New results on assortative mating by education

by on July 30, 2017 at 2:59 am in Data Source, Education | Permalink

Partner Choice, Investment in Children, and the Marital College Premium, by Pierre-André Chiappori, Bernard Salanié and Yoram Weiss

We construct a model of household decision-making in which agents consume a private and a public good, interpreted as children’s welfare. Children’s utility depends on their human capital, which depends on the time their parents spend with them and on the parents’ human capital. We first show that as returns to human capital increase, couples at the top of the income distribution should spend more time with their children. This in turn should reinforce assortative matching, in a sense that we precisely define. We then embed the model into a transferable utility matching framework with random preferences, a la Choo and Siow (2006), which we estimate using US marriage data for individuals born between 1943 and 1972. We find that the preference for partners of the same education has significantly increased for white individuals, particularly for the highly educated. We find no evidence of such an increase for black individuals. Moreover, in line with theoretical predictions, we find that the “marital college-plus premium” has increased for women but not for men.

Full-Text Access | Supplementary Materials

Here are ungated versions.

1 Christian Hansen July 30, 2017 at 3:21 am

“Children’s utility depends on their human capital, which depends on the time their parents spend with them and on the parents’ human capital.”

Trad wives are a Veblen good.
Bryan Caplan is the real feminist.

Reply

2 ChrisA July 30, 2017 at 4:14 am

How does this model cope with the standard upper class British approach of sending your kids away to boarding school at 7?

Reply

3 dearieme July 30, 2017 at 6:38 am

It doesn’t because it is a sentimental assertion not a conclusion based on evidence.

Reply

4 dearieme July 30, 2017 at 6:40 am

Maybe I’m wrong. Maybe it’s a subtle attack on the American institution of serial polygamy. When did you last see your father?

Reply

5 Art Deco July 30, 2017 at 7:56 am

There aren’t many occidental countries where conventional family relations (as my grandparents understood them) are practiced with scant exception. The median crude marriage rate of a set of 38 European countries is 5.0 per 1,000 per year and the median crude divorce rate is 1.9 per 1,000 per year, which suggests that 38% of Euromarriages are destined to end in divorce. The comparable figure for the United States is 40%.

As early as 1990, about a quarter of all the children born in France were so out of wedlock. I think it was about 1/2 in Sweden. (Greece was the most orderly in this regard: about 97% of all children there were born to married parents).

The French think well of themselves for being blase about adultery and other sexual transgressions (see every French President post – de Gaulle). That’s not something to emulate.

Reply

6 dearieme July 30, 2017 at 8:28 am

Since the behaviour started in the US it has been going on longer there. So presumably the accumulated harm of not seeing your father is greater there. Perhaps the harm is assumed rather than demonstrated? I don’t know.

“As early as 1990, about a quarter of all the children born in France were so out of wedlock.” If these non-marriages were stable, then “when did you last see your father?” would be irrelevant. But that’s quite an “if”. My guess is that the “if” is wrong, but it must be a much harder topic to get good data on than the ending of marriages.

Somebody must have looked at all the children orphaned by WWII. What did they find? My no doubt worthless guess is that having your father taken from you by death might be less bad than having him taken from you by divorce or by breakdown of a non-marriage.

Reply

7 Art Deco July 30, 2017 at 10:58 am

Since the behaviour started in the US it has been going on longer there.

Not a whole lot longer. The flip in the social ecosystem here happened between 1967 and 1979 when the ratio of divorces to extant marriages trebled. (You’d had a gradual deterioration in marital durability between about 1875 and 1967 with some odd flux at times (e.g. the dissolution of war marriages during the period running from about 1944-47, but most of the change over the whole 100-odd years happened in those 12 years). The share born illegitimate passed the 20% mark in the U.S. less than 10 years before it did so in France.

So presumably the accumulated harm of not seeing your father is greater there. Perhaps the harm is assumed rather than demonstrated? I don’t know.

There’s ample research on the subject, including and especially the panel studies done by Judith Wallerstein. Of course, it was folk wisdom 60 years ago that broken families were a bad thing. There was a period of time ca. 1975 when elements of the chatterati tried to promote the idea that it was not. You’ll still find some hag-feminist who makes these claims (e.g. Barbara Ehrenreich), but that kind of thing (like the feminism of androgyny or the normalization of pederasty) is passe. However, the official idea is commonly (not universally) pureed or put aside when people make day to day decisions.

8 Li Zhi July 30, 2017 at 11:11 am

Gotsta be one of the silliest posts e.v.e.r. Unless, that is, you also give the US credit for inventing recreational sex.

9 Hua Wei July 30, 2017 at 12:16 pm

“but that kind of thing (like the feminism of androgyny or the normalization of pederasty) is passe”
Do you want your smelling salts?

10 Art Deco July 30, 2017 at 1:50 pm

Do you want your smelling salts?

Why would I need smelling salts in response to what SIECUS was promoting in 1978?

11 Hua Wei July 30, 2017 at 2:27 pm

I do not know, but forty years later it still angers you.

12 Jan July 30, 2017 at 8:03 pm

The Catholic church normalized pederasty and is still lying about it.

13 Hua Wei July 30, 2017 at 12:12 pm

“Greece was the most orderly in this regard: about 97% of all children there were born to married parents.”
Well, there is it, but there are arguments for the traditional sexual morality, too.
“The French think well of themselves for being blase about adultery and other sexual transgressions (see every French President post – de Gaulle)”.
They probably have found some better to do than caring too much about the president’s bedroom. Let the First Lady get a divorce, sue or move back with her parents – same for him if she is the cheating one.

Reply

14 Art Deco July 30, 2017 at 1:46 pm

They probably have found some better to do than caring too much about the president’s bedroom.

It’s an indicator of his character, which is pertinent information (except to people of low character or people whose manifestation of low character is constrained by the reality of women wanting nothing to do with them).

Personally, I have better things to do than fuss over whether a candidate for Congress is confused (or not) about how frequently a woman raped is impregnated by such a rape. It’s a matter of zero consequence re federal policy. It was, however, considered of considerable moment by some a few years back. (Most of whom would utter some supercilious nonsense about the government staying out of ‘the bedrooms of the nation’. (Or whatever formulation was favored by the sociopathic Mr. Trudeau).

15 Hua Wei July 30, 2017 at 3:11 pm

“It’s an indicator of his character, which is pertinent information (except to people of low character or people whose manifestation of low character is constrained by the reality of women wanting nothing to do with them).”

Yes, it is (and it is widely thought so, too – there is a reason they usually do not reveal those things themselves unless they have to and even our own Bill Clinton had – after there was nothing else to do – to appear properly repentant), so are other indicators that fail to arouse the Right’s righteous indignation, maybe for having nothing to do with exchanging fluids. “But I take advantage of the laws of the nation because I’m running a company. My obligation right now is to do well for myself, my family, my employees, for my companies. (about stiffing contractors)” – Donald Trump

But I forgot the right’s new moral of “they (investors and contractors) knew what they were getting in”

As Dobson vouched for him, Trump is making baby steps at being a Christian (timing is everything, it seems). Bill Clinton really should have tried sucessive poligamy instead of sticking with Hillary.

“Personally, I have better things to do than fuss over whether a candidate for Congress is confused (or not) about how frequently a woman raped is impregnated by such a rape.”

Yep, confused… or not… about raped women being able of getting pregnant. Who knows. Thank God it is not a matter of character to “people of high character”, they would have to exercise their consciences cor once.
“It’s a matter of zero consequence re federal policy.” Are you sure? Because he mentioned that, not me.

16 Art Deco July 31, 2017 at 2:46 pm

If free-associating is your hobby, I’m sure there’s an underemployed psychoanalyst waiting for your business.

17 prior_test3 July 30, 2017 at 4:43 am

‘Children’s utility depends on their human capital, which depends on the time their parents spend with them and on the parents’ human capital.’

And to think that for pretty much the entire existence of the human race, this statement would have been considered ludicrous.

Reply

18 BenK July 30, 2017 at 7:45 am

I’d love to see your discussion and evidence. Perhaps you are thinking of the ‘it takes a village’ in prehistory?
Or are you making some assumptions about human capital and surrogates? Or are you arguing a sort of
genetic determinism was the primary perspective (as with Oedipus, Tarzan, etc, that good blood rises to rule)?

Reply

19 prior_test3 July 30, 2017 at 8:35 am

Children (as we define the term now) were used for labor, and their parents also labored – the ones generally taking care of children unable to take care of themselves were people not able to labor.

Further, without effective birth control, the number of people being born was considerably higher than today – the amount of time that parents had to spend with children was divided among considerably more children (a number who would die before the age of 5).

It is a fairly complex topic (the definition of child being fundamental, along with the fact that most human societies in most of human history are generally not nuclear families), but there really don’t seem to have been too many villages 150,000 years ago, or 100,000 years, or 50,000 years ago. How many villages existed 20,000 years is open to discussion, though.

‘Or are you arguing a sort of genetic determinism was the primary perspective’

I do not believe much in anything that could be considered ‘genetic determinism.’

Reply

20 Scoop July 30, 2017 at 7:51 am

“time their parents spend with them”

That part is ludicrous.

Tyler’s blogging partner spent a good portion of an entire book showing that twin studies have disproven that the idea that parenting time/style affects the people kids become about as anything can ever be discredited in the social sciences.

Reply

21 James Oliver July 30, 2017 at 4:54 pm

But you can make your children’s and/or your own enjoyable or miserable while they are children and that is something.

Reply

22 rayward July 30, 2017 at 6:40 am

People born between 1943 and 1972 are much more likely to attend college than those born before, especially in the case of women, so it’s not surprising that they would marry someone who also attended college since that’s where people often meet their mates. As for black individuals, I recall an article awhile back about highly educated black women having difficulty finding mates because there weren’t enough highly educated black men. As for my own generation and those before, it was common for male lawyers to marry female legal secretaries, whereas today it’s far more common for male lawyers to marry female lawyers. Similarly, male doctors used to marry female nurses, now male doctors marry female doctors. My grandparents were the exception: both were doctors, having met while training in the same specialty. They married in 1908. In other words, male lawyers would have married female lawyers and male doctors would have married female doctors all along if there had been more female lawyers and female doctors.

Reply

23 Art Deco July 30, 2017 at 11:09 am

especially in the case of women,

Enrollment in baccalaureate-granting post-secondary institutions was 39% female in 1928. Professional schools (in divinity, law, medicine, pharmacy, and engineering) were predominantly male, making up 89% of the student body therein. The remainder were over 40% female and no more male dominated then than they are female dominated today).

In 1928, about 6% of the relevant age cohorts were enrolled in such institutions.

Reply

24 Li Zhi July 30, 2017 at 11:26 am

The problem with that factoid is that “baccalaureate granting” institutions (which allowed women) often had an alternative track/school for them…the Mrs. track. I’d think that the clearest measure would be how many women graduated with a bachelors (or higher), not how many shared the same campus as the male cohort.

Reply

25 prior_test3 July 30, 2017 at 11:49 am

It is more than that – Pembroke, for example, admitted women. Brown didn’t.

Here is some of the history, including just how long it took from the idea to the implementation of a separate university for women at Brown – ‘The founding of the Women’s College Adjunct to Brown University in October 1891, later renamed the Women’s College in Connection with Brown University, provided an organizational structure to allow women to attend that institution; Brown College remained as the men’s college. The system resembled those at Columbia University (Columbia College for men, Barnard College for women) and Harvard University (Harvard College for men, Radcliffe College for women).

Brown’s single-sex status had first been challenged in April 1874, when the university received an application from a female. The Advisory and Executive Committee decided that admitting women at the time was not a good proposal, but they continued to revisit the matter annually until 1888. Subsequent discussions led to the creation of the Women’s College October 1, 1891.’ https://en.wikipedia.org/wiki/Pembroke_College_in_Brown_University

But don’t worry, as to be expected in America that Art Deco seems to yearn for, separate but equal was of course practiced – ‘The “coordinate” status of Pembroke College was valued because it allowed women to take courses with Brown students yet still maintain the advantage[clarification needed] of a single-sex education. This included a separate student government, separate newspaper and separate social clubs.’ (Love that ‘clarification needed’ – I’m sure Art Deco can be counted on to provide it.)

Reply

26 Art Deco July 30, 2017 at 1:38 pm

Mrs. track. I’d think that the clearest measure would be how many women graduated with a bachelors (or higher),

The statistics on the female share of degrees awarded in 1928 are as follows:

Baccalaureate degrees: 44.7%
Graduate degrees 34.8%
Professional: 4.8%
Research doctorate: 13.6%

That does not include teaching certificates or nursing certificates.

Reply

27 prior_test3 July 30, 2017 at 2:41 pm

Have any links?

Lots of assertions, rarely any links. And in the case of Brown/Pembroke, your numbers are not reflective of reality. One can safely assume the same would be true of Harvard/Radcliffe.

28 A clockwork orange July 30, 2017 at 3:47 pm

Art Deco is still crooning chanson of Curtis Loewe on his Chaise Longue.. If faience is #fancy, then I might do a vlookup again to conclude the matter in regard to the eloquent theory of the north-northwest fallacy.

29 Art Deco July 31, 2017 at 2:48 pm

You can examine the Statistical Abstract your bloody self. The Census Bureau digitized all of them, including the 1930 volume.

30 Jan July 30, 2017 at 8:08 pm

Don’t mind Art. He decides what he wants to howl on here, and then tracks down whatever bullshit stat seems to back-up his position.

Reply

31 Anonymous July 31, 2017 at 12:18 am

If he was wrong, I assume you would have said that instead

32 Jan July 30, 2017 at 8:10 pm

I had this thought as well. However, they may have controlled for these trends somehow in the study, and I do think–all things being equal–that assortative mating is on nonetheless increasing.

Reply

33 Evans_KY July 30, 2017 at 7:22 am

Darwinian fitness and subsequent mate selection by the female based on said fitness. We remain animals at our core.

Reply

34 Lanigram July 31, 2017 at 11:12 am

+1

Straw dogs

Reply

35 Thiago Ribeiro July 30, 2017 at 7:30 am

Let’s be honest: it is all about money. Americans worship the Almighty Dollar. If one is not rich, one cannot afford be “loved”. The class society becomes a caste society with all the inherent iniquities. People become desperate.

Reply

36 A clockwork orange July 30, 2017 at 3:54 pm

Has tyler ever heard of drums and shadows?

Reply

37 Art Deco July 30, 2017 at 7:42 am

Economists are bored with their proper subjects and now try to horn in on sociology, anthropology, and psychology, sounding perfectly spergatroid in the process.

Reply

38 dearieme July 30, 2017 at 8:30 am

Physicists used to do that. It often worked well when they brought new instruments to other fields. Do economists have the equivalent of “new instruments”? What are they?

Reply

39 Trump Fan July 30, 2017 at 10:50 am

Doesn’t look like it. Just another example of “assume discredited theory X is true.”

Reply

40 Art Deco July 30, 2017 at 11:02 am

No. It’s an example of ‘assume people are widgets’.

You can argue that works passably when you’re describing economic behavior (as long as you recognize that ideal types are not realities and that people tend more often to be satisficers than optimizers).

Reply

41 Bill July 30, 2017 at 8:27 am

From reading the paper: So, wealthy educated parents invest in their children because one spouse has more time (which has also increased due to technology changes in the home) and can do so, thereby increasing the human capital of the child. Educated stay at home moms.

Reply

42 chuck martel July 30, 2017 at 12:59 pm

“wealthy educated parents invest in their children because one spouse has more time”

Probably not. Throughout human history until the day after tomorrow, almost everywhere, a major role of the child has been to look after its younger siblings. It’s a part of the educational process that once created mothers and fathers. Additionally, children of all but the most affluent families were expected to perform labor that contributed to the family, especially important in agrarian societies, regardless of the family’s wealth. There’s no reason to believe that any one particular mother in any era would spend a lot of time with her children rather than playing panguingue, gossiping, chasing the mailman or dozing on the sofa.

Reply

43 Victoria Wilson July 30, 2017 at 10:09 pm

This paper offers several interesting views. 1. Private parties are producing public goods- not the schools, not government social workers. 2. The labor hours spent in such an endeavor are a measure of the potential outcome- Marx had, after all, something to contribute to economics 3.The reaction to this paper by the commentators illustrates why the mix of economics and sociology has taken so long to evolve- it triggers all sorts of diverting analysis as the measurement of social choices threatens to validate or refute one’s philosophical outlook.

Reply

Leave a Comment

Previous post:

Next post: