Search

Background: The decline in marriage and increases in cohabiting unions during childbearing ages represent a major change in family structures and family dynamics. Yet no comprehensive description has covered period trends in the prevalence of marital and nonmarital unions or partnership entry and exit rates across educational levels. Objective: We describe period trends in the proportion of individuals in union and the flow of union formation and dissolution at ages 18-49 across educational levels for men and women in 1989-2019 in Finland. Methods: We use register data with full histories of coresidential unions regardless of marital status. We calculate the age-specific yearly prevalence of marriages, cohabitations, and all coresidential unions, and rates of union formation and separation. Results: The prevalence of marriage declined across educational groups. Increases in the prevalence of nonmarital cohabitation are slightly smaller than declines in marriage; as a result, the total union prevalence declined, especially among low-educated men and women. Union formation rates have declined since around 2010, and separation rates have increased notably, especially among the lower educated and in age groups below 35. Conclusions: The increase in nonmarital cohabitation has not fully covered the decline in the stock of marriages. The increases in union instability are notable. All observed trends and educational gradients are remarkably similar among men and women. Contribution: We provide the first comprehensive population-level estimates of trends and patterns of the union stock and the flow of marital and nonmarital unions by education for both men and women of childbearing ages.

Background: Divorce and separation have become common life-course events in many European countries. Previous studies show that separated individuals are likely to move from homeownership to renting and to experience a period of residential instability. However, little is known about postseparation homeownership levels in the long run. Objective: This paper investigates homeownership levels after union dissolution. We extend previous research by examining changes in homeownership levels after separation by time (since union dissolution) and across population subgroups. We study whether and how postseparation homeownership levels are associated with repartnering and gender. Methods: We use Finnish register data and logistic regression analysis. Results: Compared to partnered individuals, homeownership levels among recently separated individuals are low. With increasing time since union dissolution, homeownership levels increase. However, the levels are high only among repartnered persons and remain low among separated individuals who remain single. Homeownership levels are slightly lower among separated women than among separated men. Contribution: The study shows the short- and long-term effects of separation on individuals' housing careers. After separation many individuals move from homeownership to rental accommodation, and most previous renters continue to rent. Separated individuals who form new unions are likely to move or return to homeownership. By contrast, those who remain unpartnered following separation have a low likelihood of living in owner-occupied dwellings. This likely reflects both individuals' choices and their constrained opportunities in a country with high homeownership aspirations and levels.

Objective: This paper studies postseparation fertility behavior. The aim is to investigate whether, and if so how, separation affects second birth spacing in Western European countries. Methods: This analysis makes use of rich survey data from Belgium, France, Germany, Italy, Spain, and the United Kingdom, as well as from Finnish register data. We thus cover the behavior of a large proportion of the population of Western Europe. We also use descriptive measures, such as Kaplan‒Meier survival functions and cumulative incidence curves. In the multivariate analysis, we employ event history modeling to show how education relates to postseparation fertility behavior. Results: There are large differences in postseparation fertility behavior across European countries. For Spain and Italy, we find that only a negligibly small proportion of the population have a second child after separating from the other parent of the firstborn child. The countries with the highest proportion of second children with a new partner are the United Kingdom, Germany, and Finland. In all countries, separation after first birth leads to a sharp increase in the birth interval between first and second births. Contribution: Our study is a contribution to the demographic literature that aims at understanding birth spacing patterns in Western Europe. Furthermore, we draw attention to the role of postseparation policies in explaining country differences in fertility behavior in contemporary societies.

Background: Childlessness has increased in many European countries. Partnerships and parenthood are obviously closely related, but there is relatively little knowledge on how childlessness is linked to contemporary union dynamics that involve high rates of separation and unmarried cohabitation. Objective: To situate (biological) childlessness in longitudinal dynamics of union formation and stability, we take a life-course approach to union trajectories that consist of states entered via the formation and dissolution of cohabitations and marriages. Concretely, we identify groups of similar union trajectories of individuals between the ages of 18 and 39 who are childless at age 42. Methods: We analyse register data on Finnish men and women born in 1969 and 1970 (childless N=3,241) with sequence, cluster, and multinomial logistic regression methods. Results: Four clusters of typical union trajectories were identified among the childless and assigned these labels: 1) Never Partnered (45Š), characterized by never having entered a coresidential partnership, or just having entered a cohabitation near age 40; 2) Briefly Cohabited (25Š), characterized by mostly living single after a brief cohabitation spell; 3) Cohabitors, Often Serial (19Š), marked by typically discontinuous cohabitation; and 4) Married (11Š). The Never-Partnered cluster is male-dominated. Men with a rural background and less-educated men and women are overrepresented among the Never-Partnered childless. Conclusions: For the great majority of the childless in our study cohorts, union trajectories are marked by either the (almost) complete absence of coresidential unions or fragmentary cohabitation histories. Contribution: The study contributes to the literature by showing that union histories, including never partnering as well as cohabitation instability, are key for understanding contemporary childlessness. Comments: We recommend that you read the paper on screen or print a color version.

Background: Previous studies suggest that in some countries socioeconomic differences in family formation are highly gendered, whereas gender-neutral patterns are reported in other countries. Most previous studies focus on single events and therefore it is unclear how the gender differences and neutralities in family transitions combine into variation in longer family-formation trajectories. Objective: We explore how family trajectories vary by educational attainment and gender. The research asks whether there are gender differences in family trajectories by education. We focus on the trajectories of women and men in Finland between the ages of 18 and 39, and on the 1969 and 1970 birth cohorts. The trajectories consist of states entered via the formation and dissolution of cohabitation and marriage and the birth of the first child. Methods: We give a sequence representation of Finnish register data comprising monthly histories of union dynamics and childbearing. We focus on the number and order of family states. Results: We find notable differences in family trajectories by educational attainment; however, the gender differences in the trajectories within educational groups are negligible. For instance, the proportion of never-partnered and childless at age 39 is largest among those with low education, regardless of gender. Further, at age 39, highly educated women and men are most likely to live in the same union in which they became first-time parents. Contribution: This study adds to previous literature by showing that in an egalitarian Nordic welfare state, longitudinal family-formation trajectories are highly stratified by education but remarkably gender-neutral.

The high and increasing incidence of divorce, with the various consequences for adults and children, has aroused interest among social scientists in understanding the contributory factors. Prominent economic and psychosocial theories suggest that the husband’s social and economic resources tend to stabilize a marriage, whereas the wife’s economic success tends to destabilize it (the gendered hypothesis). Register-based follow-up data from Statistics Finland on first marriages in Finland that were intact at the end of 1990 and divorces in 199193 (n=21,309), and Poisson regression were used to analyze the impact of the socio-economic positions of the spouses on the risk of divorce. This thesis consists of three articles published in international refereed journals, and a summary article. The aim of sub-study I was to disentangle the influences of various aspects of the spouses’ socio-economic positions on divorce risk and to reveal the causal pathways through which each socio-economic factor was related to it. Sub-study II investigated the joint effects of both spouses’ socio-economic positions. Finally, sub-study III explored the possibility that the effect of spouses’ socio-economic positions on divorce risk might vary according to the duration of the marriage. When examined individually, divorce risk was inversely associated with socio-economic status for all its various indicators (i.e. each spouse’s education, occupational class, economic activity, and income, as well as housing tenure and housing density) except the wife’s income. All of these factors had an independent effect. The independent effect was weak for both spouses’ occupational rankings and housing density, however, and it was positive for the wife’s income. The divorce risk for couples with both partners at the lowest educational level was lower than expected on the basis of its overall inverse association with each spouse’s education. Employed and homemaker women with employed husbands had comparatively stable marriages, whereas couples in which the husband, the wife, or both partners were unemployed had an elevated risk of divorce. The husband’s high income decreased the risk, and the wife’s high income increased it regardless of the level of the other spouse’s income, but the divorce-promoting effect of the wife’s high income was especially strong when the husband’s income was low. The comparatively high divorce risks for spouses with little formal education and those in manual-worker occupations were found to be specific to marriages of relatively short duration. In contrast, factors such as unemployment, the wife’s high income, and living in a rented dwelling were found to increase the risk regardless of marital duration. Overall, the socio-economic resources of the spouses, irrespective of which spouse had contributed them, decreased the risk of divorce, supporting the gender-neutral hypothesis. However, some aspects of the wife’s resources (absolute and relative to those of her husband) tended to increase the risk. The finding that the less structural socio-economic factors affected divorce risk in a very similar way in marriages of varying duration highlights their importance as factors predicting marital stability.

Background: Despite the increasing prevalence of cohabitation, knowledge of how socio-economic homogamy affects the stability of cohabiting unions is scant. Few studies have compared the effects of homogamy in both ascribed and achieved socio-economic status on union dissolution. Objective: Our aim is to determine how homogamy and heterogamy in educational level and parental social class affect the risk of cohabitation dissolution in Finland. Methods: We use unique Finnish register data that includes information on non-marital cohabitation. Cox regression is used to analyse the risk of dissolution in 20,452 cohabitations. We examine the dissolution rates in all possible combinations of partner status, and analyse how these estimates deviate from the main effects of each partner's status. Results: According to the findings, homogamy in parental social class is of little consequence in cohabitation dissolution, although cohabitations between people from upper-white-collar and farmer families are disproportionately likely to dissolve. Educational differences between partners are more significant determinants of cohabitation stability: extreme heterogamy is associated with an increased separation risk, and homogamy decreases the separation risk among cohabitors with a higher university degree. Conclusions: In line with the perception that personal achievement is more significant than social origins in contemporary union dynamics, similarity in educational level increases cohabitation stability more than similarity in socio-economic origin. Although previous Nordic studies report little or no association between educational homogamy or heterogamy and marriage dissolution, our study shows that educational differences do matter in cohabiting unions.

This is the first report of a research project that focuses on the impact of socioeconomic factors on divorce risk in the context of other determinants. In this paper; divorce risk differentials are examined by two measures of family composition and various indicators of spouses' socioeconomic status. Divorce risk differentials are described also by two temporal variables; which are used as control variables in the other analyses. This is a register-based follow-up study, covering Finnish first marriages which were intact at the end of 1990 and judicial divorces between l 991 and l 993. A piecewise exponential hazards model is used. When the temporal factors were recontrolled for; divorce risk increased with increasing age of the youngest child, and divorce risk decreased with increasing numbers of children in the family in every age group of the youngest child. Also, when the temporal factors were held constant, socioeconomic status was inversely related to divorce risk, when socioeconomic status was measured by either of the spouses' education, occupational class or economic activity, husbands income or housing tenure.

BACKGROUND Previous research on fertility has focused on women, and less attention has been paid to men and couples. OBJECTIVE The aim of this study is to examine how the socioeconomic resources of cohabiting and married partners affect entry into parenthood in a relatively gender-egalitarian welfare society. METHODS The study is based on Finnish register data and uses event-history analysis to predict first births from both partners' socioeconomic characteristics. RESULTS The results show that each partner being employed (as opposed to studying) and having a higher income seems to encourage entry into parenthood. As compared to employed couples, either partner being currently unemployed or having recent spells of unemployment had very weak effects, whereas either partner being economically inactive seems to discourage childbearing. Although the resources of male partners also have an effect, the female partner's situation appears to be equally or even more influential. The effects of female partners' characteristics are almost as great when male characteristics are controlled as when they are not, and women's and men's characteristics do not interact with each other. Moreover, with regard to income and educational attainment beyond age 30, for example, the woman's resources have a stronger positive effect than the resources of the male partner. CONCLUSIONS Together with several previous studies from the Nordic countries, this study lends support to the idea that the influence of women's and men's economic resources on family formation are perhaps much more symmetrical than conventional theories suggest. COMMENTS The significance of women's own resources, net of the male partner's resources, suggests that previous studies have not overestimated their positive impact.

BACKGROUND Remarkably little is known about the significance of consensual unions for fertility. This is true everywhere, but the lacuna is more important in the Nordic countries where there is so much childbearing outside of marriage, mostly in consensual unions. The purpose of this paper is to help fill this hole in our knowledge for Finland. OBJECTIVE Unusually good register data enable us to study recent fertility trends by union status (married, cohabiting, neither) using records for some 112,000 Finnish women, or 11Š of all women at fertile ages. METHODS Our description of fertility is based on group-specific duration-based TFRs, which is the number of children borne by a woman who remains in the group throughout her reproductive life, as computed from the fertility rates for a synthetic cohort. This is an intuitively appealing metric that has been taken into systematic use only recently. RESULTS We find substantial fertility differences between women who cohabit, women who marry directly (i.e., without pre-marital cohabitation), and women who marry their cohabitational partner. As one would also expect in Finland, cohabiting women have much lower fertility than married women. The marital TFR is highest among the directly-married and declines monotonically as the length of pre-marital cohabitation increases, even when premarital childbearing is included in the count. As far as we know the latter relationship has not been shown before, because extensive data for complete cohabitational unions have not been available for other populations. CONCLUSIONS The Finnish data are unique, even among the Nordic countries, in that they contain individual-level life histories of family dynamics that cover consensual unions from their very start. Fertility analysis would benefit if data similar to the Finnish were to become available, because analyses that rely on civil status as an indicator of union status barely add anything to what we already know about today?s family dynamics.