Verified by 
Is it true that about half of all marriages end in divorce? That should be an easy question to answer. After all, divorce is a very clear event that leaves an official paper trail. But the answer continues to be controversial, to this day.
What are the current claims about the rate of divorce?
In 2010, Paul Amato published a review of research on divorce in the prestigious Journal of Marriage and Family. Here is the key takeaway:
“At the end of the 20th century, 43% to 46% of marriages were predicted to end in dissolution. Because a small percentage of marriages end in permanent separation rather than divorce, the common belief that about half of all marriages are voluntarily disrupted is a reasonable approximation.”
So Paul Amato’s answer was yes: It is true that half of all marriages end in divorce or a permanent separation.
In 2014, Claire Cain Miller, writing in the New York Times, said we had it all wrong:
“It is no longer true that the divorce rate is rising, or that half of all marriages end in divorce. It has not been for some time.”
Miller said that more people who married in the 1990s reached their 15th wedding anniversary than people who married in the 1970s or 1980s. She added, “If current trends continue, nearly two-thirds of marriages will never involve a divorce, according to data from Justin Wolfers, a University of Michigan economist.”
That projection, that the percentage of marriages ending in divorce in the future will be just a little over one-third, is just that – a projection, a guess. It also came with an important caveat:
“…the decline in divorce is concentrated among people with college degrees. For the less educated, divorce rates are closer to those of the peak divorce years.”
Miller’s article caused quite a stir. The demographer Steven Ruggles counter-argued that the divorce rate has actually been increasing over time:
“As Sheela Kennedy and I demonstrated in our recent article "Breaking Up is Hard to Count: The Rise of Divorce in the United States, 1980-2010" Demography (2014), available at http://www.hist.umn.edu/~ruggles/Articles/breaking_up.pdf, the much-vaunted decline in divorce is an artifact of bad data and poor measurement. As we show, the only reliable data on current U.S. divorce rates derive from the American Community Survey (ACS). Controlling for the aging of the married population, the ACS data reveal a continuing and dramatic increase in the risk of divorce since 1990. The rise of divorce is especially striking among older adults: among those aged 55 to 64, the divorce rate has quadrupled over the past three decades.”
Justin Wolfers then published his own defense of his claims about the decrease in the divorce rate in the New York Times. Ruggles countered again, making statistical arguments and suggesting that Wolfers was in the minority in his belief that the divorce rate is declining:
“The consensus of most demographers, as Schoen and Canudas-Romo (2006) put it, ‘it is premature to believe that the probability of divorce has begun to decline.’”
After all that back and forth (and more), Professor Scott Stanley had the same question I did: So what does Paul Amato (who wrote the review of the research in 2010) think now? He asked him and got an answer that was remarkably similar to the conclusion of the review paper. Amato believes that today’s lifetime risk of divorce is between 42 and 45 percent. “And if you throw in permanent separations that don’t end in divorce, then the overall likelihood of marital disruption is pushing 50 percent.”
Why is there so much controversy over the divorce rate?
#1: There is no one way to measure the divorce rate.
Here are a few of the ways that divorce has been measured:
The crude divorce rate: The number of divorces for every 1,000 people in the population.
The Census Bureau uses this measure. It is not a great measure because it depends on the proportion of people who are married. If there are proportionately fewer married people – and that proportion has been decreasing for decades – then there are proportionately fewer people who have any chance of getting divorced.
The refined divorce rate: The number of divorces for every 1,000 married women.
This is a better measure than the crude measure but it still does not answer the question that people seem to care about the most: What is the likelihood that a marriage will end in divorce?
The probability that a marriage will end in divorce: Researchers typically study this by looking at people from different cohorts – for example, people born in a particular year. The only way to know for sure how many of those people’s marriages ended in divorce is to follow them until they are all dead or divorced or widowed. For some of the people in a particular cohort, that’s going to take a long time (even if you could accurately keep track of them all). By now, we have a pretty good idea of the rate of divorce for people born, say, in 1910. But what does that tell us about the likelihood that a couple who marries in 2017 will divorce? The early 20th century was a whole different time with a whole different set of norms and demographics.
So what social scientists often do now is to look at the percentage of couples from different cohorts who reach a particular wedding anniversary. Justin Wolfers used data like that to make the claim that the divorce rate is falling. For example, he noted that “76 percent of people whose first marriages occurred in the early 1990s went on to celebrate their 10th anniversary, up from 73 percent for those married in the early 1980s, and 74 percent for those married in the early 1970s.” Of course, that doesn’t mean that the divorce rate will be only 24 percent for people who married in the early 1990s. They got to their 10th anniversary but they may not get to their 20th, or even their 11th. And even though it may look like their marriages are more stable, there is a happy obstacle in their futures. They are living longer than the generations before them, and that gives them more potential years to get divorced. So we don’t know what their lifetime likelihood of divorce will be. We have to wait and see.
#2: Statistics on divorce rates depend on the accuracy and comprehensiveness of the data collection.
Some states do not regularly report divorce rates to the federal government. A measure of divorce that depends on people’s reports rather than official documents can also be flawed. For example, if people are asked whether they got divorced in the past 12 months, they sometimes say yes even if their divorce was 13 months ago or is not yet official. Perhaps there are biases in the other direction, too, such as denying a divorce that really did happen.
So what is the divorce rate, really?
The odds that a marriage would end in divorce really were close to 50 percent in the past. There is little controversy about that. The question is whether the rate is now declining.
There is also little controversy about one particular group – the Baby Boomers. They continue to divorce at a high rate. The younger generations do not seem to be divorcing at the same rate. But of course, there’s still time, and if they live longer, they will have even more opportunity to divorce than their elders.
Also a matter of widespread agreement: Divorce rates are different for different social classes. People with more income and more education have lower divorce rates. It is worth restating the important qualification that appeared in the original New York Times article that got people all riled up:
“…the decline in divorce is concentrated among people with college degrees. For the less educated, divorce rates are closer to those of the peak divorce years.”
I think that means that even those who insist that the divorce rate is declining do not think it is declining for everyone. People who do not have a college degree account for about two-thirds of all adults 25 and older. Their divorce rates are “closer to those of the peak divorce years,” meaning close to 50 percent.
The most optimistic estimate, based on a prediction rather than a summary of actual past divorces, is probably the one in the Times: “If current trends continue, nearly two-thirds of marriages will never involve a divorce…” Again, though, that comes with the caveat that the decline “is concentrated among people with college degrees.”
My own best guess? The chances that a marriage will end in divorce is probably somewhere between 42 and 45 percent.
