What I don’t understand, is who did upper middle class women marry in the Mad Men era, if the men married their secretaries?
1. The distinction between upper middle class and lower middle class was not as sharp back then.
2. College was not as much of a status marker back then. Few women went to college, and even most middle-class men did not.
3. So class selection was not as strong. A (upper-) middle-class woman who married someone without a college degree did not think of herself as marrying down.
4. The result was a lot divorces from middle-class marriages of that period. I believe that it is Stevenson and Wolfers who point out that the period of the 1970s was a transition from marriage as production complementarity (I’ll bring whom the bacon, you fry it) to consumption complementarity (let’s make sure that our leisure interests coincide). This form of marriage turns out to be much more class-selective.
The bottom line is that the upper middle class men and women both married down. In the 1970s, they got divorced, and the sorting along class lines got sharper.
What I’d like to know: how strongly correlated were the IQs for males and females getting married in 1940, 1950, 1960, 1970, etc. How much did assortative mating increase over that period of time?
The one thing to remember is the Mad Men era is the historical outlier of cross class marrying not the long term reality. I bet before WW2 cross class marrying was very uncommon and the class structure changed significantly after the disruptions of the war. (A large middle class was created with labor shortage and society enforced discrimination in the labor market.)
In terms of today, the BIG change is people are waiting until 28 – 30 and meeting their spouse after they are 25. So as people age past High School and Graduate College, they tend to interact with less people and especially with less people outside their class. So people date as much of different classes as they did 50 years ago, but with birth control, they are avoiding shotgun marriages.
How much is assortive mating a function of geography. Specifically, are those writing on the subject approaching it from a professional or geographic (think metropolitan, maybe East Coast) perspective. I live in “fly over country” and don’t find it all that uncommon to meet married people from distinctly different occupations and educational levels. I suppose at the elite level it is a reality, but does that really matter.
Isn’t assortative mating about the evolution of sorting away from status-based sorting (or rather the evolution of status itself) towards IQ-based sorting?
Keep in mind that in the early to mid 20th century, not only did a relatively small minority have college degrees, but there were significantly more male than female college grads, so college-educated women didn’t have to ‘marry down’ educationally (and, indeed, joke about going to school for an ‘MRS’ degree was a thing when my parents were in college in the 50s).
Richard Herrnstein’s 1971 article in The Atlantic “I.Q.” made well-known the theory that assortative mating by IQ was the wave of the future. It’s unfortunate that it’s hard to find the full text online since it’s a historically important essay.
I read it around 1973 so my memories are very fuzzy. My impression is that Herrnstein didn’t necessarily have a lot of data for his hypothesis; it was more his thought experiment about what gentile small town life must have been like in the old days. (Herrnstein was a New York City boy.)
It would be interesting to hear what Gregory Clark thinks of the theory. My impression from his two books is that there likely was more assortative mating by IQ in the past than Herrnstein would have imagined, but I don’t recall any decisive tests one way or another.
Thank you for taking the time to answer my question.
The shorter answer is “In the Mad Men ERA, ‘upper-middle class’ did not exist.”
I personally think “Aflluent” is a much better term for this group of people. “Upper-middle class” as a term has a lot of humblebrag and self-justification built in. “Upper-middle class” may have bene more accurate when the term was coined, but not anymore. The affluent still use it and think of themselves that way because they are principally trying to distinguish themselves as apart from the wealthy, but it disguises the reality (c.f. Murray, Putnam, etc) that their incomes, job descriptions, economic interests, tastes, and mores all differ fairly radically from the middle-middle class. The sharpest class division in contemporary society is around the 75th percentile of income, not the 99th.
– While there have always been people who make say, 2x the median income (ballparking 100k vs 50k today), in the 60s/70s the number of people meeting this threshold was much smaller than today
– In the 60s/70s, the geographic and occupational concentration of such people was much much less than it is today. Back then, the typical person who would have been “upper-middle class” had a job that was upper-middle management. They would have spent most of their day working with and managing people who earned less than they did at the same company, living perhaps on a nicer street, but not so distant that their kids would be in different schools. Today, as has been documented in a lot of places, the growth pay differences has been larger between firms than within them. The people today who make 2x median are much more likely today to work with and live in a neighborhood with primarily other people who earn similarly. This change is why they can be described as a ‘class’ in a way that would have been less accurate 40 years ago
– Until the early/mid-70s, all the job descriptions and career tracks into affluence and wealth were effectively closed to women. Doctors, lawyers, corporate management, and anyone reasonably described as professional had to marry women who earned less because practically all of those careers were effectively male-only. The baby boom generation was the first in which enough women had access to higher-earning careers that assortative-mating was possible on a scale big enough for sociologists to invent a term for it.