Bryan Caplan  

Divorce and Motivated Reasoning in the WaPo

Longer Pub Hours, Fewer Car Ac... My San Francisco Fed Talk...
In The Economic Naturalist, Robert Frank remarks:

Psychologist Tom Gilovich has suggested that someone who wants to accept a hypothesis tends to ask, "Can I believe it?"  In contrast, someone who wants to reject it tends to ask, "Must I believe it?"

I immediately thought of Gilovich's insight while reading Scott Keyes' op-ed on divorce in the Washington Post.

"Can I believe it?":

No-fault divorce has been a success. A 2003 Stanford University study detailed the benefits in states that had legalized such divorces: Domestic violence dropped by a third in just 10 years, the number of husbands convicted of murdering their wives fell by 10 percent, and the number of women committing suicide declined between 11 and 19 percent. A recent report from Maria Shriver and the Center for American Progress found that only 28 percent of divorced women said they wished they'd stayed married.

"Must I believe it?":

While some studies show that children of divorced parents do experience worse life outcomes -- including diminished math and social skills, a higher chance of dropping out of school, poorer health, and a greater likelihood of divorce themselves -- Stanford sociologist Michael Rosenfeld points out that there is no way to test definitively whether children of divorced parents were already more likely to experience such outcomes. And as Stephanie Coontz, a historian and the author of "Marriage, a History," explains, what's most critical is the high-conflict environment that kids grew up in before their parents separated.

<sarcasm>How fortunate that there is a "way to test definitively" whether domestic violence and female suicide would have fallen if divorce laws hadn't been liberalized!</sarcasm>

"Must I believe it?" continued:

Would making divorce less accessible encourage partners to stay together, as conservatives hope? Probably not. Waiting periods and mandatory classes "add a new frustration to already frustrated lives," Rosenfeld notes. In other words, a cooling-off period isn't cooling anybody off.

"Can I believe it?" continued:

More problematic, these roadblocks "could easily exacerbate the situation and harm kids," Coontz says, noting that divorcees are "more likely to parent amicably if they haven't been locked into a long separation process."

Keyes' double standard vexes me even though I think (a) government should play no role in marriage, and (b) twin and adoption evidence shows little or no effect of divorce on kids' adult outcomes.  All of the following still remain highly plausible:

a. There are a lot of so-so marriages.

b. The cost of divorce affects the divorce rate for couples in so-so marriages.

c. Most kids of couples in so-so marriages strongly prefer for their families to stay together.

Oh, and if liberalized divorce has been so great, why did this happen?  Don't tell me what you can believe.  Don't tell me what you must believe.  Just tell me what makes sense to you.

COMMENTS (8 to date)
adbge writes:

If marital stability does have a genetic component, one should use the question "Are your parents divorced?" to filter potential long-term mates.

A Google scholar search backs this up:

Divorce/separation between birth and 5 years predicted early menarche, first sexual intercourse, first pregnancy, and shorter duration of first marriage. Separation in adolescence was the strongest predictor of number of sex partners.
ThomasH writes:

Is the question why divorce among low income people rose more than divorce among high income people?

Maybe the reduction on the cost of divorce is relatively greater for low income people and so for a similar elasticity of demand, you get more divorce? Or maybe as low incomes stagnated it made less sense for a partner to stick around hoping the other partner's income would rise?

MingoV writes:
Tom Gilovich has suggested that someone who wants to accept a hypothesis tends to ask, "Can I believe it?"
Apparently, he doesn't know any skeptics who say, "Show me the evidence." whether they want to believe it or not.

Low income is just a correlating factor in divorce. People with low incomes tend to marry early. They tend to marry without applying good judgment. They tend to have children earlier in their marriages. They tend to have more children despite inadequate income. A divorce doesn't hurt financially because even with alimony and child support payments, the noncustodial parent (almost always the husband) can live better. The custodial parent often qualifies for federal aid: Medicaid almost always and some SNAP aid much of the time. Thus, divorce can be a win-win event for the couple, though it's a lose event for taxpayers.

Joe Coco writes:

I believe a majority of our societal woes can be attributed to the negative externalities of social security and no fault divorce.
No fault divorce creates unacceptable incentives for the vicious spouse to act badly. Let's assume bad spouse A racks up all the debts, and good spouse B accumulates all the assets. In a no-fault divorce scenario, bad spouse A gets half the assets and good spouse B gets half the liabilities. Once the divorce takes place, bad spouse A gets lots of community support for carrying their share of the post divorce load, while good spouse A often carries more than their court appointed load.
To reduce the impact of moral hazard, society needs to forego the short term efficiency of no fault divorce in favor of assigning fault to broken marriages. Once we get the incentives correct in light of natural law, we will see a decline in divorce.

NZ writes:

No consideration seems to have been given to the question of how no-fault divorce (i.e. drastically lowered costs of divorce) impacts present and future society-wide attitudes towards marrying in the first place.

If the laws and customs that buttress an institution are systematically demolished, it sends the message that the institution is not very important or meaningful.

In this case it also dilutes the meaning of marriage itself, which is supposed to be a lifelong bond. How seriously can anyone take the phrase "till death do us part" if everyone knows a divorce is easily obtained?

Eli writes:

Jonathan Haidt's book The Righteous Mind also quotes Tom Gilovich. It's a great summary of the process through which we confirm our biases.

Enial Cattesi writes:
How seriously can anyone take the phrase "till death do us part" if everyone knows a divorce is easily obtained?
A good question would be "Was that always so?" Marriage as an institution had many forms. Why would this form be any better than the old ones? Why should "till death do us part" be a more bounding contract than other pledges?

To make divorce harder it is just enabling someone else's prejudices and beliefs about how the world should be on the general population.

Marriage used to be a contract between two people. As usual some third parties (the church, the state) want to control this form of association and so we arrived at this mess.

But about the question how did this happened: how many professionals are shunning marriage all together?

NZ writes:

@Enial Cattesi:

"Was that ['till death do us part'] always so?"

For the most part yes, it was, at least in the history of Western (i.e. European and North American) society. I don't know much about the history of marriage in other parts of the world, but I suspect marriage until death was the prevailing pattern there too.

I suspect this because marriage for life is indeed superior with respect to the fundamental goal of marriage: to solidify the bonds that support the formation of a stable family unit. (Essentially, the rearing of children.)

You're correct that making divorce harder is just enabling someone's prejudices and beliefs about how the world should be, but isn't that true of any social norm? Are social norms and customs to be disposed of simply because they carry this implicit evangelism? And who is this "else" you speak of?

More fundamentally, is "the general population" just a mushy gob of blank slates with no inherent inclinations, tastes, or behavioral patterns?

Comments for this entry have been closed
Return to top