[Review] Coming Apart, by Charles Murray

July 2015

I recently finished it. Thanks to Holden for the recommendation.

Murray makes some good observations about general trends in American social classes, starting with an impressive catalog of the ways that America’s elites and lower classes have drifted appart over the past half-century. (Murray restricts his analysis to whites for most of the book to try and disentangle the effects of race and class. Race relations have probably improved since the 1960s, which could confound the effects of deteriorating class relations.) Chapter 2 had a nice quiz on the theme of “which of these things do you have in common with the majority of Americans” (including things like “have you visited one of the 10 biggest restaurant chains that isn’t Chipotle” or “do you know anyone in the Army”) on which I scored 2/100.

Murray then spent about 10 chapters discussing all the indicators on which America’s lower class has slipped dramatically over the last 50 years, categorizing them into four “virtues:” industriousness, honesty, religious participation (and community participation more generally), and being married. This section was basically reasonable as observational data; I believe that Murray didn’t fudge the statistics and I mostly-ish buy his implication that we should be worried about these indicators.

There are a few problems with this section of the analysis, mostly having to do with confounding. The first is that his primary proxy for the virtue of “honesty” is criminal convictions. There are a bunch of exogenous reasons why convictions might increase that Murray doesn’t address very well (e.g. stricter enforcement and sentencing, leaded gasoline, the war on drugs). He also sort of glosses over the fact that the increase in lower-class crime seems to be turning around now, suggesting that this is due to increased incarceration (even though incarceration and crime increased together for quite a while.)

Another issue is with his section on marriage. Murray points out that even after adjusting for a bunch of obvious confounders, having your biological parents stay married is correlated with much better outcomes for kids. He then uses this (plus the fact that marriage is really easy to track) as a justification for using marriage as an indicator and getting worried about the decline in marriage stability. I basically just don’t buy this. Even after adjusting for all the “obvious” confounders, it’s perfectly plausible that most of the residual relationship between marriage and outcomes is the result of further confounding.

Despite that, the evidence Murray musters for worsening outcomes in America’s lower class–particularly in terms of work and community engagement–is largely convincing and hence worrying. A large drop in hours-worked and labor force participation is a problem no matter what the cause, so its badness is unlikely to be an artifact of counfounding the same way marriage or criminality might be.

Unfortunately, in the last part of the book (where Murray proposes solutions) he basically goes completely off the rails, probably because he felt a need to work libertarianism in somehow and realized that it wasn’t going to make it in unless he really stretched for it. Murray concludes that this decline in outcomes is all happening because the US became more like a welfare state, because receiving welfare robs people of the incentives to participate in their community, stay married, work hard, and not commit crimes. He simultaneously argues that this state of affairs makes people less happy, because to be happy you need to face meaningful challenges (e.g., having to work so you don’t starve instead of having to work so you don’t have to go on welfare).

Unfortunately the obvious counter-point is that there are plenty of welfare states that don’t seem to be having this problem of increasing nationwide deterioration and ennui. Murray doesn’t touch this at all, instead conveniently abandoning any pretense towards quantitativeness or statistics. Instead, he just spits out prophesies that the European welfare-state model is doomed and hopefully America will learn from their mistakes before it’s too late. So that’s pretty disappointing.


Enjoyed this post? Get notified of new ones via email or RSS. Or comment:

email me replies

format comments in markdown.