Abstract
Introduction
More liberal attitudes toward lesbians and homosexuals in the general population have led to considerably more same-sex couples living together throughout the Western world during the past decades. 1 A growing number of same-sex couples also have children, either from former couple relationships, or by acquiring children together through adoption, assisted fertilization, or use of a surrogate mother. According to numbers from the United States Census Bureau, 15% of American same-sex couples have children in their households. 2 With more children growing up in families headed by a same-sex couple, scholars have become increasingly more interested in the stability of same-sex unions. Moreover, research suggests that in terms of cohabitations or marriages, these unions tend to be less stable than different-sex cohabiting and marital unions, and the gap in stability is particularly large for couples with children. 3 Data from the Scandinavian countries, which have had legislation for same-sex partnership or marriage since the 1990s, indicate considerably higher divorce rates in same-sex marriages than in different-sex marriages. 4 This applies to males and even more to females living in same-sex marriages. However, in a more recent study from Sweden, only lesbian marriages were more prone to divorce than heterosexual marriages. 5 Moreover, Raley and Sweeney note that in settings where same-sex partnerships were more recently legally-recognized through civil unions or marriage, same-sex couples in formalized relationships sometimes have higher levels of stability than different-sex married couples. 6 This may be due to the fact that early adopters of registered partnership formation are more likely to be long-term couples who have been waiting to formalize their relationships.7, 8
Such ambiguities regarding the stability of same-sex marriage point to the need for more research on these marriages. However, in a review of research on marital satisfaction and stability in the 2010s, Karney and Bradbury concluded that studies continue to focus on different-sex couples despite the widespread legalization of same-sex marriage during the decade. 9 Accordingly, when scholars have considered union stability among same-sex couples, few studies have included same-sex couples who are married. 10 This is most likely due to a lack of available data and generally small sample sizes of same-sex couples. 11 Also, Sassler and Lichter pointed out that there is a paucity of data on sexual minorities in general, particularly longitudinal data. 12 Thus, to get more accurate knowledge of divorce risk in same-sex marriages relative to different-sex marriages, large groups of marriages need to be included over a period of several years.
Background
As of 2022, 31 countries worldwide have legalized same-sex marriages, 13 but most of these legislations have been introduced quite recently; for example, in 11 of the 16 European countries since 2010. 14 With increasing numbers of countries formally recognizing same-sex marriages, an interesting research issue is whether the legal formalization of same-sex unions may have any impact on their stability. One possibility is that the legalization of same-sex marriage may strengthen commitment and stability in terms of long-term monogamous relationships, 15 similar to what marriage is claimed to do for different-sex relationships.16,17, 18 Research has begun to test the effects of marriage laws on dissolution rates among same-sex couples. 19 Overall, this body of work demonstrates that marriage is important to same-sex couple stability; the ability to legally marry, coupled with general social tolerance toward marriage, is associated with a stronger desire for long-term relationships and stronger monogamy beliefs among sexual and gender minority populations. 20 In keeping with this, it is well established that different-sex marriages are generally more stable than different-sex cohabitations,21, 22 and assumedly this is reflecting poorer outcomes in cohabiting unions. However, the differences are generally reduced or eliminated upon accounting for selection into either cohabitation or marriage, and the transition to marriage per se may not result in any improvements in relationship quality. 23 Regardless of how these findings are interpreted, studies of different-sex marriages may not be applicable to same-sex marriages for various reasons: marriage may reflect different motives or levels of commitment for same-sex couples compared to different-sex couples. Heterosexuals getting married are primarily reflecting acceptance of normative family values and the goal of living in an enduring couple union. For same-sex couples, however, politics of gay and lesbian rights may also be part of the decision to marry. 24 Moreover, being a sexual minority may also influence same-sex couples’ choice of formalizing the couple relationship as well as the stability of the union formation. 25 Finally, various demographic characteristics may differentiate between same-sex and different-sex couples; for instance, heterogeneity of age between the partners and the proportion living in rural areas, which may influence the decision to marry and the stability of the marriage. 26 , 27 Thus, there is a need for more research on same-sex unions, particularly related to the stability of same-sex marriage.
The gay rights movement in Norway started officially in 1950 with the establishment of a formal organization for homosexuals. 28 At that time, there was little attention given to homosexuality as a social issue, and most gays and lesbians were hiding or even suppressing their sexual desires, and almost none were living together as couples. Homosexual activity between males was also illegal according to the law, whereas lesbian sexuality was legally and socially unnoticed. The existence of a gay rights movement was therefore barely recognized by anyone other than homosexuals themselves. This lack of recognition from the wider community continued for the next 15 years, until 1965, when the Norwegian Broadcasting Corporation, the only public broadcaster in Norway at that time, launched a radio program of 80 min devoted solely to a variety of homosexual-related issues. In the aftermath of the radio program, public awareness of homosexuality gradually increased and made discrimination against homosexuals a continuously more current issue in the public and political discourse, and seven years later, in 1972, the penal code against homosexual activities was repealed. This was primarily a symbolic victory for the gay rights movement, since very few individuals had ever been convicted under this law. Still, it had immense impact on the fight for gay rights and in the subsequent years the Norwegian population gradually developed more accepting attitudes towards homosexuals, and several professional or public schemes and regulations related to gay rights were discontinued or put into practice. Among these were the repeal of homosexuality as a psychiatric diagnosis by the Norwegian Psychiatric Association in 1977, guarantee of full rule of law for homosexual teachers in the schools in 1978, full rights for homosexuals in the armed forces in 1979, a separate penal protection for homosexuals in 1981 and, perhaps the most important reform, the introduction of the Partnership Act in 1993. With this law, Norway was among the very first countries worldwide with an official legitimization of same-sex unions in terms of registered partnership, giving homosexual couples basically the same rights and duties as married couples. In 2009 the partnership act and the traditional marriage act were converted into a common marriage act for both heterosexual and homosexual couples.
This long history of same-sex legal unions makes data from the Norwegian official marriage register ideal for investigating the stability of same-sex marriages. Firstly, it makes it possible to compare the stability of formalized same-sex unions with different-sex marriages over a long period. Secondly, the marriage register also provides data on demographics that may be related to the formalization and stability of same-sex unions. Here, one characteristic is of particular interest; namely whether the couples are living in an urban or rural area. Several studies have demonstrated more homophobic attitudes and discrimination against gays and lesbians in rural areas than in urban areas.29,30,31, 32 However, the traditional negative rural attitudes and discrimination against homosexuals may have declined along with the general trend of more liberal attitudes toward homosexuality during the past couple of decades.33, 34 For instance, Anderson et al. found widespread support for same-sex marriage in a largely rural state in the upper Midwest without significant differences based on rural residence, suggesting that the presumption of rural bias and hostility toward alternative sexualities has been mitigated. 35
Most studies using a rural/urban perspective of same-sex marriages are from the USA and are based on relatively small samples, focusing mainly on different attitudes toward, and discriminations against, gays and lesbians or on more general LHBT rights, particularly the right to marry. Thus, little is known of any rural/urban differences in other parts of the world or the extent to which same-sex marriage stability differs between rural and urban areas.
The present study aims at providing this kind of information from a Norwegian setting. More specifically, it addresses four different research issues related to same-sex marriage:
What are the main demographic characteristics of same-sex male and female marriages compared to different-sex marriages? To what extent are female and male same-sex marriages more or less prone to divorce compared to different-sex marriages? To what extent has the stability of formalized same-sex unions changed during the past 25 years? To what extent are same-sex relationships more or less stable in rural areas compared to urban areas?
Method
Data
The dataset consists of all male and female same-sex marriages formed from 1993 until 2018 in Norway (N = 5187). The data file is available for download 36 and described in detail. 37 All marriages were investigated for each successive year to see if the pair was still married. They were identified by a unique identification number in the database of Statistics Norway, a method that made it possible to detect divorces even of persons who divorced and remarried in the same year, something that would not be detected if only the individual persons were controlled for being married each successive year. When investigating divorces, some argue for using reports of separation instead of divorce. 38 However, separations are reversible and sometimes not reported in the Norwegian database. Using divorce thus provides more accurate and reliable data on the stability of same-sex marriages. Moreover, the data were attained through a special order from Statistics Norway and are not accessible in any data published on their website (www.ssb.no). Before the data were delivered, marriages formed within a calendar year were aggregated, making it possible to perform analyzes at a cohort level. For more details, see the supporting material. The size of the population of same-sex marriages varies significantly from one year to the next.
This study is linked to a previous study on different-sex marriages and compares the same-sex marriages in the present study to the data from that study. 39 The connected study investigates differences in divorce rate among urban and rural marriages for more cohorts than those included in this study, but does not have any information for same-sex marriages. The definition of rural and urban marriages is the same in both studies (see Supplemental Material for details).
Even if the data includes both same-sex registered partnerships (1993–2008) and same-sex marriages (2009–2018), we chose to use the term
Analysis
Analyzes were conducted in R,
40
and the EpiR package was used for relative risk analysis.
41
Effect sizes were calculated using Hedges’ g that corrects for different sample sizes followed with 95% confidence intervals as described by Hedges and Olkin using the online tool of Lenhard and Lenhard.42,
43
The Hedges’ g was calculated using the formula
The term
Results
The analyzes were performed separately for male and female same-sex marriages, respectively, in the total population and rural and urban areas. In all analyzes, the data of same-sex marriages were compared with data from all different-sex marriages in Norway formed in the corresponding years.
Population
Total Sample
From 1993 until 2018, there were 524,564 marriages in Norway, and 0.99% of these (5,187) were same-sex marriages. The lowest percentage was in 1997 with 0.46%, and the highest was in 2018, the last year included in this study, with 1.89%.
In 1993 there were 111 male same-sex and 43 female same-sex marriages. The latter steadily increased in the following years, with a transient dip in 2008, the year before same-sex marriage was legislated. The all-time high was reached in 2017 with 213 marriages. On the other hand, male same-sex marriages decreased after the first year, and then it took more than 20 years before these marriages passed the number from 1993. However, similarly to female same-sex marriages, male same-sex marriages also revealed a dip in 2008. As illustrated in panel a of Figure 1, there were more male same-sex marriages than female same-sex marriages in the first seven years. From 2000 onward, female marriages outnumbered male marriages. Thus, the increase of female same-sex marriages has been much steeper since 1993 than the increase in male same-sex marriages.
Population of same-sex male and female marriages—total, urban, and rural sample.
Urban and Rural Sample
In urban areas, the portion of same-sex marriages has increased from 1.9% to 3.1% of all marriages. As illustrated in panel b of Figure 1, there was a drop in male and female same-sex marriages in 2008, the year before same-sex marriage was legislated, followed by an incline in 2009. In rural areas, same-sex marriages have increased from 0.3% to 1.2% of all marriages. A steeper increase of female same-sex marriages was observed compared to male same-sex marriages, as illustrated in panel c of Figure 1.
Age at Marriage in Same-sex Marriages
Total Sample
Mean age at marriage was 43.1 for the oldest partner and 35.6 years for the youngest partner among males, with an average age difference of 7.5 years. For females, the mean age for the oldest partner was 38.7 and 34.1 years for the youngest, with an average age difference of 4.6 years. In comparison, the mean age at marriage was 35.8 and 32.9 years, respectively, in different-sex couples, with an average age difference of 2.9 years. Mean age at marriage has increased from 1993 to 2018 in the different-sex marriages, from 31.9 years to 38.6 years for the male partner who most often is oldest, while there has been a declining age trend in female couples (from 43.6 years to 37.8 years for the oldest partner) and male couples (from 44.2 years till 42.9 years for the oldest partner).
Urban Sample
The mean age at marriage for urban male couples was 43.7 years for the oldest partner and 36.3 years for the youngest partner, with an average age difference of 7.3 years. For female couples, it was 38.6 years for the oldest partner and 34.2 years, with an average age difference of 4.4 years. The mean age at marriage among different-sex couples was 35.1 years and 32.4 years, respectively, with an average age difference of 2.7 years.
Different-sex couples have had an increase in age at marriage from 1993 until 2018 (from 32.4 to 37.4 for males), while same-sex couples have had a declining trend in terms of 43.5 years to 37.8 years for the oldest person in female couples, and 44.6 years to 43.8 years for the oldest partner in male couples. In 2018, age for urban female same-sex marriages did not differ significantly from age in different-sex marriages (effect size Hedges g = 0.035, CI = −0.168–0.238); however, male same-sex marriages did (effect size Hedges g = 0.609, CI = 0.392–0.826).
Rural Sample
The mean age at marriage for rural male couples was 42.1 years for the oldest partner and 33.9 years for the youngest partner, respectively, with an average age difference of 8.2 years. Corresponding figures for female couples were 38.3 years and 33.8 years, respectively, and 4.6 years. The mean age at marriage for different-sex couples was 36.1 years for the oldest partner and 33.1 years for the youngest partner, with an average age difference of 3.0 years. Notable, there was a larger age difference in rural compared to urban areas, especially among male marriages.
Age at marriage for same- and different-sex couples showed the same trend in rural areas as in urban. However, in rural areas, there was no significant difference in age between different-sex and same-sex couples in 2018 (for females: Hedges g = 0.013, CI = −0.405–0.146; for males: Hedges g = 0.128, CI = −0.209–0.465).
Divorce Rate
Female same-sex marriages had the highest total divorce rate throughout the period, followed by male same-sex marriages. The highest total divorce rate was observed among female same-sex marriages formed in 2003, with 59.1% divorced before 2018. See Table 1 for further details and the Supplemental Material for figures illustrating the successive divorce rate for each cohort. Of the 43 female same-sex marriages formed in 1993, 15 marriages were ended by divorce during the 25-year follow-up period, indicating a total divorce rate of 34.9%. Similar figures for male same-sex marriages were 23.4% (26/111) and 35.6% for different-sex marriages (6,402/17,977). Figure 2 illustrates the total divorce rates for each of the cohorts in the total sample and the urban and rural samples.
Total divorce rate for each cohort—total, urban, and rural sample.
Relative risk for same-sex female and male marriages at 5–25 years.
Relative Risk
We assessed the relative risk at five-, 10-, 15-, 20-, and 25-year follow-ups. All cohorts reaching those follow-up points were included. Thus, for the five-year follow-up, the cohorts 1993–2013 were included. For the 10-year follow-up, the 1993–2008 cohorts were included, and for the 15-year follow-up, the 1993–2003 cohorts were included. Due to small sample size when splitting between urban and rural areas at the 20- and 25-year follow-ups, we assessed only the total sample at these time points.
The risk of divorce was significantly higher for female same-sex marriages compared to different-sex marriages. The relative risk decreased along with a longer follow-up period. For male same-sex marriages, there was no significantly higher risk of divorce compared to different-sex marriages at any time of follow-up. Results are presented in Table 1.
In 2009, the arrangement of registered partnership was converted into same-sex marriage. Thus, we wanted to assess whether the legislation impacted the divorce risk. Same-sex registered partnerships (“marriages” formed from 1993 until 2008) were thus aggregated and assessed in comparison to same-sex marriages (marriages formed from 2009 and on). Since same-sex marriages were legislated in 2009, the data did not reach the 10-year follow-up. Therefore, the risk difference between those who formed a registered partnership and those who formed a same-sex marriage was assessed at only a five-year follow-up. Due to the sample size, only the total sample was assessed. Female same-sex registered partnerships had a relative risk of 1.06 (95% CI = 0.86–1.31) compared to same-sex marriages, whereas male same-sex registered partnerships had a relative risk of 1.10 (95% CI = 0.76–1.60) compared to same-sex male marriages.
Trends in Divorce Rate
The relative risk analyzes presented in Table 1 are based on combined data from all available cohorts. To investigate trends in the data, we assessed the successive divorce rate for all cohorts at five-year follow-up for the total and urban and rural samples. Assessing at five-year follow-up provided data for all cohorts from 1993 until 2013.
We identified a declining divorce trend at five-year follow-up in the total and urban samples of same-sex marriage and in different-sex marriage (see panel a and b of Figure 3), whereas, in rural areas, same-sex marriages increased throughout the period (panel c in Figure 3).
Divorce rate at five-year follow-up for the total urban and rural sample.
Discussion
For the total and urban samples, the population of male same-sex marriages was considerably higher the first year compared to the following 22 years. This indicates that many male same-sex couples had been waiting to formalize their couple relationship and used the opportunity as soon as it was possible. Female same-sex couples did not show the same pattern. Instead, the number of marriages increased steadily during the whole period. In the first years, there were more male than female same-sex marriages. However, similar to what has been observed in Sweden, 45 female same-sex marriages have had a steeper incline and are now the most common type of same-sex marriage in Norway. By 1997, same-sex marriages amounted to almost 0.5% of all marriages and further increased to almost 2% in 2018.
The relatively small number of same-sex marriages in 2008 compared to the following year was probably because of the legislation which was introduced in 2009. Thus, many of the same-sex couples who were planning to formalize their relationship may have waited until that year to get married instead of starting a registered partnership. However, the new legislation of same-sex marriages may not seem to have had any impact on the numbers of same-sex marriages in the subsequent years, similar to what Kolk and Anderson reported from Sweden. 46 Thus, the same-sex legislation may in itself have had little enduring effect on individuals’ decisions to formalize their same-sex couple unions. Moreover, neither did the legislation have any effect on the divorce rate. This may also correspond well with findings from studies of relationship quality when cohabitating couples decide to marry, in which there are few, if any, improvements in relationship quality. 47
Moreover, the percentage of same-sex marriages throughout the period has been higher in urban areas than rural areas. This may indicate that same-sex marriages are still less accepted in rural areas, even 25 years after the formalization of same-sex unions was legalized in Norway. Thus, even though there has been a general trend of more liberal attitudes toward homosexuality during the last couple of decades,48, 49 our findings suggest that this trend is not equally as strong in rural areas as urban areas. Alternatively, the findings may not primarily reflect a difference in the attitudes toward homosexuality across urban and rural areas, but rather any tendency of people in rural areas to be more favorable toward traditional family values in general, compared to people in urban areas. 50
Regarding age at marriage, it is well known that same-sex couples are older and more heterogeneous than their different-sex counterparts when they marry, 51 which was confirmed in the present study. We also observed that the age at marriage has declined for same-sex couples, while it has inclined for different-sex couples during the last 26 years. Thus, there has been a converging trend between different-sex and same-sex couples regarding age at marriage. Perhaps this also reflects more liberal attitudes toward homosexuality and same-sex marriage, empowering increasingly more young gay and lesbian couples to “come out of the closet” and get married. However, men who enter same-sex marriages are generally older than women entering same-sex marriages, and male couples are also more age heterogeneous. Moreover, the age at marriage was highest for same-sex couples in urban areas, while the age difference was highest among same-sex couples in rural areas. Such diversities between urban and rural same-sex marriages have not, to our knowledge, been identified before. These findings suggest that both the partners’ sex and age and the urban-rural dimension should be taken into consideration in studies of same-sex marriage.
Although there was wide variability in divorce rates across the different marriage cohorts, the divorce rate was consistently higher for female same-sex marriages than male same-sex marriages. When combining the marriage cohorts, female same-sex marriages, but not male same-sex marriages, had significantly higher divorce risk than different-sex marriages, similar to what has been reported from Norway and Sweden in previous studies in Scandinavia.52, 53 This was apparent after five, 10, 15, and 20 years of marriage. At 25 years, the divorce risk did not differ between female same-sex marriages and different-sex marriages, whereas male same-sex marriages had significantly lower divorce risk than different-sex marriages. However, only the initial 1993 cohort was followed for 25 years, and this cohort may differ from subsequent marriage cohorts since some of the marriages formed during that year may consist of couples who had waited a long time for the possibility to formalize their relationship in a registered partnership. These couples may have been more committed to their relationship and less prone to divorce compared to same-sex couples in general.
Still, both male and female same-sex marriages in the total sample and in the urban sample have had a steadily declining divorce rate since 1993, in parallel with different-sex marriages at five-year follow-up. A possible explanation could be that couples who decided to marry (or form a registered partnership) in the early years after same-sex union legislation did so partly as a political statement, whereas this may have been less relevant for couples forming more recent same-sex marriages. 54 Thus, the earlier couples may not have been committed to the marriage in the same way as more recent couples. Alternatively, the latter couples may have experienced more societal support for their marriage and less homophobia, disapproval, and hostility. Both these factors may have contributed to the declining divorce risk. The trend toward declining divorce risk in same-sex marriages may also reflect sexual minorities’ tendency to adapt to normative heterosexual romantic relationship values and ideology during the past few decades. 55 However, it is notable that the declining divorce risk coincided with the trend toward an increasingly younger age at marriage among same-sex couples. This may seem to contrast with studies indicating that young homosexuals are less likely than heterosexual young adults to value faithfulness and lifelong commitment. 56 In contrast, a recent study from Norway assessing divorce trends in general indicated that marriage age has a relatively modest effect on divorce risk, particularly 10 years or more into the marriage. 57
For same-sex marriages in rural areas, an opposite trend of increasing divorce risk was observed. This may reflect a more general variability in divorce risk across different married groups. For instance, different-sex couples 60 years or older have had an increasing divorce risk, described as the grey divorce revolution, during the past few decades even though the divorce risk has declined in the general population. 58 Why a similar trend is apparent also for same-sex couples living in rural areas of Norway, and particularly for female same-sex couples, is open to question. However, it supports the previously mentioned notion that the more liberal attitudes toward lesbians and homosexuals in the general population may not have pertained to rural areas to the same extent as urban areas. Therefore, married same-sex couples living in rural areas may have experienced an increasing divide between their expectations of societal acceptance and their actual experiences of any homophobic attitudes and behaviors, inflicting relational strains upon them, which may explain the inclining divorce risk in rural same-sex marriages.
The main strength of this study was the data from a relatively large number of same-sex marriages throughout 26 years, probably longer than any other studies in this field. Moreover, in previous research, very few studies have made separate analyzes for male and female same-sex marriages, respectively, or such marriages in an urban and rural area. No studies have, to our knowledge, combined same-sex marriages for males and females with an urban-rural dimension. In the present study, however, these characteristics were analyzed both separately and in combination, thus adding valuable insight into the complexity of same-sex marriages.
The main shortcoming was the lack of information on qualitative aspects of same-sex marriages, such as the level of marital challenges, relationship satisfaction, any experiences with couples therapy, and the reasons for divorce among those who split up. This kind of information would have provided even more insight into these marriages. The same applies to various social characteristics of the partners, such as education level, occupation, income, and religious affiliation. Since these data are not available at an individual level from any official register in Norway, the data would need to be collected separately in a survey among individuals living in same-sex marriage. Moreover, the lack of information on societal attitudes toward homosexuality and same-sex marriage was also a limitation of this study. Any such information would be of particular interest in explaining the differences in divorce risk between urban and rural areas. These data would also need to be collected separately and eventually matched with the official marriage and divorce register data. Thus, to gain more insight into the complexity of same-sex marriage, future research would need to carry out relevant surveys among married same-sex couples and preferably also in the general population, and eventually combine these data with similar register data used in the present study.
Supplemental Material
sj-docx-1-jfh-10.1177_03631990221122966 - Supplemental material for Same-sex Marriage Over 26 Years: Marriage and Divorce Trends in Rural and Urban Norway
Supplemental material, sj-docx-1-jfh-10.1177_03631990221122966 for Same-sex Marriage Over 26 Years: Marriage and Divorce Trends in Rural and Urban Norway by Rune Zahl-Olsen and Frode Thuen in Journal of Family History
Footnotes
Declaration of Conflicting Interests
Funding
ORCID iD
Supplemental Material
Author Biographies
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
