Divorce and the Business Cycle

INSERT DESCRIPTIONNaum Kazhdan/The New York Times

There’s an interesting discussion over at Room for Debate on how the recession is affecting family life. A key question is whether recessions lead to a rise in divorce or not.

Here’s Betsey Stevenson (my frequent coauthor and significant other):

While the economically vulnerable are in less stable relationships, there is little evidence that the economic cycle drives divorce. The big swings in marriage and divorce follow dramatic social, rather than macroeconomic change. Blame the wars, the sexual revolution, and changes in the meaning of marriage, rather than cyclical ups and downs.

Here’s Barbara Dafoe Whitehead:

The divorce rate soared during the stagflation of the mid-1970’s, continued to rise during the deep blue-collar recession of the early 1980’s, and thereafter stabilized at historically high levels.

And here are the facts (updating Betsey’s and my Journal of Economic Perspectives paper):


The chart shows the two standard ways of measuring the divorce rate, and the shaded areas on the chart highlight the recessions. Contrary to Whitehead’s claim, the divorce rate actually fell during the early 1980’s recession. And she’s wrong to say it stabilized after that, as divorce rates continued to fall. In fact, divorce rates have been falling for the past 30 years. It’s time for social commentators to start to recognize these facts.

To my eye, the most amazing fact in this chart is just how little the divorce rate is affected by the business cycle. The 1970’s recessions didn’t see a sharp uptick in divorce, but rather the continuation of the pre-existing trend. Divorce rates declined during the 1980’s recession and barely changed during either the 1990 or 2001 recessions. And if we examine data over the longer run, again we see that divorce doesn’t seem to move around much with the business cycle:


If anything, the Great Depression suggests that divorce rates may decline during dramatic slowdowns. Unfortunately my charts report only the most recent complete annual data, which are for 2007, while the recession began in December of that year. To learn about the consequences of this recession, we have to turn to the most recent NCHS data. And these data tell us that in seven months through to July 2008, the number of divorces is down about 6 percent compared with the equivalent period in 2007. That is, the ongoing decline in divorce is continuing.

Allen Reynolds

Really interesting - that's a positive social trend we don't hear often enough.

Interesting too the divergence between "Divorces per 1000 people" and "Divorces per 1000 married couples". Had to think about that for a moment. The answer, of course, is a decline in the marriage rate.

What's with the spike around 1948? Did everyone come back from the war and get divorced three years later??

Leland Witter

Given the difficulty in selling a house, I wonder how many couples might postpone actually divorcing to wait out the housing market a bit rather than both getting killed on selling and even possibly having to pay if they are under water.


People can't afford divorce lawyers, separate housing, to lose health care, etc. The costs of divorce are high compared to having a lousy roommate.


I seriously doubt that many people are so calculating as to consider housing values as a deterrent to divorce. It seems much more likely that pyschological side effects of the economy will have a more direct effect on divorce rates if any. In particular when it comes to financial factors, I think people are more hesistant to end a marriage if there is less certainty whether they would be able to provide for themselves or their children on their own.

That brings up an interesting question. Is there a difference between divorce rate trends of couples with children vs without during different economic cycles?


So basically there's a spike when marriages unraveled after WWII and then a more permanent increase when society liberalized - and the divorce laws became less punitive for women - but otherwise divorce doesn't act much as a good subject to price adjustment.


I'd guess late 1940s was a change in the legal situation, but that's just a guess, and 60 seconds of searching didn't turn up any support for it. Anyone know?


Interesting that things turned around in 1980. Since this is the Freakonomics blog and all, it seems natural to ask if there could be any connection between Roe v. Wade and the divorce rate. The logic would go like this: Fewer unwanted pregnancies lead to fewer unwanted marriages lead to fewer divorces. It would make sense that the effect would take a few years to show up in the divorce rate. Say a hypothetical pregnancy doesn't lead to a child in 1974 and the corresponding hypothetical shotgun wedding doesn't happen either. Six years later (after they hypothetical couple has time to realize they can't stand each other), the corresponding divorce doesn't happen.

I expect to see this addressed in the sequel to Freakonomics.

Steve Peterson

I'd bet there is a significant financial impact -- most notably, when more women started getting out of the house and having careers of their own, and thus weren't reliant on marriage for their income, divorces went up.

Subsequently, during economic downturns the financial risk of going from a two income family to a two household family becomes more unappealing.

Regarding the post-war spike -- a spouse you don't get along with that well is much easier to get along with if you're stationed 5000 miles or more away, but the old problems come to the surface when you're finally re-united and coming home after a big event is also a good reason for both partners to give the relationship a serious second look.

Eric M. Jones

There are SOOOO many factors which figure into this. I have read that the average length of a marriage in 1860 was about the same as that in 1960, if you didn't include the marriage termination by the death of one partner.

In 1900 there had to be many fewer marriages and many more out-of-wedlock births, many fewer records kept.

The chart shows that divorces went way up after the baby boom generation came of age and started to use the birth control pill and pot, then declined after AIDS came on the scene and VCR allowed porn to entertain married adults.

"....the spike around 1948? Did everyone come back from the war and get divorced three years later??- Allen Reynolds"

Yes, Allen. It's easy to stay married when one partner is on the other side of the planet. There is a similar bump after WW1 and the war in Vietnam.

David S.

My guess is that the late 1940's was not a change in the legal rules but that commenter 5 is right. Just a large number of war marriages, the "war brides", coming apart quickly after years of not seeing each other.


Interesting data. But they can't capture how many de facto marriage relationships are breaking up, only the de jure ones.

I wonder how that graph would look?


It would somewhat make sense that you wouldn't see a spike in divorces during a recession, since a divorce tends to be expensive. Though my Dad would argue that each of his divorces was worth every penny.

The data is certainly puzzling... since financial issues are often cited as a top cause of marital problems.

John Logan

What's wrong with this picture?

Let's have a close look at the NCHS link above - straight from the Technical Notes page of the National Vital Statistics Report on Marriages and Divorces:

"NCHS has ceased publishing divorce counts and rates based on provisional data from the combined 50 states because it is no longer statistically feasible to calculate estimates of monthly divorce counts for nonreporting states (California, Georgia, Hawaii, Indiana, Louisiana, and Minnesota). Instead, NCHS publishes only the 12-month divorce rates shown in Table A, based solely on the combined counts and populations for reporting states and the District of Columbia. Aggregate 1-month, cumulative counts and rates, and 12-month counts for divorces have been discontinued because of their very limited usefulness when based on less than the full complement of states. Monthly marriage and divorce counts as reported by each state will continue to be shown in Table 2."

Which, in layman's terms means that divorce rates for the US as a whole are guesses based on only those states reporting. Considering the size and dynamics of California alone (which stopped reporting back in 1992) even the NCHS admits their numbers are skewed.

That said, considering that today we have the largest population of adults who come from divorced families, and who study after study concludes are 3 times more likely to divorce themselves it's no wonder that a US Census report released in 2007 based on the 2004 SIPP Report predicted that people marrying today stand only a 33% chance of ever seeing their 25th anniversary. That's hardly a decline in divorce rates.

Ryan hit the nail on the head regarding divorce in economic downturns. Better together and unhappy than divorced and bankrupt (and probably still unhappy).



With respect to the comment, 'The big swings in marriage and divorce follow dramatic social, rather than macroeconomic change."

I agree -- the advent and adoption of social networking sites (i.e. facebook), and not the recession, will create a measurable increase in the divorce rate.

Of course, development of the data will take some time.

Thoughts on this?


the business cycle just amplify the scope of trends:if the pare trends to solve problem through divorce,then recession bring more divorce;vice versa


It's not the recession/housing prices/money keeping people from divorce. If that were the case, you would see a serious spike in the 3 years following the end of the recession.


:::::Better together and unhappy than divorced and bankrupt (and probably still unhappy).::::::

this attitude just leads to extramarital affairs. Maybe disease, too. I often think lots of couples after they raise their children end up in this situation and just stay with marriage.Maybe it is why so many people i see are overweight and don't really care about their wellbeing -because of that idea that money is what you should base your most important decisions on.

That is why it's so hard for me to find a true friend after my divorce. You get lumped into a divorced "meat" category but also a high maintenance group that people either expect want group therapy or a quick good time. I divorced for entirely a different reason than most people, for my peace of mind and that my own personal health insurances. I lost the group health insurance, the home. Outsiders, like lawyers and even family members who concentrate on money and data never ever get the entire picture and probably only see that you are unhappy because you are alone.



After living through many a tough time living in a boom and bust area that is tied to the softwood industry I think that all the tough times wear down couples and depending on the couple and the good times to offer a reprieve from the stress and turmoil...some might hold on longer than others.

In our case it lasted 25 years...and when it came it was when the better times were in sight so technically one can look at the stats and say the marriage ended in a period of positive economic growth but the damage was done many years previously.


How about the dramatic drop in marriage rates? There has been an almost 50% drop in per-capita-marriage rates between 1969 and 2004. Here is the data:


While a 20% drop in divorce rates is good news, an almost 50% in new-marriage rates is even bigger news. What are the factors behind this societal change?


I believe the labels on the right and left vertical scales on the graphs have been swapped. The rate of divorce per 1000 married couples must be larger than the rate of divorce per 1000 people: even if 100% of people are married, the rate of divorce per 1000 couples will be 50% of the rate of divorce per 1000 people.