Alparslan Tuncay writes,
I look at assortative mating based on the permanent wage, which is constructed by removing age and year components from wage, and using couple rank correlation in the permanent wage as the measure of assortative mating. I then document that assortative mating increased from 0.3 for families formed in the late 1960s (initial cohort) to 0.52 in the late 1980s (final cohort).
Pointer from Tyler Cowen. This is something that many of us have speculated about, but here is some actual documentation. Tuncay also looks into the causes of assortative mating (he points out that marrying later makes it possible to know lifetime wage more reliably) and the consequences (he says that it accounts for 1/3 of the increase in family inequality).
This paper is an example of the road to sociology, but in a good way. First, it relates an economic outcome (inequality) to a sociological phenomenon. Second, it goes counter to the oppressor-oppressed ideology.
“Second, it goes counter to the oppressor-oppressed ideology.”
I am not so sure that Progressive will give up on that one. Regardless of the causes of sorting — and invidious discrimination, not…love — will be adduced as causing most of the sorting. And in any event…disparate impact will not be ignored. Since policing love will prove hard, the disparate impact of such sorting will just be taxed. The marriage tax penalty will be expanded, and it will be added to include all forms of partnering that do not involve marriage.
Hasn’t society always had a degree of assortative mating though? I believe that the 1950s post WW2 diminished but this has been society norm not the outlier. Additionally, current income assortative mating would not have worked before 1970 as there was a defined sex discrimination in the workplace. So the George & Barbara Bush who had more similar backgrounds but less income/education look ‘less assortative’ than say Bill and Hillary Clinton whose backgrounds vastly different but similar income/education. (So Bushes fit the pre-feminist society assortative mating.)
1) Not only do you know the income better past 25 but you also know the person better. So I believe this is one reason the divorce rate has dropped since 1980.
2) One aspect of divorce rates is it was the college educated couple divorcing a lot higher (1965 – 1985) versus working classes.
3) In terms of inequality is it assortative mating or simply women have equal treatment in the workplace that is hard measure?
4) I still say societies are moving toward the Singapore/Sangerism solution in which the working and middle classes learn the cheapest way to live is small families and late marriage.
Hasn’t society always had a degree of assortative mating though?
America has always had assortative mating and has only ever had “soft” arranged marriages where the would-be-bride had the power to refuse any suitor. American women, in accordance with their biological imperative, have always enjoyed the right to choose.
Beautiful paper. Wish I could do work like that. Nevertheless, the dismissal of the documented increase in educational homogamy (see page 6) because “it has been shown to have a negligible impact on household income inequality” appears to ignore the reality of uppen income households. The reality is that educational homogamy results in higher levels of one-income households among the higher income brackets. The deleterious effect of day care on children’s welfare and future earnings is well documented. Equally well documented is the lower workforce participation rate of married Ivy League women and that these women have more children. From an evolutionary biology perspective it would appear that there other important forms of inequality, not just household income.
Evolution is built in to our biology.
I think that the smarter people in the progressive/left have at least a faint awareness that most of the current drivers of racial and income inequality are due to realities that it may not be possible to change with policy. At this point, we’ve implemented the progressive playbook about as fully as possible, and yet the disparities that these policies were intended to eliminate persist.
When you look at the drivers of income inequality at the household level, what you see is that the top quintile is composed of fully employed, married people with college degrees (A) and the bottom quintile is composed partially employed single earners with less than a high-school education (B). Affirmative action, school desegregation, income transfers, and all of the other components of the progressive tool-kit have not and will not ever change the set of norms, behaviors, attitudes in a way that will close the gap between households A and B in any significant way.
This is roughly akin to discharging a full magazine only to find that the monster is still staggering towards you. I think that this as much as anything is behind the palpable sense rage and panic that’s engulfing the left on these issues. If we’ve reached a state of formal equality under the law and disparities still persist, it must be because of the unconscious bias lingering in the hearts of even off-the-shelf-progressive types like white Ivy-league sociology professors, etc.
I had once hoped that the said failures might induce progressives to do a bit of soul searching and embrace policies like school voucherization, a negative income tax in place of the minimum wage, etc – but that hope is long gone.
“Second, it goes counter to the oppressor-oppressed ideology.”
I don’t think that’s possible because oppressor-oppressed claims seem largely unfalsifiable. If high-wage men marry high-wage women, then that one can say that’s an example of how those with low lifetime wages are excluded from marrying opportunities.
However, one could also claim that non-assortative mating is evidence of privilege. For example, if high-wage men tended to marry low-wage women, then that could be viewed as evidence that men were threatened by educated professional women. One could also claim that high-wage men were using their wealth to “steal” women away from low-wage men.
The algorithm for formulating oppressor-oppressed claims is to (1) cast any non-interaction, such as not marrying outside one’s class, as neglect or exclusion and (2) cast any interaction, such as marrying outside one’s class, as exploitation/appropriation or denying others the opportunity for interaction.