equality and family dynamics. To deepen the knowledge on power distribution and gender dynamics of couple relations, we investigate how women and men's childbearing intentions influence actual childbearing behavior. The Young Adult Panel Study (YAPS) has information on both partners' childbearing intentions in 2009, which we follow for five years with register data on childbearing. The results indicate that women's childbearing intentions are more important than men's intentions in determining actual childbirths.

This thesis contains four empirical studies that examine patterns in family formation in Sweden around the turn of the new millennium. This is an interesting context for the topic because Sweden is often seen as a forerunner in family-demographic change as well as in gender equality and value developments. The empirical studies provide insight into the state of affairs of these developments. In all four studies, event-history analyses are applied to Swedish population register data. Two of the studies focus on first-marriage formation, while the other two focus on first childbearing. Study I reveals that starting from 1998 there was an increase in marriage-formation rates, which was not due to compositional changes in the population. This is in contrast to the common image of Sweden in the forefront of contemporary marriage decline. Study II shows that marriage rates for men and women display clear peaks at exact ages 30, 40, 50, and 60 as well as in the year 2000. The peaks demonstrate that in Sweden, marriage formation is not fully determined by conventional structural factors or individual characteristics. It illustrates that the choice to marry may be taken quite lightly in a context where differences between cohabitation and marriage are relatively small. Study III shows that the transition to parenthood varies across occupational groups. For both men and women, birth rates are positively related to own earnings. Birth rates are also high for those who work in a caring or teaching oriented occupation or in a gender-typical occupation. The findings may reflect differences in the possibility to take on parental responsibilities and they demonstrate the role of gendered behavior. Self-selection processes may also be important. In Study IV, the transition to parenthood for men and women is examined in relation to the sex composition of the workplace. Birth rates increase with the share of employees of the individual’s own sex, net of factors such as own earnings, public or private sector employment, occupation, and industry. This shows the importance of gender and gender structures in the work environment for childbearing behavior. 

Denna avhandling består av ett introduktionskapitel och fyra empiriska studier som undersöker mönster i familjebildning i Sverige kring millennieskiftet. Detta är en intressant kontext för ämnet eftersom Sverige ofta ses som en föregångare inom familjedemografisk utveckling och jämställdhets- och värderingsförändringar. I samtliga studier tillämpas livsförloppsanalys på registerdata över befolkningen. De två första studierna fokuserar på förstagångsäktenskap och de andra två på barnafödande. Studie I visar att från 1998 skedde en uppgång av giftermålstalen, som inte berodde på sammansättningsförändringar i befolkningen. Detta står i kontrast till den typiska bilden av Sverige som föregångare i en allmän nedåtgående giftermålstrend. Studie II visar att giftermålstalen för män och kvinnor uppvisar tydliga toppar vid ”jämna” åldrar (30, 40, 50 och 60) samt under år 2000. Topparna åskådliggör att i Sverige formas giftermålsbeslut inte helt av konventionella strukturella faktorer eller individuella egenskaper. Detta kan spegla att i en kontext där skillnaderna mellan samboende och äktenskap är relativt små kan giftermålsbeslut fattas relativt lättvindigt. Studie III visar att benägenheten att skaffa ett första barn varierar mellan yrkesgrupper. För både män och kvinnor är födelsetalen positivt relaterade till inkomster och till arbete inom omsorg och utbildning. Detta återspeglar sannolikt skillnader i möjligheten att försörja en familj och att ta ansvar för omsorg av eventuella barn samt att män och kvinnor troligtvis väljer yrken som passar deras intressen. Benägenheten att skaffa barn är också större för de med könstypiska yrken, vilket indikerar att könsrollssocialisering spelar en central roll.  I studie IV undersöks sambandet mellan arbetsplatsens könssammansättning och benägenheten att skaffa ett första barn. För både män och kvinnor är benägenheten att skaffa barn högre ju större andel av kollegorna som tillhör ens eget kön. Detta visar att genus och könsstrukturer i arbetsmiljön kan påverka barnafödandebeslut.