## Do 50 percent of marriages really end in divorce?

I’m sure you’ve come across or heard this statistic: 50 percent of marriages end in divorce. Why might people believe this?

One way that people may draw this conclusion is by doing some quick and rough math using marriage and divorce data from the CDC. The data show that around two million couples were married each year. The number of divorces and annulments was 944,000 in the year 2000 and 746,971 in 2019. Based on these data alone, the ratio is not quite 1:2 especially as divorce rates have been decreasing in recent years.

There are a few reasons why this kind of math is inappropriate. One of the more obvious issues is that more states reported marriage data than divorce data. The data on marriage include nearly every state except for three of the years while the divorce data exclude anywhere between three and six states over the entire duration shown. California, the state with the largest resident population in the US, is one of the states that was consistently excluded all 19 years of the divorce data. This would lead to an undercount of divorces that took place each year.

Even if we assume that data on divorces from the excluded states bump us up to the 50 percent mark, there are other assumptions that are worth unpacking. For example, does this mean we’re assuming that half of all couples that got married in a given year also got a divorce? The data do not support this. Less than 2% of marriages end within 12 months. Nearly 90 percent of first marriages are still intact five years later. The median time from first marriage to separation is 8 years.

Also, when people cite this statistic, are they saying this about all marriages regardless of whether it’s someone’s first, second, third, and so on? The research doesn’t support this either. First marriages are less likely to end in divorce than subsequent marriages. The reasons for this are complicated because of a range of factors including the presence of stepchildren and attitudes about divorce.

## So what is the probability of divorce?

To estimate the probability of divorce, we need to follow the same married couples over the duration of their entire lives or until the end of their relationship. This is also known as a longitudinal study. In longitudinal studies of first marriages, there is some support for the notion that 50 percent of marriages end in divorce but it’s limited to a subset of marriages that took place during 1970s. For example, one study found that 48 percent of couples that married in the 1970s got a divorce within the next 25 years. The same study found that divorce rates peaked in 1981. Since then, divorce rates have been declining, especially among adults under age 50. The divorce rate has been increasing among adults aged 50 and older, which has been referred to as gray divorce. Roughly one in four divorces in 2010 were among older adults, compared with 1 in 10 in 1990. Notably, gray divorce rates have flattened in recent years. This means that the high probability of divorce was largely concentrated among people born between 1950 and 1964, commonly referred to as Baby Boomers.

Current trends indicate that divorce rates will continue to fall especially as fewer couples get married and wait longer to do so. In other words, the average couple that gets married for the first time today has a higher chance of staying together than their parents. How much more likely? Researchers can’t say for certain, especially since the type of couples who are getting married today are different from the type of couples that married in the past. For example, marriages have become more selective of couples with certain attributes like higher levels of education and less student loan debt. One way to interpret this trend is that social expectations around marriage have made the practice more exclusive to those who can afford it. Greater access to economic resources also means that the relationship is less likely to chronically suffer from the financial hardships–and the stresses that come with it–that affect couples with less resources. Additionally, in some ways, marriage has become a symbol of accomplishment whereas in the past it was considered a necessary precursor for purchasing a home and becoming a parent. Nowadays, more and more couples view marriage as a “nice-to-have” but it’s no longer requirement for family life. This means that the type of couples who get married may also hold particular values and preferences that make them less likely to divorce than couples who view divorce more positively.

With all these nuances and caveats in mind, a review of the most recent data suggest that, on average, as many as two in five couples (40%) will divorce but it could be as low as one in three (33%). Remember that some of the best data are based on couples who were married long ago in the past. It may be not be accurate to assume that the circumstances that lead to divorces in the past are relevant to couples today. The same goes for predictors of long term happiness among couples who never separated. One thing that researcher consistently find is that economic stressors can severely undermine a successful marriage. While most couples cannot immediately improve their economic circumstances, research shows that couples can improve their relationship by modifying certain emotional responses (including lack of response) that trigger or prolong conflicts. Of course, learning to problem-solve and communicate differently under constant emotional duress is no easy feat for any couple or individual. This takes time and effort, and with a little luck, it’s possible that any married couple reach their happily ever after.