Post
Conversation
![David Watson 🥑](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="96" height="96"><rect fill-opacity="0"/></svg>)
Post your reply
You have heard it said: "50% of marriages will end in divorce." But that's not true anymore. Divorce rates have been falling, and that means more marriages are making it.
Looking at the 10-year survival rate of marriages: we see a persistent trend towards stability since the height of the divorce revolution.
When we use later-year divorce rates from past decades to project into the future, we get a narrow band of outcomes. Whether we get rates like from the '70s and '80s (where we now see "gray divorce"), or the '60s or '50s(!), the cumulative divorce rate will be ~40%
Things to consider: we don't have a full SIPP dataset for 20 years of marriage among "millennium marriages" (just 2000-2002), so perhaps divorces will be revised higher. Perhaps 2010 marriages will look more like the '60s? We'll see. In any case, things are looking up.
What percent of weddings now are shotgun weddings (i.e., child born less than, say, 7 months later) compared to in the past?
Probably a combo of both self-selection (people who marry today want it more, and people who don't just don't get married), and changed norms around living together (people who would feel they were incompatible living together now find out before they begin, and don't marry).
Right on #1 but wrong on #2: cohabitation prior to marriage is linked to higher divorce odds
Less important that they are marrying later. We see lower divorce risk for first marriages across age subgroups
People below the upper middle class tend not to marry anymore. Upper middle-class marriage is more stable.
That divorce is down sounds good but it is a symptom of people stopping partnerships altogether. The loneliness crisis will be gigantic.
Yes selection may be the main driver. We may be splitting into two worlds: marital stability and partnering wild-west
Nice crude marriage rates. Wonder why 1950s look so low? Could it be a huge increase in the denominator? My point: selection about marriage is important for why divorce is dropping NOW but hardly tells the whole story. This is false: "more marriage means higher divorce rate"
Shouldn't that line be dropping at some point? I mean, it is hard to believe there's a better chance of divorce after 40 years than after 5 or 10.
This shows cumulative probability. You're right that yearly divorce risk goes down after ~ 5 years
To be expected as marriage shrinks to an upper class, highly educated, more religious practice. Now if only we could only get to the marriage rates AND stability of pre-revolution times.
I think that you need to account for the reduction in overall marriages...
Marriages that do happen are the ones that will last- strong family oriented folks- likely with religious backgrounds. They last.
Folks who would have been divorced in years past now do not get married
Yes there is a lot of selection here: religion, college education, income
Time for a lesson in conditional probability and Bayesian priors.
Compare the overall US marriage rate by decade from the 1950s until the 2010s.
Marriages are not stronger. The type of people who usually divorce (low income percentile, HS degree) are getting married at much lower rates. They never enter the pool to drag down the average.
Are marriages really getting stronger, or are only the strongest marriages still happening?
It seems like fewer people are marrying overall, and those who do are part of cultures where marriage is still highly valued. And in those cultures, divorce is less common.
Fewer people are getting married. This graph doesn’t mean what you think it does.
The flip side of men finding marriage easier in the past is that lots of people who “shouldn’t” have been married were married, and ended up splitting up. The really late marriage of these days means people are more serious when they do get married
Quote
JayMan
@JayMan471
And you know what, that right-wing nostalgia isn’t totally wrong. The Baby Boom was a time when men had life on “easy mode”, and more attractive women were more easily had by men for whom those women would be now out of their league x.com/fuckedupyogis/
Because the marriage rate dropped. If you increase the marriage rate back to previous levels, the retention will drop.
But the marriage rates themselves have fallen, so this is not very relevant.
thank you good sir for sharing this /bows to grant. Now I'n not a stats doctor but what are your thoughts on less people getting married (societal changes, expenses, standards for marrying someone etc) which is sort of causing a self selection bias that of the
Selection effects because the high risk divorce people just don't get married. You should count common law marriages
This isn’t shocking at all. Fewer people are getting married nowadays overall. A lot of long term situationships becoming normalized. So those who are getting married nowadays, tend to be more religious and higher SES than average — both of which have always been associated with
This graph almost exactly looks like the graph of marriage rates by decade. It’s likely because fewer people are getting married, and thus those who select into marriage are less likely to divorce.
Because less people by age 35 are getting married….
Below is an estimate of the percentage of people in the United States married at least once by age 35, organized by decade of birth, based on available data from U.S. Census Bureau reports, Pew Research Center, and other
That’s because marriage rates as a whole have fallen in those recent decades.
Therefore the people getting married now are more towards the upper middle class and less at the lower end of the class spectrum. Which means the people getting married in the 2000s and 2010s have
This likely reflects the fact that marriage rates have dropped so sharply.
With much less social expectation to get married, fewer people are getting married in the first place. So it would not be surprising if those that do are more likely to be truly committed to each other.
Because people don't bother getting married anymore, they just have multiple baby mamas
A big part of this is that the filtering process for marriage is a lot stronger nowadays. People marry for love
Yes. GenX and boomers are liars and cheats. Pretty obvious in the rest of culture too
This needs to be redone as "years of successful marriage per unit of population".
Marriage rates have fallen significantly among the groups least able to sustain them, so of course the average looks better, but those people at the bottom are definitely *not* better off.
I think this just reflects that people who would have gone through a divorce just don't even get married.
Can you do the same chart with either absolute numbers of marriages or number of marriages per 1000 people?
Not a shock. There are fewer marriages. The marriages that aren't happening tend to be the ones that wouldn't have worked anyway.
There are less marriages though. So the share of population that ends up unmarried is probably pretty stable
There really is not a lot of data after the year 2000, so I’m not sure how they’re coming up with that conclusion. This is based on the amount of years that somebody has been married.
?? no. Look at the data. What the chart shows is that you have to give those marriages enough time until they reach the 40-50% Share level.
This makes perfect sense with the rise in cost of living and the cost of divorce. Having gone through it myself, few people can live on a single income and most households are dual income. Doesn’t mean all the marriages are happy.
Marriage stats are filtered now. People who would have been a divorce stat 30 years ago never get married to begin with.
Maybe it’s the cycle where Strong people -> good times, good times -> weak people, weak people -> bad times, bad times -> strong people. Gen z is becoming strong.
That's cause religious people are increasingly overrepresented among people who still get married despite the high rate of divorce
Yeah but I think it’s mainly because people just stopped getting married.
Shockingly? Marriage is trending down so the only people who get married nowadays are those that are actually committed.
There are less of them . So the people that actually do get married actually want to be.
Marriages are happening later, when people are less stuoid. Also I think a lot of millennials are repulsed by divorce having gone thought it as a child. That’s my state of mind anyways
40% is still a lot.
If someone said, you have a 50% chance of a bad outcome and someone corrected you to say 'actually it's only 40%,' that wouldn't be very helpful.
As someone who unfortunately has to do a lot with divorce laws in various states, I would guess the ruinous costs of getting divorced in a failing economy have a lot to do with it.
That graph is unclear. Do the decades (e.g. 1950s, 1960s, etc.) reflect the birth dates or the marriage dates?
If the chart was % of people within certain age groups (tracked over time) that are still in their first marriage I don’t think the trend would be as positive. It’s only a small win to reduce divorces by people never getting married in the first place. A map would also be
I wonder if it has anything to do with those generations seeing so much trouble from their parents divorces.
Yeah well it's better because the marriages are younger. You have to calculate over the entire lifetime of the person. Lots of people divorce in late midlife and Gen YZ haven't reached that point yet. But the rise in never married is a serious problem.
They aren’t stronger they just haven’t had 50years of data yet. All trends go up.
When you marry the government you give the woman all the power to divorce and make a bag.
If people just married to God and that’s it, that would prevent all this from happening.
Not shocking, there are way fewer marriages than in past decades. People who marry these days tend to really want to be married.
fewer people are marrying too, so those that would have had a high propensity to end in divorce are just shacking up
