Gray Divorce in Sweden: Divorce for Swedes Over 60 Is Rising

When considering what increases divorce rates, it is most natural to think about characteristics of individuals or couples. For instance, alcoholics are more likely to divorce and so are poorer couples. However, even if we were to add dozens and dozens of these kinds of biographical details, we would still fall short of understanding why societal level divorce rates change over time. It isn’t that biographies don’t matter, but rather that the intersection of biography and history tells us so much more. In Linda Kridahl and Martin Kolk’s recent report describing why divorce rates have increased at older ages (60+) in Sweden,¹ they take us through a story that includes individual risk factors and why these factors are more common among older married Swedes today than they were in the past.

Those who weren’t trained in age, period, and cohort models in their demography graduate programs often need a bit of practice before they can think about a group of people moving through time together (a cohort) being shaped by social conditions and historical events. That’s why it’s helpful to look first at why divorce rates have decreased markedly for those married at younger ages in the United States before turning to why divorce rates for Swedes over age 60 are going up instead of down.

If the decline in divorce rates among younger adults seems puzzling, it is likely because an individualized perspective could predict the opposite. That is, most marriages occurring in the U.S. today are entered via cohabitation, and people who marry without cohabiting first have lower divorce rates than those who cohabit first. If the majority of people entering marriage do so with a history that elevates the risk of divorce, how can divorce rates be going down? The answer relies on understanding that as marriage rates have fallen for younger adults, the ones who still marry are no longer typical. The majority of U.S. men don’t marry until after they turn 30, and women marry only a couple of years younger. Back when most people were marrying in their early twenties, those who did so were (by definition) typical—their marriage age wasn’t related to peculiar characteristics.

Today, people who marry younger than age 25 are peculiar. They live in a society that offers cohabitation as an alternative to marriage, that encourages young people to get established before starting a family, and that emphasizes the importance of waiting until your own brain is fully developed (about 25) before choosing a lifetime partner. Yet these couples choose to marry anyway. Before the 1960s when cohabitation was avant-garde phenomena, cohabitants had a peculiar disregard for the social norm of marriage. Today, young marrieds have a peculiar regard for the institution of marriage. Most of those who are married young are at lower risk of divorce for this reason. It then makes sense that the group as a whole has lower divorce rates.

The divorce rate of a cohort doesn’t depend only on the events happening in their generation; it also depends upon which people in that cohort are married. It isn’t possible to get divorced without marrying first. 

Young marrieds in the United States have demonstrated an atypical valuing of marriage. What are married 60-pluses in Sweden like? Kridahl and Kolk are able to tell us the story from very rich data—Sweden maintains a central database of all individuals residing in the country, a population register that includes dates of birth, migration, and death, as well as entry and exit from marriages.² That means that when researchers investigated divorce among those 60 and older, they knew how long the couple had been married and whether either or both partners had been previously married.

Just as young marrieds in the U.S. as a group are different today than they were in the 1950s, older marrieds as a group are becoming more different over time in Sweden.

This information allows us to imagine each successive cohort as being comprised of different kinds of married people. Just as young marrieds in the U.S. as a group are different today than they were in the 1950s, older marrieds as a group are becoming more different over time in Sweden. For example, Kridahl and Kolk show older married Swedes are less likely to be in long duration unions and more likely to have been married before. Both of these characteristics are risk factors at the individual level: Divorce risk declines as marital duration increases, and first unions are more stable than subsequent ones. 

When described like this, we don’t have to try to imagine what would provoke an elderly Swedish couple who reached their Golden Anniversary to split thereafter. Instead, we need to imagine differences between the group of Swedes born in the early 1940s and those born in the late 1960s: the earlier cohort had a bigger share in their first, long-term marriages, while the later cohort had more remarried individuals. The average age at remarriage was about 50, and most of the divorces were to those in their 60s (not their 70s or 80s).

Those born in the early 1940s averaged having about decade of marriage under their belt before 1974 when the introduction of no-fault divorce caused rates to spike in Sweden. In contrast, most born in the late 1960s weren’t even married before divorce became relatively commonplace. These are different groups of people. Divorce rates have gone up among older Swedes because more of those who are married are in riskier marriages. The pool of people eligible for remarriage disproportionately excludes people whose religion, income, family history (or anything else) contributes to the stability of their first marriages. It also disproportionately includes people who have agreed to end previous unions along with a smattering of widows, widowers, and exes, who would have stayed married if their partner had cooperated. In other words, both selection and experience contribute to higher order marriages being riskier.

If I thought at the individual level alone, I might wonder why older Swedes don’t stick together the way my parents have. My parents are at virtually no risk of divorce partly because neither one has the cognitive capacity necessary to follow through on filing one even if they wanted to, and both are very attached to the house they have lived in together since 1968. But just three houses down lives Fred, who divorced his wife after their five children were grown and still lives in the same house with a new wife. They will likely stay together, too; divorce rates at older ages in the U.S. are up, but they aren’t skyrocketing.³ However, imagine the difference in the grey divorce rate on a block where 11 of the 12 couples have a history like my parents’ versus a block where most couples have a history like Fred’s.

We don’t have register data in the United States to replicate Kridahl and Kolk’s study here, but the data we do have lead me to believe that divorce at older ages will keep increasing here for some of the same reasons it has in Sweden.

Laurie DeRose is a senior fellow at the Institute for Family Studies, Assistant Professor of Sociology at the Catholic University of America, and Director of Research for the World Family Map Project.

1. “Grey divorce” typically refers to people aged 50+; Kridahl and Kolk investigate divorce trends at even older ages.

2. Swedish register data now also capture the beginning and ending of cohabitations, but Kridahl and Kolk observed cohorts born 1940-60, i.e., those aged 60+ in 2000-02. Cohabitations were not registered when these cohorts were entering adulthood.

3. Older divorce rates are skyrocketing in percentage terms, but percentage growth can convey a misleading picture for rates that start low. For example, divorce after age 65+ occurred to 0.18% of Americans in 1990 as was up over 300% when it reached its 2019 level of 0.54%, but that’s still just over 5 per 1000.