Four Forces Watch

Richard Reeves says,

What’s been driving the kind of economic separation has been a combination of two main factors: One, well-established earnings inequality and higher returns to higher education at the top. And then actually you see this stunningly unromantic term, assortative mating — i.e. college graduates marrying other college graduates, which means they double down on that income. And then they’re able to put that into housing which gets them access to a good school and so on. Then you take all the tax subsidies that are available. So, I’ve got a new paper on 529 but there’s also mortgage interest deduction which ends up in a way subsidizing this kind of separation.

He is explaining the increased separation between the top 5th and the rest of the population in terms of income.

There is an interesting discussion of the notion that in order to have upward mobility you have to have downward mobility. Actually, that is not necessarily the case. Suppose that the bottom of the income distribution is populated largely by young workers and immigrants. As they move up the ladder, instead of replacing them with once-rich families on the way down, you replace them with a new generation of young workers and with new immigrants. This is not merely hypothetical.

15 thoughts on “Four Forces Watch

  1. Confucius was right about “rectification of names” being essential to enable clear thinking.

    “Returns to Education” is a pretty misleading fallacy of attribution.

    And the meaning of “social mobility” is fluid enough to permit a fixed normative valence even when the details of the implications are suspect. In particular, the understood public and political meaning differs strongly from the way economists measure and analyze the income distribution data.

    Yes, one can adjust for age cohorts, talent, race, gender, etc. or just ignore all that politically sensitive complexity by focusing on a less controversial segment, Coming Apart style.

    But no one seems to care very much about that, which is revealing.

    The whole point of the mobility discussion is to try and find a plausible proxy for “level in iniquity / social injustice.” And there is no way to do that without some implicit assumption or model regarding what the current distribution in the “equitable society” – the deviation between the current data and the distribution in that equitable society being some kind of meaningful metric for injustice.

    But if we can’t agree on what the equitable distribution looks like, we can’t agree on the current deviation, and so we have no commonly shared way to discuss, interpret, or analyze the same data. Which means the whole exercise is worse than futile, just a pretense and charade of “mathy” quantitative academic rigor to serve as seemingly respectable fodder for yet more normative sociology.

    • Handle is correct. In particular this kind of “analysis” is used as a pretext to villify the top 20% of the income distribution.

      The top 0.1% has an alliance with the bottom 80% against the top 20%. Exhausting that top 20% with yet more taxes and regulation serves the interests of the top 0.1% in preventing rivals from threatening their formidable monopolies on desirable real estate and political influence. The bottom 80% is easily manipulated with plain jealousy.

      • I definitely see this. As a top 20% wage earner for many years, my effective federal income tax rate was always in the range of 24-25%. So I’m sensitive to the argument.

        The various proposed tax plans will skew this even more. If we thought “47%” was a lot of ppl not paying income tax, wait to see what happens next.

    • Concerning the relation between “assortative mating” (Reeves) and “what the equitable distribution looks like” (Handle): I suggest a model such as below might be both useful and value neutral.

      Given an initial population with Gaussian distribution on some metric (such as “wealth”); and given a way to measure “assortative mating”, compute (by simulation of several generations) what distribution the population asymptotically tends to.

      Two definitions of assortativity in the literature, both due to Spearman, seem appropriate. With either definition and for any given population size, any mating yields a number that may be scaled onto a range of 0 to 100 (from no assortment to perfect assortment).

      Such simulation must be able to estimate (or guesstimate) the metric (“wealth”) of children, given the metric of the parents. We may quibble about any such method, but at least any given method is transparent, and thus open to evaluation and criticism.

  2. It’s mentioned in the article but not emphasized enough.

    There are NO gov’t policies that equalize two kids’ chances of similar natural ability when one is raised by his or her married bio-parents, and one is raised by a single, unmarried parent.

    Lots and lots of research, mostly indirect, shows this very clearly. This parental choice is the single biggest driver of the inequality, and needs to be talked about more honestly.

    On mortgage deductions, there should be lifetime maximum limits on the deductions (maybe $500,000 for life; $50,000 per year).

    Perhaps a national zoning tax to pay for the gov’t and Social Security.

    Top 20% of 310mln is top 62m people, mostly married college grads DI (double income), tho now with 1 or 2 kids (rather than DINKs of the 80s). Also a tax on those who have college degrees and less than 3 kids — with more support for kids.

    While the point about low income/ low wealth quintiles could be more from temporary young folk in the low quintiles moving up, there will be some large of poor adults whose lifestyle choices plus limited natural ability and bad work habits means they are 2/3 choosing poverty.

    The real main issue should be increasing the median wealth, so that the richness of the top, and the growth of their incomes, doesn’t bother the middle income quints so much (20-40-60-80). Tho most in the 20-40 quint would argue that the 60-80 quint is doing fine.

    Where is good graph showing how much income these quint boundaries are making; and how much wealth is in that quint? Maybe with top 1% broken out and two #s for the 80% (with & without top 1%).

    • Getting people to marry, stay married, and have 2.1 kids is a difficult task. The lower class can’t stay married. The upper class doesn’t get divorced but either don’t get married or get married so late in life they don’t produce replacement rate children.

      This appears to be a culture problem (since at a given level of heredity conservative social believes correlate with higher marriage and fertility rates), but good luck getting people to tackle the issue.

      Socioeconomic status is a zero sum game. However, there is a man for every woman and vice versa. We can’t all be rich, but we could all have families and satisfying communities. That’s the only goal that has any real chance of succeeding.

  3. The libertarian/conservative turning against the upper class (think David Brooks article also) is getting kind of weird and this used to be the core Republican base. (I am not in this category.) I am guessing it is because they went D higher than normal but exit polls showed Trump won these voters. However, this turn against this class is kinda weird though.

    1) In terms of assortative mating, aren’t we exaggerating this reality compared to past history? (especially compared to pre-WW2 days) The primary difference between today versus the 1950/1960s is most upper middle class women were housewives on doctors or lawyers. (They were often the leaders at local churches and charities.)
    Upper middle used to George/Barbara Bush versus Bill/Hillary Clinton. It was the sex discrimination labor market was the reality not assortative mating.

    2) I think the primary reason for increased assortative mating is people are marrying later in life and don’t meet their spouse until 25+. (In general, most people start being more selective of interactions around 25.) The other point not brought up is the divorce rate is dropping and it is primarily because college graduates (say Charles Murray) used to be the highest divorce rates in the 1970s. So this new marriage mating has had positive effects.

    3) In reality isn’t the 1950s US economy an historical outlier of economic inequality? It followed the Depression, Depression Baby Bust (look it up), World War 2 and other developed nations bombed out where there was a labor shortage which led to worker union power. The job growth of Eisenhower was ~1.16% per year which is lower than any other President than Bush Jr. and only slightly higher than Obama. (1.13% which included the 2009 job decreases.) (Oddly enough our current economy is lot closer to the 1950s with relatively low labor supply.)

    4) It does seem to me that the primary failures of society today are people in post-High School years where 1950/1960s economy/community were better at getting these people down a productive path.

    • Multiple reasons for this.

      One is we use immigrant labor. My factories almost exclusively use immigrant labor. We interview a lot of whites but we almost never hire them.

      Why not? They have shitty work history. They got fired from jobs for not showing up. They failed high school. They’re stupid enough to complain about their previous job in an interview. They have a criminal record. They can’t pass a drug test.

      The white working class is gone. The jobs aren’t. We actively hire and cant find people.

      This is a massive cultural breakdown. And it’s single motherhood. That is the cause. It’s downstream from culture. The elites said free love and liberation and “don’t need no man.” And their kids fail at life. They expect us to treat them like their mother. And instead we fire them.

      Ymmv

      • Interesting. It sounds like your suggesting that what happened in the black community has belatedly taken off among working class whites. I forget who it was but some black person ominously forecasted that ‘we’re the canary in the coal mine’ in relation to the white population.

        It seems pretty class specific though. Sure some children of affluent whites slip through the cracks and nominally marriage is declining among them, but in practice affluent whites are still fairly behaviorally conservative, e.g. monogamous, two parent households, strong emphasis on education, etc.

        • The problem with upper class whites is they avoid total failure by not taking any chances in life. This is most evident in their low TFR. Can’t be a single mother if you don’t marry and have children.

      • Immigrant single parent rates skyrocket between the first and second generation, along with other indicators of social pathology (crime, etc). It’s fair to say that people come here with their native peasant culture and within a generation assimilate to American underclass culture. In the long run everyone behaves the same as natives with the same IQ.

    • ” I think the primary reason for increased assortative mating is people are marrying later in life…”

      I would think the primary reason is that more women are attending university and participating in the labor force.

      • Correct, but it seems to be more attending university then being in the labor force. Years of schooling, not merely IQ or hours worked, is a better indicator. Perhaps its in part what they are teaching at universities these days.

  4. Not sure about the assortative mating bit. Outside of the ivy league (and similar institutions) liberal arts colleges skew strongly female. So maybe the model works for the top 1%, but not the top 20%.

    • The female skew has only recently been so strong AND there was a “backlog” of marriageable college male grads w/o female grad possible mates that could absorb some higher number of female grads. That backlog has now pretty much been completed, so the excess female grads have fewer recent male grads, and far fewer earlier male grads still interested in marriage.

      Far more male grads marry female non-grads than female grads marrying male non-grads. On marriage and children, women’s choices define the culture.

Comments are closed.